Abstract
The germplasm base of strawberries is restricted. The major cultivated strawberry species, Fragaria ×ananassa, originated ≈250 years ago when South American F. chiloensis subsp. chiloensis forma chiloensis and North American F. virginiana subsp. virginiana accidentally hybridized in European gardens. Since that time, only a handful of native clones have been used by breeders. As a novel way to expand the germplasm base of the strawberry, we preselected native clones of F. virginiana and F. chiloensis for a wide range of horticulturally important characteristics and then reconstructed F. ×ananassa by crossing superior clones of each. Before crossing between species, we undertook one round of selection within species to maximize diversity. Reconstruction appeared to be an effective method of strawberry improvement, because superior families and individuals were identified that had outstanding vigor, high productivity, seed set, fruit color, and firmness. None of the fruit were of commercial size, but one reconstruction family, FVC 11 [(F. virginiana Frederick 9 × LH 50-4) × (F. chiloensis Scotts Creek × 2 MAR 1A)], had individuals with fruit weights of almost 20 g.
The founding genetic base of the commercial strawberry, Fragaria ×ananassa Duchesne in Lamarck, is limited. It originated ≈250 years ago when a few clones of South American F. chiloensis chiloensis (L.) Miller subsp. chiloensis forma chiloensis and North American F. virginiana Miller subsp. virginiana accidentally hybridized in European gardens (Darrow, 1966). The systematic breeding of strawberries was started in England in 1817 by Thomas A. Knight using only a small number of native and cultivated clones. North American genetic improvement was initiated in the mid-1800s with a restricted group of European F. ×ananassa cultivars, South American F. chiloensis, and North American F. virginiana genotypes. This germplasm base played the predominant role in public and private breeding programs for the next 100 years.
Although impressive breeding progress was made using this narrow germplasm base, other horticulturally useful genes are likely available in native populations of Fragaria, because both octoploid species have extensive geographical ranges that encompass a broad range of biotic and abiotic stresses (Hancock et al., 2004; Staudt, 1999). Contained within the wild germplasm is a wide range of interesting flavors and aromas, unusual resistance to heat, drought and salinity, almost a continuum of photoperiod sensitivities, and tolerance to a wide variety of diseases and pests (Hancock, 1999; Hancock et al., 1990; Luby et al., 1991).
To date, almost all the novel native genes that have been incorporated into cultivated material have come through back-crossing (Hancock, 1999; Hancock et al., 1993a). At least eight wild clones have been introgressed into F. ×ananassa since the 1920s (Sjulin and Dale, 1987) bringing in such traits as day-neutrality, red stele and strawberry aphid resistance, drought and salinity tolerance, and winter-hardiness (Barritt and Shanks, 1980; Bringhurst and Voth, 1984; Daubeny, 1990; Galletta et al., 1989).
Back-crossing from wild genotypes has allowed for the rapid incorporation of a few genes into the genetic background of F. ×ananassa. However, only a limited number of native clones have been used in this manner, leaving much genetic diversity untapped. Also, potentially useful diversity at non-selected genes has likely been lost when back-crossing and tightly linked deleterious genes can be carried into late generations. These two disadvantages are compounded when only a single non-selected native clone is used rather than a group of elite selections from a broad screen of native material.
An alternate strategy for germplasm enhancement would be to pre-select native clones of F. virginiana and F. chiloensis for a wide range of horticulturally important characteristics and then reconstruct F. ×ananassa by hybridizing superior clones of each. Hancock et al. (1993a) suggested a multiple stage process of reconstruction (Fig. 1): 1) select elite clones of F. chiloensis and F. virginiana from published reports and personal experience; 2) intercross the elite selections within species and select the superior progeny; 3) intercross these elite selections again within species and select the most promising genotypes; 4) reconstruct F. ×ananassa by making interspecies crosses among the elites of F. virginiana and F. chiloensis; and 5) select superior genotypes of reconstructed F. ×ananassa that can be used in further breeding and/or varietal release.
Breeding scheme to reconstruct Fragaria ×ananassa using native clones of F. virginiana and F. chiloensis.
Citation: HortScience horts 45, 7; 10.21273/HORTSCI.45.7.1006
Breeding scheme to reconstruct Fragaria ×ananassa using native clones of F. virginiana and F. chiloensis.
Citation: HortScience horts 45, 7; 10.21273/HORTSCI.45.7.1006
Breeding scheme to reconstruct Fragaria ×ananassa using native clones of F. virginiana and F. chiloensis.
Citation: HortScience horts 45, 7; 10.21273/HORTSCI.45.7.1006
There are several advantages to using the reconstruction approach. The most obvious is that genetic diversity will be greatly expanded within the F. ×ananassa gene pool. This will not only afford breeders with an expanded germplasm base, but unique epistatic interactions might appear that are of horticultural use. If sufficiently large populations of native material are screened, a breeder should be able to select for a wide group of positive characteristics with the minimum level of deleterious combinations. This method may also produce higher levels of genetic heterozygosity than conventional back-crossing methods, because the starting material will be more diverse than the typical breeding populations.
Of course, there are several potential disadvantages to the “reconstruction” approach. When a high number of traits are being selected simultaneously, numerous linked genes with negative impacts are often carried in the breeding population (Galletta et al., 1989). In back-crossing strategies, deleterious associations are less of a problem, simply because fewer genes are being selected. However, F. chiloensis and F. virginiana have been interbred on numerous occasions without any reports of chromosomal, physiological, or morphological abnormalities (Darrow, 1966, Luby et al., 2008). The existence of separate sexes in the wild species (dioecy) might also present a problem, although sex can be controlled through the selection of hermaphroditic parents or by incorporating the trait in later stages when existing hermaphroditic F. ×ananassa are used. The genetics of sex in the octoploid strawberries is regulated by a single locus or closely linked ones in which female is dominant to hermaphrodite, which is dominant to male (Ahmadi and Bringhurst, 1991; Spigler et al., 2008).
We recently undertook a study to test how well native clones of F. chiloensis and F. virginiana combine genetically. We intercrossed 15 genotypes of the two species that had high vigor, good yields, and resistance to common foliar diseases as well as large, attractive fruit that were unusually firm for wild clones (Luby et al., 2008). The progeny were planted in Minnesota and Ontario and evaluated for disease resistance, winter-hardiness, spring bloom date, fruit set, seed set, fruit size, and photoperiod sensitivity. We found that substantial breeding progress could be made by reconstructing F. ×ananassa if care is taken to select elite, complimentary genotypes of F. virginiana and F. chiloensis.
We report on another reconstruction effort that differed from the previous effort in two major ways: 1) we used a broader germplasm base that included elite clones of Chilean germplasm that were horticulturally superior to the F. chiloensis previously studied; and 2) we performed a round of improvement within species before making the interspecific crosses to maximize levels of genetic diversity. We did not sib the best performing interspecies F1 hybrids as previously suggested in Hancock et al. (1993a), because we felt few deleterious genes would actually segregate out in a single generation of inbreeding an octoploid. This approach yielded superior reconstructed F. ×ananassa with very broad adaptations, high yields, excellent fruit quality, and near commercial-sized fruit.
Materials and Methods
Previous germplasm evaluations.
A large number of studies have collected and evaluated wild germplasm. Members of our group have examined over 600 genotypes of F. virginiana ssp. glauca from the northern Rocky Mountains (Hokanson et al., 1993; Luby et al., 1992; Sakin et al., 1997) and nearly 2100 genotypes of F. virginiana ssp. virginiana from the central United States and Ontario along with a few representatives from Alaska, Alberta, New York, Pennsylvania, and western North Carolina (Dale et al., 1993; Luby and Stahler, 1993; Stahler et al., 1995). These clones have been evaluated for a wide range of traits, including photoperiod sensitivity, fruit size, sex, female fertility, disease resistance, nematode resistance, and environmental adaptations (Hancock et al., 1993b; Lewers et al., 2007; Pinkerton and Finn, 2005; Serçe and Hancock, 2005a, 2005b; Serçe et al., 2002).
We have also evaluated over 1000 wild genotypes of F. chiloensis from California for morphological variation and physiological tolerances (Hancock and Bringhurst, 1979, 1989; Jensen and Hancock, 1982). Others have screened more than 100 genotypes of F. chiloensis from Chile for phenotypic variation (Cameron et al., 1991; Gambardella et al., 2005; Lavín et al., 2005), physiological tolerances (Lavín, 1997), and fruit flavor characteristics (Carrasco et al., 2005; Nishizawa et al., 2005). Over 4000 F. chiloensis clones from the Pacific Northwest have been screened for resistance to aphids (Crock et al., 1982; Shanks and Barritt, 1974), black vine weevils (Doss and Shanks, 1987; Shanks et al., 1984), spider mites (Shanks and Moore, 1995; Shanks et al., 1995), nematodes (Pinkerton and Finn, 2005), foliar diseases (Luffman and Macdonald, 1993), physiological characteristics (Cameron and Hartley, 1990), and variation in reproductive traits (Dale et al., 1993; Luby et al., 1991; Luffman and Macdonald, 1993).
Recently, we selected an elite group of 38 wild strawberry accessions that appeared to be horticulturally superior and represented a wide range of the natural diversity found in F. chiloensis and F. virginiana (Hancock et al., 2001a, 2001b, 2001c). These genotypes were evaluated in five different states in the United States for a number of characteristics, including plant vigor, flower number, flowering date, harvest date, runner density, fruit weight and color, seed set, and foliar disease resistance. A much larger sample of 270 genotypes of wild F. virginiana and F. chiloensis from the National Clonal Germplasm Repository at Corvallis, OR, was also compared in a greenhouse at Michigan State University (MSU) for 14 morphological, reproductive, and production characteristics (Hancock et al., 2003). All the parents used in the current study were evaluated in these comparisons.
First round of crosses within F. virginiana.
We began our reconstruction work by selecting a small group of 16 elite F. virginiana clones that we felt represented the range of diversity in the North American material from cool climates (Table 1). Sixty-two families were generated among these at Simcoe, Ontario, in the winter of 1996, including High Falls 22 × LH 10-6, LH 28-1, LH 30-4, LH 39-15, LH 10-1, N8417, RH 18, RH 23, RH 30, and RH 43; Montreal River 10 × LH 10-6, LH 30-4, LH 10-1, N8417, RH 23, and RH 43; LH 10-6 × N8417, RH 23, and RH 43; LH 28-1 × Eagle 14; LH 30-4 × Eagle 14, Montreal River 10, Hemlo 2, Frederick 9, and RH 23; LH 39-15 × Eagle 14, Montreal River 10, Hemlo 2, Frederick 9, N8417, RH 23, RH 30, and RH 43; LH 40-4 × Eagle 14, Hemlo 2, and Frederick 9; RH 18 × Eagle 14, Montreal River 10, Hemlo 2, Frederick 9, LH 10-6, LH 28-1, LH 30-4, LH 39-15, and LH 10-1; RH 23 × Eagle 14, Hemlo 2, Frederick 9, and LH 10-1; RH 30 × Eagle 14, Montreal River 10, Hemlo 2, Frederick 9, LH 10-6, LH 28-1, LH 30-4, and LH 10-; and N8417 × Eagle 14, Hemlo 2, Frederick 9, LH 10-6, LH 28-1, LH 30-4, and LH 10-1. These genotypes were the same F. virginiana used in the previously published reconstruction study (Luby et al., 2008).
Characteristics of the wild octoploid Fragaria used in the reconstruction crosses.z
Seeds were germinated and grown to the four-leaf stage and six to 12 individuals from each family were set together in rows in May 1996 at Benton Harbor, MI. The plants were placed at 1.2 × 1.2-m spacing and their runners were trained by cross-cultivation into a 0.6 × 0.6-m square. In the summer of 1997, each genotype was subjectively evaluated for yield, fruit size, firmness, flavor, aroma, and color and the most elite genotypes were selected.
First round of crosses within F. chiloensis.
In 1999, we crossed three Chilean landraces (F. chiloensis subsp. chiloensis f. chiloensis) at East Lansing, MI, with superior combinations of high soluble solids, excellent flavor, strong aroma, and large size but poor color and low yields (2 BRA 1A—Chile, CA 1541—Peru, and NAH 4—Ecuador) with another ancient cultivar, Darrow 72, with many large red fruits, and several native clones (F. chiloensis subsp. chiloensis f. patagonica) with high fruit numbers (HM 1—Oregon and 2 MAR 1A—Chile), unusually large fruit size and dark internal color (Scotts Creek—California), or multiple disease resistance (RCP 37—California) (Table 1; Hancock et al., 2005). Of these genotypes, only HM 1 and RCP 37 were used in the previously published reconstruction study (Luby et al., 2008).
Seedlings of each family were germinated in East Lansing, MI, and when they had all produced their first true leaves, 20 to 40 representatives of each family were shipped overnight, unrefrigerated, to Corvallis, OR. On arrival, they were transplanted into a commercial potting mix in 4 × 4 × 4-cm pots in a greenhouse and grown to the four-leaf stage. In May, individuals from each family were set together in rows in the field at 1.2 × 1.2-m spacing and their runners were trained periodically by cross-cultivation into a 0.6 × 0.6-m square. In 2001, the plants were subjectively evaluated weekly during the next harvest season for yield, fruit size, firmness, flavor, and color and the most elite genotypes were selected (Table 2).
Characteristics of the elite selections from the first round of intraspecific crosses among selections from F. chiloensis and F. virginiana.
In Michigan, 20 to 40 plants of each family were also transplanted into a commercial potting mix in 4 × 4 × 4-cm pots and set in a randomized complete block design in a greenhouse (for full details see, Hancock et al., 2005). All flowers and runners were removed in 2000. In 2001, data were collected on each plant for bloom date, harvest date, crown number, runner number, peduncle length, flower number, fruit weight, skin color, and flesh color and the most elite genotypes were selected (Table 2). To enhance pollination, a camel hair brush was used to mix pollen from all open flowers in the greenhouse on a 3- to 4-d sequence. The results of this study were reported in Hancock et al. (2005).
Second round of crosses between elite F. virginiana and F. chiloensis hybrids.
The elite selections of F. virginiana and F. chiloensis (Table 2) were intercrossed in 26 combinations in 2005 (Table 3). Seedlings of each family were germinated at MSU and half were shipped to Oregon when they had their first true leaves. These plants were transferred to 4 × 4 × 4-cm pots and grown to the four-leaf stage. In May at Benton Harbor, MI, and Corvallis, OR, the plants were set at 1.2 × 1.2-m spacing and their runners were trained by cross-cultivation into a 0.6 × 0.6-m square. Ten to 76 plants of each family were set adjacent in rows.
Hybridizations made in the first round of interspecific crosses.
In 2006, five random plants were selected in each family at both locations and were given a subjective vigor rating from 1 (low) to 9 (high) based on the relative size of the mother plant and her daughters. These same plants were observed on a weekly basis to determine when the first flowers and ripe fruit appeared. When each of the selected genotypes was in full bloom, the number of flowers was counted on five randomly selected inflorescences from the mother and daughters of each genotype. When ≈50% of the fruit were ripe in each block of plants, the average percentage of ovules set as filled achenes was visually estimated on a random sample of five fruit from each genotype using a scale of 1 to 10 (representing 10% intervals). The weight of those same berries was determined along with the percentage of their surface area that was colored and the depth of color. Firmness was also evaluated on a subjective scale from 1 to 9. In August, leaf disease incidence was estimated on a subjective 1 to 9 scale.
In Michigan, a random sample of five fully ripe fruit was also evaluated for soluble solids, pH, and titratable acidity (TA). Soluble solids (SS) were determined using a handheld refractometer (Westover Model RHB-32; Southwest United Industries, Tulsa, OK). TA was determined from 10 mL of juice diluted to 100 mL with distilled water, titrated with 0.1 N sodium hydroxide (NaOH) to pH 8.2, and expressed as percentage citric acid (mass/mass) on a fresh weight basis.
The data were analyzed using SAS procedures (SAS Institute Inc., 2005). The descriptive statistics were calculated using TABULATE. The data from families grown in Michigan and Oregon were subjected to analysis of variance (ANOVA). ANOVAs were analyzed in two ways. First, combined analyses were performed on the data collected from each location. In this analysis, a random model was used in which all factors (location, replication, family, location × family interaction and error) were considered random. Because location × family interactions were found to be significant for most of the variables, ANOVAs were constructed for each location separately as well. In this analysis, the families were also considered random.
Results and Discussion
Crosses within F. virginiana.
Among the intercrossed F. virginiana clones, fruit sizes were generally quite small, but it was not unusual to find genotypes with extremely aromatic fruit that were 2 to 3 cm in width (data not shown). In subjective observations, the most useful parents for fruit size appeared to be High Falls 22 and Montreal River 10, whereas RH 30 appeared to produce the highest proportion of progeny with deep red fruit color. The families producing genotypes with the largest fruit sizes were: High Falls 22 × LH 10, High Falls 22 × N8417, LH 39 × Montreal River 10, LH 40 × Eagle 14, Montreal River 10 × N 8417, Montreal River 10 × RH 23, Montreal River 10 × RH 43, RH 30 × Frederick 9, RH 30 × LH 10, RH 18 × LH 28-1, RH 30 × Montreal River 10, RH 30 × LH 30, and RH 10 × Montreal River 10. From these elite families, five hybrids were selected for further use as reconstruction parents (Table 2): RV-1 (Montreal River 10 × RH 23), RV-2 (Montreal River 10 × RH 43), RV-3 (RH 30 × Montreal River 10), RV-4 (RH 30 × LH 30), and RV-5 (High Falls 22 × LH 10). Unfortunately, RV-5 was lost before further crosses were made. All the parents of these genotypes were used in the previously published reconstruction effort (Luby et al., 2008).
In another study investigating the genetics of day neutrality in octoploid strawberries, Serçe and Hancock (2005b) evaluated a number of F. virginiana crosses, including Frederick 9 × LH 50-4, Frederick 9 × RH 30, LH 50-4 × RH 30, Eagle 14 × Frederick 9, and Montreal River 10 × Frederick 9 with a large group of additional families of F. ×ananassa × F. virginiana, F. ×ananassa × F. chiloensis, and F. ×ananassa × F. ×ananassa. Although this study focused primarily on flowering patterns, the F. virginiana families were also subjectively evaluated for fruit size, fruit color, firmness, and flavor. Of these families, two genotypes were selected for further reconstruction work: RV-6 (RH 30 × LH 50-4) and RV-7 (Frederick 9 × LH 50-4) (Table 2). Although RH 30 and Frederick 9 were in the previously reported reconstruction effort (Luby et al., 2008), LH 50-4 was not. LH 50-4 was found to be one of the largest fruited clones of the F. virginiana ssp. glauca evaluated by Sakin et al. (1997).
Crosses within F. chiloensis.
Significant differences were observed across the F. chiloensis families for flowering and fruiting season, peduncle length, fruit size, color, and SS and number of fruit, crowns, and runners. This information was previously published in Hancock et al. (2005). NAH 4 and Pigeon Point transmitted the largest fruit size. Scotts Creek and 2 MAR 1A were the best parents for fruit color. HM 1 produced progeny with the highest flower numbers. Progeny of 2 MAR 1A and 2 BRA 1A were the latest fruiting. The cross Scotts Creek × NAH 3 yielded the highest number of selections with large fruit, excellent color, good yields, and high SS.
A number of families were selected as superior in Oregon, including: Scotts Creek × Darrow 72 (very early fruiting, large fruit with deep color), NAH 4 × Darrow 72 (very early, very large fruit), NAH 4 × 2 MAR 1A (very late fruiting, very high flower numbers, long peduncles, and large fruit), HM1 × Darrow 72 (early, high crown numbers), NAH 4 × CA 1541 (very early, large, and very well-colored fruit), NAH 4 × 2 BRA 1A (late fruiting, very high flower numbers, long peduncles, and very large fruit), and Darrow 72 × CA 1541 (very early, high flower numbers, and good fruit color).
The most superior families selected in Michigan were: Scotts Creek × 2 MAR 1A (late fruiting, high flower numbers, many crowns, and dark red fruit color), HM 1 × 2 BRA 1A (late fruiting, high crown numbers, and dark red fruit color), HM 1 × 2 MAR 1A (late fruiting, high crown numbers, high flower numbers, and dark red fruit color), NAH 4 × 2BRA 1A (late fruiting, very high flower numbers, long peduncles, and very large fruit), NAH 4 × 2 MAR 1A (very late fruiting, very high flower numbers, long peduncles, and large fruit), Scotts Creek × 2BRA 1A (late fruiting, high flower numbers, and deep-colored fruit), Scotts Creek × Darrow 72 (very early fruiting, large fruit with deep color), Scotts Creek × NAH 4 (very large, well-colored fruit), NAH 4 × Darrow 72 (very large fruit), and NAH 4 × RCP 37 (early fruiting, large, very dark red fruit).
Of the selections made in Michigan and Oregon, seven F. chiloensis hybrids were chosen as reconstruction parents: RC-1 (Scotts Creek × 2 MAR 1A), RC-2 (NAH 4 × BRA 1A), RC-3 (NAH 4 × Darrow 72), RC-4 (Scotts Creek × CA 1541), RC-5 (Scotts Creek × BRA IA), RC-6 (NAH 4 × 2 MAR 1A), and RC-7 (Scotts Creek × NAH 4) (Table 2). Unfortunately, RC-4 was not successfully used as a parent in subsequent crosses. RC-1, RC-2, RC-3, RC-6, and RC-7 were superior performers in both locations. FC-4 was one of the most elite genotypes in Oregon but not Michigan, whereas RC-5 was elite in Michigan but not Oregon. Another hybrid made by Dale and observed during fruiting in the MSU greenhouse was also selected as a reconstruction parent [RC-8 (Sable Beach 8 × Del Norte) × (Lions Head 3 × Del Norte)]. This hybrid had unusually high numbers of deep red, almost round fruit of better than average size for F. chiloensis, and its parents were almost disease-free in the field in Ontario (Dale, personal communication). The parents of all these RC selections were not used in the reconstruction effort reported by Luby et al. (2008).
Reconstruction crosses between F. virginiana × F. chiloensis.
Twenty-two families were evaluated in Michigan and Oregon (Table 3). The reconstructed F. ×ananassa were in general broadly adapted. Virtually all the hybrids survived at both locations and most had vigor ratings above 6.0, although the climatic adaptations of the original F. virginiana and F chiloensis selections were likely quite distinct. Typically, F. virginiana is found in habitats with very cold winters and hot summers, whereas F. chiloensis is located in climates that are more moderate in summer and winter.
Significant differences were observed between locations for all the traits except berry weight and flesh color (Table 4). Higher mean values were observed in Oregon for flower number, achene set, skin color, leaf disease, flowering date, and ripening date, whereas higher mean values were observed for vigor and firmness in Michigan (Table 5). Significant location × family interactions were also observed for vigor, flower number, skin color, firmness, leaf disease score, and flowering date but not for achene set, berry weight, or flesh color. Most of the intersite differences were likely the result of environmental variation, although vigor and firmness could have been influenced by individual investigator bias.
Mean squares, df, and significance for combined and separate analyses of variance of families obtained from Fragaria chiloensis × F. virginiana crosses, which were grown in Michigan and Oregon.
Values of several horticulturally important traits of families obtained from Fragaria chiloensis × F. virginiana crosses, which were grown in Michigan and Oregon.
Although none of the crosses produced genotypes of overall commercial quality, many of the families carried individual characteristics that were at or near cultivar quality (Table 5). Mean family values for vigor were as high as 9.0 in Michigan and 7.8 in Oregon. Highs for mean flower numbers per inflorescence were 13.0 in Oregon and 9.0 in Michigan. Mean fruit set exceeded 85% in both locations. Families were observed with averages of 100% external color and over 80% internal color in Michigan and Oregon. Mean firmness values exceeded 7.0 in some families at both locations. Flowering and fruiting dates varied by as much as 16 and 12 d at the individual sites, and disease ratings varied from 1.8 to 7.8. In Michigan, mean SS were found to be as high as 11.7%, whereas TA was as low as 0.71% citric acid (Table 6). SS/TA ratios varied from 2.93 to 13.57.
Values of soluble solids, pH and acidity of families obtained from Fragaria chiloensis × F. virginiana crosses, which were grown in Michigan.
Scott and Lawrence (1975) suggested that to get fruit quality back to industry standards is very difficult when using native germplasm, especially with the small, soft-fruited F. virginiana (Scott, 1959). Previous studies have found that at least three rounds of back-crossing back to F. ×ananassa were necessary to recover genotypes meeting commercial standards (Bringhurst and Voth, 1978, Scott and Lawrence, 1975). Although we did not find any progeny in this study that had commercial-sized fruit, one of the families had fruit that averaged 7.6 g in Michigan and 12.8 g in Oregon, and the largest fruited genotype in this family had primary fruit that averaged 19 g in Michigan (Fig. 1). This value far exceeded the weight of its progenitors found in other studies: Frederick 9 (1.9 g), LH 50-4 (2.1 g), Scotts Creek (3.2 g), and 2 MAR 1A (1.8 g) (Hancock et al., 2001c). Clearly, genes for large fruit are carried in small-fruited, native genotypes. The largest fruited families contained genes from the F. chiloensis genotypes Scotts Creek, 2 MAR 1A, and Adam Dale's complex hybrid, FC-8 [(Sable Beach × Del Norte) × (Lions Head 3 × Del Norte)] and the F. virginiana genotype LH 50-4. 2 MAR 1A is only modest sized by wild standards, but Scotts Creek, Dale's hybrid, and LH 50-4 are unusually large.
Some of the elite families combined a number of superior traits (Tables 5 and 6). FVC 8 {[(RH 30 × LH-50)] × [(Sable Beach 8 × Del Norte) × (Lions Head 3 × Del Norte)]} had unusually high flower numbers, seed set, and berry weight in Oregon and among the darkest flesh at both locations. FVC 10 (RH 30 × LH 50-4) × (Scotts Creek × NAH 4) were unusually large fruited at both locations with very high vigor and had unusually high seed set and firmness in Michigan and dark flesh color in Oregon. FVC 11 (Frederick 9 × LH 50-4) × (Scotts Creek × 2 MAR 1A) had exceptionally large fruit at both locations and unusually high seed set in Oregon. FVC 16 had very high vigor in Michigan and unusual flower numbers and fruit weight in Oregon. FVC 17 was very well colored and had high fruit numbers and large fruit weights in Michigan and Oregon and a high seed set in Oregon. FVC 18 had unusual vigor at both locations, high SS in Michigan, and high fruit numbers in Oregon. FVC 30 had very high vigor and low fruit acidity in Michigan and unusually high averages for seed set, fruit weight, and skin color in Oregon. FVC 9 (Montreal River 10 × RH 23) × (Scotts Creek × NAH 4) had unusually large fruit and deep flesh color in Michigan and exceptionally well-colored skin in Oregon.
All but one of these populations was represented in the individual selections made in Michigan and Oregon (Table 7). The only one missing was FVC 9. Overall, 13 elite genotypes were selected in Michigan and 12 in Oregon. Representatives of three families were selected in both locations, RVC 8, RVC 10, and RVC 30. Genotypes from FVC 10 and 16 were only selected in Oregon, whereas FVC 17, 18, and 28 were only selected in Michigan.
Elite FVC selections made in Michigan and Oregon from the elite Fragaria virginiana and F. chiloensis gene pools.
The genotypes selected offer a wide combination of subspecies, geographic locations, and adaptations. Probably the most diverse is FVC 11, which is composed of four subspecies from four distinct geographical regions, including [F. chiloensis ssp. pacifica (Scotts Creek from California), F. chiloensis ssp. patagonica (2 MAR 1A from Chile), F. virginiana ssp. virginiana (Frederick 9 from Ontario), and F. virginiana ssp. glauca (LH 50-4 from Montana)] (Figure 2). The original parents carried a broad range of adaptations, including cold-hardiness (LH 50-4), powdery mildew [Sphaerotheca macularis (Wallr.:Lind)], and leaf scorch resistance [Diplocarpon earlianum (Ellis & Everh.) F. A. Wolf] (Frederick 9), high drought and salt tolerance (Scotts Creek and 2 MAR 1A), and a high photosynthetic rate (2 MAR 1A) (Hancock et al., 2001a).
Fruit of FVC 11-3 from the cross of F. virginiana (Frederick 9 × LH 50-4) × F. chiloensis (Scotts Creek × 2 MAR 1A).
Citation: HortScience horts 45, 7; 10.21273/HORTSCI.45.7.1006
Fruit of FVC 11-3 from the cross of F. virginiana (Frederick 9 × LH 50-4) × F. chiloensis (Scotts Creek × 2 MAR 1A).
Citation: HortScience horts 45, 7; 10.21273/HORTSCI.45.7.1006
Fruit of FVC 11-3 from the cross of F. virginiana (Frederick 9 × LH 50-4) × F. chiloensis (Scotts Creek × 2 MAR 1A).
Citation: HortScience horts 45, 7; 10.21273/HORTSCI.45.7.1006
Conclusions
Like in the study of Luby et al. (2008), the reconstruction of F. ×ananassa by crossing elite genotypes of F. chiloensis and F. virginiana appears to be an effective strategy for strawberry improvement. Families and individuals were identified that had outstanding vigor, high productivity, seed set, fruit color, and firmness. None of the fruit were of commercial size, but one family, FVC 11, had individuals with fruit weights of almost 20 g, a fruit size deemed acceptable only 20 years ago outside of California. The reconstruction effort described here resulted in much larger fruited progeny than the one previously published (Luby et al., 2008). This greater success was likely the result of superior wild genotypes being used as parents. It is also possible that intercrossing geographically diverse selections within the species resulted in increased heterozygosity and greater heterosis in the reconstructed populations.
The question remains as to whether further intercrossing within reconstructed populations will yield new cultivars. It is possible that many of the heterotic interactions gained in the first interspecific cross may be lost in subsequent generations. We plan to take our elite FVC selections and intercross them and back-cross them to selected cultivars of F. ×ananassa to determine which strategy will be most effective in incorporating the diversity collected in the FVC selections in breeding improvement.
We also intend to generate a new family of wild F. virginiana × F. virginiana crosses to better capture the available diversity. We feel that we have represented F. chiloensis well with the inclusion of the South American material and Scotts Creek from California, but new F. virginiana clones have emerged that warrant testing as reconstruction parents (Hancock et al., 2001c, 2003). These include F. virginiana ssp. grayana (Vilm.ex L. Gay) CFRA 1170 and 1180 from Kentucky, F. virginiana ssp. virginiana Mill. CFRA 1385 from Quebec, and F. virginiana ssp. platypetala (Rydb.) Staudt CFRA 0110 from Oregon. All of these have fruit larger than any of the clones previously used. These will be intercrossed and then the most elite performing hybrids will be hybridized with our best FVC hybrids.
Literature Cited
Ahmadi, H. & Bringhurst, R.S. 1991 Genetics of sex expression in Fragaria species Amer. J. Bot. 78 504 514
Barritt, B.H. & Shanks C.H. Jr 1980 Breeding strawberries for resistance to the aphids Chaetosiphon fragaefolii and C. thomasi HortScience 15 287 288
Bringhurst, R.S. & Voth, V. 1978 Origin and evolutionary potentiality of the day-neutral trait in octoploid Fragaria Genetics 90 510
Bringhurst, R.S. & Voth, V. 1984 Breeding octoploid strawberries Iowa Sta. J. Res. 58 371 381
Cameron, J.S. & Hartley, C.A. 1990 Gas exchange characteristics of Fragaria chiloensis genotypes HortScience 25 327 329
Cameron, J.S. , Sjulin, T.M. , Shanks, C.H. & Muñoz, C.E. 1991 Collection of Fragaria chiloensis in central and southern Chile 108 118 Dale A. & Luby J. The strawberry into the 21st century Timber Press, Inc Portland, OR
Carrasco, B. , Hancock, J.F. , Beaudry, R.M. & Retamales, J.B. 2005 Chemical composition and inheritance patterns of aroma in Fragaria ×ananassa and F. virginiana progenies HortScience 40 1649 1650
Crock, J.E. , Shanks, C.H. & Barritt, B.H. 1982 Resistance in Fragaria chiloensis and F. ×ananassa to the aphids Chaetosiphon fragaefolii and C. thomasi HortScience 17 959 960
Dale, A. , Daubeny, H.A. , Luffman, M. & Sullivan, J.A. 1993 Development of Fragaria germplasm in Canada Acta Hort. 348 75 80
Darrow, G.M. 1966 The strawberry history, breeding and physiology Holt, Rinehart and Winston New York, NY
Daubeny, H.A. 1990 Strawberry breeding in Canada HortScience 25 893 894
Doss, R.P. & Shanks, C.H. 1987 The influence of leaf pubescence on the resistance of selected clones of beach strawberry [Fragaria chiloensis (L.) Duchesne] to adult black vine weevils (Otiorhynchus sulcatus F.) Sci. Hort. 34 47 54
Galletta, G.J. , Draper, A.D. & Maas, J.L. 1989 Combining disease resistance, plant adaptation and fruit quality in breeding short day and day-neutral strawberries Acta Hort. 265 43 51
Gambardella, M. , Cadavid, A. , Díaz, V. & Pertuzé, R. 2005 Molecular and morphological characterization of wild and cultivated native Fragaria in southern Chile HortScience 40 1640 1641
Hancock, J.F. 1999 Strawberries CABI Publishing Wallingford, UK
Hancock, J.F. & Bringhurst, R.S. 1979 Ecological differentiation in perennial, octoploid species of Fragaria Amer. J. Bot. 66 367 375
Hancock, J.F. & Bringhurst, R.S. 1989 Yield component interactions in wild populations of California Fragaria HortScience 23 889 891
Hancock, J.F. , Callow, P.W. , Dale, A. , Luby, J.J. , Finn, C.E. , Hokanson, S.C. & Hummer, K.E. 2001a From the Andes to the Rockies: Native strawberry collection and utilization HortScience 36 221 225
Hancock, J.F. , Callow, P.W. , Serçe, S. & Schilder, A.M.C. 2001b Relative performance of strawberry cultivars and native hybrids on fumigated and nonfumigated soil in Michigan HortScience 36 136 138
Hancock, J.F. , Finn, C.E. , Hokanson, S.C. , Luby, J.J. , Goulart, B.L. , Demchak, K. , Callow, P.W. , Serçe, S. , Schilder, A.M.C. & Hummer, K.E. 2001c A multistate comparison of native octoploid strawberries from North and South America J. Amer. Soc. Hort. Sci. 126 579 586
Hancock, J.F. , Callow, P.W. , Serçe, S. & Son, P.Q. 2003 Variation in the horticultural characteristics of native Fragaria virginiana and F. chiloensis from North and South America J. Amer. Soc. Hort. Sci. 128 201 208
Hancock, J.F. , Dale, A. & Luby, J.J. 1993a Should we reconstitute Fragaria ×ananassa? Acta Hort. 348 85 93
Hancock, J.F. , Harrison, R.E. , Luby, J.J. & Hancock, J.F. 1993b Morphological variation in Fragaria virginiana from the Rocky Mountains Acta Hort. 348 94 101
Hancock, J.F. , Drake, C.A. , Callow, P.W. & Serçe, S. 2005 Genetic improvement of the Beach strawberry HortScience 40 1644 1645
Hancock, J.F. , Maas, J.L. , Shanks, C.H. , Breen, P.J. & Luby, J.J. 1990 Strawberries 489 546 Moore J.N. & Ballington J.R. Genetic resources in temperate fruit and nut crop Internat. Soc. Hort. Sci Wageningen, The Netherlands
Hancock, J.F. , Serçe, S. , Portman, C.M. , Callow, P.W. & Luby, J.J. 2004 Taxonomic variation among North and South American subspecies of Fragaria virginiana Miller and F. chiloensis (L) Miller Can. J. Bot. 82 1632 1644
Hokanson, K.E. , Harrison, R.E. , Luby, J.J. & Hancock, J.F. 1993 Morphological variation in Fragaria virginiana from the Rocky Mountains Acta Hort. 348 63 68
Jensen, D. & Hancock, J.F. 1982 A multivariate analysis of California strawberries Bull. Torrey Bot. Club 109 136 148
Lavín, A. 1997. Caracterización botánica, fisiológica y agronómica de ecotipos chilenos de Fragaria chiloensis (L.) Duch., recolectados en las X and XI regiones de Chile. Informe final. Proyecto Fondecyt 1940083 (1996). Centro Experimental Cauquenes (INIA).
Lavín, A. , Barrera, C. , Retamales, J.B. & Maureira, M. 2005 Morphological and phenological characterization of 52 accessions of Fragaria chiloensis (L.) Duch HortScience 40 1637 1639
Lewers, K.S. , Turechek, W.W. , Hokanson, S.C. , Maas, J.L. , Hancock, J.F. , Serçe, S. & Smith, B.J. 2007 Evaluation of elite native strawberry germplasm for resistance to anthracnose crown rot disease caused by Colletotrichum species J. Amer. Soc. Hort. Sci. 132 842 849
Luby, J.J. , Dale, A. , Hancock, J.F. & Serçe, S. 2008 Reconstructing Fragaria ×ananassa utilizing wild F. virginiana and F. chiloensis: Inheritance of winter injury, photoperiod sensitivity, fruit size, gender, female fertility and disease resistance in hybrid progenies Euphytica 163 57 65
Luby, J.J. , Hancock, J.F. & Ballington, J.R. 1992 Collection of native strawberry (Fragaria ssp.) germplasm in the Pacific Northwest and Northern Rocky Mountains of the USA HortScience 27 12 17
Luby, J.J. , Hancock, J.F. & Cameron, S. 1991 Expansion of the strawberry germplasm base in North America 66 75 Dale A. & Luby J.J. The strawberry into the 21st century Timber Press Portland, OR
Luby, J.J. & Stahler, M.M. 1993 Collection and evaluation of Fragaria virginiana in North America Acta Hort. 345 49 54
Luffman, M. & Macdonald, P.J. 1993 Fragaria germplasm at the Canadian clonal germbank Acta Hort. 348 102 108
Nishizawa, T. , Nagasawa, S. , Mori, Y. , Kondo, Y. , Sasaki, Y. , Retamales, J.B. & Lavín, A. 2005 Characteristics of soluble sugar accumulation in commercially grown Fragaria chiloensis HortScience 40 1647 1648
Pinkerton, J. & Finn, C.E. 2005 Responses of strawberry species and cultivars to the root- lesion and northern root-knot nematodes HortScience 40 33 38
Sakin, M. , Hancock, J.F. & Luby, J.J. 1997 Identifying new sources of genes that determine cyclic flowering in Rocky Mountain populations of Fragaria virginiana ssp. glauca Staudt J. Amer. Soc. Hort. Sci. 122 205 210
SAS Institute Inc 2005 SAS users guide; SAS/STAT, version 6 SAS Institute Inc Cary, NC
Scott, D.H. 1959 Size, firmness and time of ripening of fruit of seedlings of Fragaria virginiana Duch. crossed with cultivated strawberry varieties Proc. Amer. Soc. Hort. Sci. 74 388 393
Scott, D.H. & Lawrence, F.J. 1975 Strawberries 71 97 Janick J. & Moore J.N. Advances in fruit breeding Purdue University Press West Lafayette, IN
Serçe, S. & Hancock, J.F. 2005a The temperature and photoperiod regulation of flowering in Fragaria chiloensis, F virginiana, and F ×ananassa genotypes Sci. Hort. 103 167 177
Serçe, S. & Hancock, J.F. 2005b Inheritance of day-neutrality in octoploid species of Fragaria J. Amer. Soc. Hort. Sci. 130 580 584
Serçe, S. , Callow, P.W. , Ho, J.J. & Hancock, J.F. 2002 High temperature effects on CO2 assimilation rate in genotypes of Fragaria ×ananassa, F chiloensis and F virginiana J. Amer. Pomol. Soc. 56 57 62
Shanks, C.H. & Barritt, B.H. 1974 Fragaria chiloensis clones resistant to the strawberry aphid HortScience 9 202 203
Shanks, C.H. & Moore, P.P. 1995 Resistance to two-spotted spider mite and strawberry aphid in Fragaria chiloensis, F. virginiana, and F. ×ananassa clones HortScience 30 596 599
Shanks, C.H. , Chandler, C.K. , Show, E.D. & Moore, P.P. 1995 Fragaria resistance to spider mites at three locations in the United States HortScience 30 1068 1069
Shanks C.H. Jr , Chase, D.L. & Chamberlain, J.D. 1984 Resistance of clones of wild strawberry, Fragaria chiloensis, to adult Otiorhynchus sulcatus and O. ovatus (Coleoptera: Curculionidae) Environ. Entomol. 13 1042 1045
Sjulin, T.M. & Dale, A. 1987 Genetic diversity of North American strawberry cultivars J. Amer. Soc. Hort. Sci. 112 375 385
Spigler, R.B. , Lewers, K.S. , Main, D.S. & Ashman, T.-L. 2008 Genetic mapping of sex determination in a wild strawberry, Fragaria virginiana reveals earliest form of sex chromosome Heredity 101 507 517
Stahler, M.M. , Ascher, P.D. , Luby, J.J. & Roelfs, A.P. 1995 Sexual composition of populations of Fragaria virginiana (Rosaceae) collected in Minnesota and western Wisconsin Can. J. Bot. 73 1457 1463
Staudt, G. 1999 Systematics and geographic distribution of the American strawberry species: Taxonomic studies in the genus Fragaria (Rosaceae: Potentilleae) University of California Press Berkeley, CA
