Phenotypic and Genotypic Analysis of a U.S. Native Fine-leaved Festuca Population Reveals Its Potential Use for Low-input Urban Landscapes

Continued reduction in limited natural resources worldwide increasingly necessitates the incorporation of low-maintenance and low-input plant materials into urban landscapes. Some fine-leaved Festuca grass species have been used in formal gardens and native urban landscapes because of their inherent tolerance to abiotic stresses, but native, ornamental types (tall and non-spreading with multicolored culms and panicles) are not common in landscapes of the western United States. A native fine-leaved Festuca collection made in Montana (designated FEID 9025897) by the U.S. Natural Resources Conservation Services possesses such ornamental characteristics but has not been evaluated for its horticultural potential. Therefore, a study was designed to assess its phenotypic and genotypic attributes by cloning 270 FEID 9025897 plants and evaluating them along with native F. idahoensis and F. ovina PIs (five) and commercial checks (five) for genetic diversity and plant morphology for 2 years (2010–11). Plant genetic constitution was determined using amplified fragment length polymorphism (AFLP) analysis. Plant height, width, biomass, relative vigor (visual rating of 0 = dead to 5 = green, abundant growth), persistence (number of plants alive per plot), and regrowth after clipping (visual rating of 0 = none to 5 = most) were estimated by evaluation of plants under replication at Hyde Park, UT. Based on AFLP-based coancestry analysis, FEID 9025897 plants possessed considerable genetic affinities with F. idahoensis. Morphological traits as averaged over both years varied in height (13.9 to 105.0 cm), width (9.9 to 66.2 cm), biomass (0 to 170.4 g), vigor (0.2 to 4.7), persistence (0 to 3.9), and regrowth (0 to 4.0). Based on these differences, 19 (7%) FEID 9025897 plants were identified for their ornamental potential that possessed multicolored (red, orange, and yellow) culms and varied in morphology with 2-year means of height (79.8 cm), width (45.2 cm), biomass (88.5 g), vigor (2.9), persistence (1.8), and regrowth (3.7). The popularity of ornamental grasses for use in urban landscapes, parks, median strips, parking lot borders, and for erosion control on slopes has increased in recent years (Loram et al., 2008; Wilson and Knox, 2006). This increase is partially the result of the broad range of flowering times, panicle size, leaf width and color, and plant form of modern cultivars, which allows for their use in horticultural applications ranging from formal gardens to informal native urban landscapes (Wilson and Knox, 2006). Such plantings are considered to be an integral part of ecological systems worldwide, where they provide immeasurable aesthetic value (Waliczek et al., 2005; York, 2001) and considerable economic return to a range of horticultural industries (Hughes and Hinson, 2000; Johnson and Christensen, 1995; Meyer, 2011; Rathwell et al., 1995). Native and non-native grasses are considered central to many U.S. urban landscapes (Beard and Green, 1994; Fender, 2008). In the western United States [U.S. Department of Agriculture (USDA) hardiness zones 3 to 5; annual precipitation 254 to 610 mm], little bluestem (Schizachyrium scoparium), western wheatgrass [Pascopyrum smithii (synonymous with Agropyron smithii)], prairie junegrass [Koeleria macrantha (synonymous with K. cristata)], needlegrass (Stipa spartea), buffalograss (Buchloe dactyloides), and blue grama (Bouteloua gracilis) are being increasingly used for low-input urban horticultural applications (Wilson, 2011). To a lesser extent, fine-leaved fescue species are being considered for these low-input situations (Ruemmele et al., 2003) and in more purely ornamental applications [e.g., blue fescue (F. glauca); Neal and Senesac, 1991]. The genus Festuca contains 300 genetically diverse, perennial species with both wide and narrow leaves and can have tufted or rhizomatous growth habits. Several species possess drought tolerance and have attributes useful for lowinput applications (Ruemmele et al., 2003). For instance, some Festuca species are used worldwide as turfgrass {e.g., F. rubra (red fescue), F. ovina (sheep fescue), F. arundinacea [tall fescue (synonymous with Schedonorus arundinaceus)]}, for roadsides [e.g., F. ovina, F. trachyphylla (hard fescue)], and in rangeland restoration {e.g., F. ovina, F. scabrella [altai fescue (synonymous with F. altaica)]} (Johnson 2003; Ruemmele et al., 1995). Blue fescue is the only fine-leaved fescue widely used in the western United States as an ornamental for lowinput applications (e.g., ‘Elijah Blue’ and ‘Casca11’; Blom, 2013). Because the taxonomy of Festuca grasses is ambiguous (Aiken et al., 1992; Darbyshire and Pavlick, 2007; Huff and Palazzo, 1998; Loureiro et al., 2007), there have been several taxonomic classifications of fine-leaved Festuca species (e.g., Darbyshire and Pavlick, 2007; Hultén and Fries, 1986; Received for publication 10 Mar. 2014. Accepted for publication 24 Sept. 2014. Mention of a trade name, proprietary product, or specific equipment does not constitute a guarantee or warranty by the U.S. Department of Agriculture and does not imply its approval to exclusion of other products that may be suitable. Corresponding author. E-mail: [email protected]. 706 J. AMER. SOC. HORT. SCI. 139(6):706–715. 2014. Tutin et al., 1993). The taxonomic treatment of Catalán et al. (2004) is used herein in the broad sense along with the treatment of Darbyshire and Pavlick (2007) because of its historical context to North American flora and its recognition of Festuca species worldwide. Where possible, other species designations are also cited to provide continuity with other taxonomic treatments. Several fine-leaved Festuca species form a closely related polyploid aggregate often called the Ovina Complex with a basic chromosome number of 2n = 2x = 14 (Darbyshire and Warwick, 1992; Jones et al., 2008; Ma, 2012; Pavlick, 1984). The Ovina Complex is composed of F. valesiaca [volga fescue (ploidy of 2x to 6x)], F. filiformis [hair fescue (2x)], F. idahoensis [idaho fescue (4x)], F. ovina (2x to 6x), F. roemeri [roemer’s fescue (4x to 6x)], F. trachyphylla (6x), and F. viviparoidea ssp. viviparoidea [northern fescue (4x to 6x)]. Festuca idahoensis, F. roemeri, and F. viviparoidea ssp. viviparoidea are species considered native to the western United States. In 1982, the U.S. Natural Resources Conservation Services (NRCS) Bridger Plant Materials Center (BPMC) collected seed from a native fine-leaved Festuca population in a semiarid region near Busby, MT, and designated it FEID 9025897. Because historical information about this population is limited, very little is known about the genetic or morphological characteristics of this plant material. Using five plants of FEID 9025897 in an AFLP-based analysis, Jones et al. (2008) reported that this population possessed relatively close genetic affinities with F. ovina. In 2009, the USDA Forage and Range Research Laboratory (FRRL) identified plants of FEID 9025897 in a Logan, UT, field nursery that were vigorous, relatively tall, and possessed desirable horticultural characteristics [J.E. Staub, unpublished data (Fig. 1)]. These preliminary observations suggested that selections from this population might have potential for low-input urban horticultural applications in the western United States. Thus, a replicated field trial was designed to assess the inherent phenotypic variation of the population and identify potentially useful and/or novel genotypes that require further multi-location evaluations for urban landscape applications. As a secondary objective, AFLP markers were used to characterize the genetic constitution of the population to provide information on its putative genetic ancestry and relatedness to F. idahoensis and F. ovina. Materials and Methods GERMPLASM. Seed of FEID 9025897 was originally collected (more than 10 plants) by the BPMC on the Charles E. Helvey Ranch east of the Rosebud River (T7S R39E NW1/4Sec 3; lat. 45 31#39$ N, long. 106 58#25$ W) in Big Horn County, MT. This collection site has since been disturbed and the original plant material is no longer available. Seed of the initially collected population was field increased by the BPMC in 1994 and received by the FRRL in 2004. Historical information about the population is limited and the ancestry of plants in this population is unknown. Fine-leaved F. ovina and F. idahoensis populations exist throughout this region of Montana (Darbyshire and Pavlick, 2007) and, thus, could be progenitors of FEID 9025897. Seven F. idahoensis accessions, ‘Nezpurs’ [PI 601053 (ploidy of 4x)], ‘Joseph’ [PI 601054 (4x)], PI 344597 (4x), PI 344604 (4x), PI 344609 (4x), PI 344614 (4x), and PI 344631 (4x), and three F. ovina cultivars, Durar [PI 578732 (6x)], Covar [PI 578733 (2x)], and Bighorn [PI 549274 (6x)], were used as checks (Table 1). Because the original plant material no longer exists, these checks were chosen to represent the genetic variation of these species in the United States, especially the F. ovina cultivars because they are widely distributed. Seed of all checks were obtained from the USDA Agricultural Research Service (ARS) Germplasm Resources Information Network (GRIN, 2013). Although ‘Durar’ is sometimes classified under several species names (e.g., F. trachyphylla, F. lemanii, F. ovina, F. brevipila), here it will be referred to as F. ovina sensu lato (sheep fescue in the broad sense) following Jones et al. (2008) because it is indistinguishable from F. ovina (Ma, 2012). PLOT ESTABLISHMENT AND MAINTENANCE. In Jan. 2008, seeds from the FEID 9025897 population were germinated on blotter paper, and then seedlings were planted in 164-mL nursery containers (Conetainers; Stuewe and Sons, Tangent, OR) containing a mixture (v/v) of 3:1 pumice and peatmoss in a greenhouse in Logan, UT. Seedlings were grown at 21 C day/ 15 C night with supplemental light supplied by high-pressure sodium lights [average irradiance = 400 W (1800 mmol m s); Sun System III, Sunlight Supply, Vancouver, WA] at a relative humidity between 50% and 70%. Seedlings were fertigated daily with 20 mg mL of 20N–8.7P–16.6K water-soluble fertilizer (Peters Professional; Scotts, Marysville, OH) to provide 4.0 mg mL nitrogen, 1.7 mg mL phosphorus, and 3.3 mg mL potassium. A total of 270 plants from the FEID 9025897 population were vegetatively cloned (12 clones per plant) for a replicated trial. Commercial cultivars (Bighorn, Joseph, Nezpurs, Durar, Covar) and PI accessions [10 checks (Table 1)] used as checks were grown from seed as described previously for individuals Fig. 1. Examples of selections for red coloration and plant habit within a putative Festuca idahoensis · F. ovina population, FEID 9025897 originating in Montana: (A) upright and inclined plant habits, (B) upright habit, (C) demonstrates pendulant inflorescences. J. AMER. SOC. HORT. SCI. 139(6):706–715. 2014. 707 of the FEID 9025897 population. Because of their mode of reproduction (intermating with limited self-pollination), the checks must be considered heterogeneous and heterozygous. Thus, to represent their total inherent genetic variability during morphological evaluation, checks were established from seed and not from clones. Clones of the population and seedlings of the checks were transplanted in May 2008 to a field nursery in Hyde Park, Cache County, UT (lat. 41 48#41.22$ N, long. 111 49#18.83$ W) 8 km north of Logan, UT (elevation = 1383 m), where the average annual precipitation during the experiment (2009–11) was 508 mm (average 20-year precipitation is 453 mm). The soil type was a Nibley fine mixed mesic (Aquic Argiustolls) having a neutral to slightly acidic pH (USDA, 2014). Plants were arranged in a randomized complete block (RCB) design with four clones or seedlings per plot in three replications and were spaced 0.5 m within the rows and 0.75 m between rows ( 26,667 plants/ha) with additional plants used as end and side borders. Although plants were given water at transplanting, no water or fertilizer was applied during the experiment, and plots were hand-weeded each year from May to August. Broadleaf weeds were also controlled with herbicide [mixture of 30.56% 2,4-dichlorophenoxyacetic acid (2,4-D), 8.17% mecoprop-p, and 2.77% dicamba (MEC Amine-D; Loveland Products, Greeley CO)] application once in April or May of each year at a rate of 3.0 g ha 2,4-D, 0.8 g ha mecoprop-p, and 0.3 g ha dicamba. PHENOTYPIC TRAIT EVALUATION. On 18 June 2010 and 12 May 2011, the relative plant vigor of all accessions was assessed using a 11-point visual rating scale from 0 to 5 (0.5 as units), where plant vigor (size, color, and transition from winter to spring growth) was defined as 0 = dead plant, 2.5 = plant possesses moderate biomass with some green foliage (tussock evident), and 5 = green plants having comparatively abundant above-ground biomass. On 28 June 2010 and 20 June 2011, the height (centimeters) of each plant was measured as the distance from the plant base (soil surface) to the top of the highest floret at full anthesis (florets were gathered and straightened upward for measurement). Leaves and seed spikes were gathered together by hand and harvested 10 cm above ground when culms were dry and just before seed shattering (seed maturity), and then oven-dried at 60 C to estimate plant biomass as total above-ground dry weight (grams). After above-ground harvesting, plant width was measured as the diameter of the remaining leaves and stems. For plots harvested in 2010, dried florets of each plant were mechanically threshed to separate mature seeds and chaff (i.e., stalks and poorly developed or aborted seeds) for dry weight determination. Nevertheless, plants in a great majority of plots ( 97%) did not appear to produce mature seed (i.e., florets were flat with no or vestigial embryos), an observation that was subsequently confirmed visually in 2011. Thus, seed was not threshed in 2011 and seed yield data are not presented. On 18 Nov. 2010 and 4 Nov. 2011, plants were visually rated on the same 11-point scale for regrowth, where 0 = no regrowth, 2.5 = comparatively moderate regrowth, and 5 = most regrowth. Persistence was determined by counting the number of plants alive within each plot (0 to 4) at the time when regrowth ratings were taken. Culm coloration was initially not one of the traits evaluated, but observations were noted on a few specific plots in July 2010 that were striking in color. Because brown/tan, yellow, orange, and maroon colors were also observed along the length of culms (Fig. 1) in July 2011, more extensive notes were recorded on specific plots, but the entire trial was not formally evaluated for this trait. Although selections were largely based on traits other than culm color, this was an important consideration because selections were made only from plots with attractive culm color. Seed viability was not a characteristic originally considered for evaluation of the FEID 9025897 population, but it became necessary to measure this trait to address seed sterility. To determine relative seed viability, seed germination was measured in four cultivar checks (Covar, Durar, Joseph, and Nezpurs) and 34 FEID 9025897 individuals. This population subset of FEID 9025897 was chosen to represent the range of phenotype (morphological traits) and genotype (AFLP-based ancestry coefficients; see subsequently) detected. Germination for each accession was measured as the emergence of the radicle in three samples (replications) taken from each of three replications harvested in 2010. One hundred seeds from each plot were placed on seed germination paper (Steel Blue blotter paper, 178 g m; Anchor Paper Co., St. Paul, MN) inside plastic germination boxes (110 · 110 · 29 mm), moistened with 12 mL of distilled water, and then misted once with chlorothalonil fungicide (0.9% solution, Daconil; TechPac, Lexington KY). Seeds were then stored at room temperature ( 23 C) and the number of seeds germinated was recorded daily for 4 weeks. The germination rate was calculated as the mean percent germination of the three replications. PHENOTYPIC TRAIT ANALYSIS. Morphological trait data (over 2 years) were analyzed using a linear mixed models analysis Table 1. PI accessions obtained from the U.S. Department of Agriculture, Agricultural Research Service, Genetic Resources Information Network (GRIN) used as checks for morphological evaluation of a U.S. native Festuca idahoensis · F. ovina grass population. Accession Scientific name Common name Cultivar Origin ID type PI 344597 F. idahoensis Idaho fescue Idaho Collector PI 344604 F. idahoensis Idaho fescue Idaho Collector PI 344609 F. idahoensis Idaho fescue Washington Collector PI 344614 F. idahoensis Idaho fescue Idaho Collector PI 344631 F. idahoensis Idaho fescue Montana Collector PI 549274 F. ovina Sheep fescue Bighorn Oregon Cultivar PI 601054 F. idahoensis Idaho fescue Joseph Idaho Cultivar PI 601053 F. idahoensis Idaho fescue Nezpurs Idaho Cultivar PI 578732 F. ovina Sheep fescue Durar Oregon Cultivar PI 578733 F. ovina Sheep fescue Covar Turkey Cultivar Accessions are heterogeneous and heterozygous populations obtain from GRIN. ID type as per classification under the GRIN system. 708 J. AMER. SOC. HORT. SCI. 139(6):706–715. 2014. under which residuals for all traits were tested for normality using PROC UNIVARIATE in SAS software (Version 9.3 for Windows; SAS Institute, Cary, NC). Homogeneity of variance was evaluated by plotting the residuals against the predicted values. Year effects were tested using a repeated-measures model with compound symmetry covariance structure. Year was assumed to be a fixed effect because inferences for years were limited to the 2 years (2010 and 2011) under evaluation (Smith and Casler, 2004) and these 2 years may not be representative of multiple growing seasons of perennial species. Accession was assumed to be fixed because inferences were made for the specific accessions evaluated. Replicates were considered random effects. Year and accession main effects and the year · accession interaction effect were first tested under the repeated-measures model using PROC MIXED in SAS software. If year and/or year year · accession interaction effects were significantly different from zero, accession effects were then tested within each year as a RCB design using PROC MIXED in SAS. Accession means were separated using Fisher’s protected least significant difference test by applying the lsmeans statement in SAS. Multivariate principal component analysis (PCA) was performed on lsmean values of all traits using PROC FACTOR in SAS to define accession relationships and to identify those traits that led most to accession discrimination (Kutner et al., 2004). Pearson product-moment correlation coefficients were produced using PROC CORR in SAS to assess the strength of associations among the traits examined (Székely et al., 2007). DNA EXTRACTION AND AFLP ANALYSIS. Leaf samples from at least 15 plants of each accession were collected, lyophilized, and then ground into fine powder with zinc beads inside extraction tubes using a mixer mill (Model MM 300; F. Kurt Retsch, Haan, Germany). Total cellular DNA was extracted using a DNA extraction kit (DNeasy Plant Mini Kit; QIAGEN, Venlo, The Netherlands) according to the manufacturer’s instructions and quantified with a spectrophotometer (NanoDrop ND-1000; Thermo Fisher Scientific, Waltham, MA). The AFLP polymerase chain reaction (PCR) amplification procedure was performed following Vos et al. (1995) with fluorescently labeled primers, size fractionating, and fragment size analysis according to Jones et al. (2008). The same E.AC/ M.CT primers defined by Jones et al. (2008) were used for preamplification, whereas the primer combinations of E-ACAC/ M-CTAC, E-ACAG/M-CTCA, E-ACCA/M-CTAG, E-ACCA/ M-CTTC, E-ACCT/M-CTCT, E-ACTC/M-CTTG, E-ACT/MCTA, E-ACT/M-CTG, E-ATA/M-CAA, and E-AGG/M-CGC were used for selective amplification. Approximately 4% of the samples were replicated to identify reproducible marker bands and determine marker error rates. Only the most distinct (i.e., bright) and reproducible (i.e., consistent during repeated PCR amplicon analysis) bands were used for analysis. AFLP-BASED CLUSTER ANALYSIS. Genetic relationships among individuals in the population and checks for each species (Table 1) were characterized using AFLP fragment data matrices [AFLP band present (1) or absent (0)] with Bayesian cluster analysis according to Falush et al. (2003) and Pritchard et al. (2000) using the program Structure Version 2.3.1 (Pritchard et al., 2000). Genetic relationships were defined by three iterations of Markov chain modeling (10,000 burn-in period, 100,000 Monte Carlo convergence permutations) where K = 2 using the recessive model to determine the proportion of the two parental species (F. idahoensis and F. ovina) for each individual in the population. PLOIDY ESTIMATION. The ploidy level of 16 selected FEID 9025897 plants (Table 2) and all PI accession checks was determined by flow cytometry using a 4’,6-diamidino-2-phenylindole staining kit (CyStain ultraviolet precise P; Partec, Munster, Germany) to extract and stain nuclei according to the manufacturer’s protocol. Rapidly growing and immature leaves from each accession were harvested, pooled with leaves from a diploid (2n = 2x = 14) control [W6 30595 (F. valesiaca); Ma, 2012], and macerated in 400 mL of extraction buffer. After incubation for 30 to 60 s, the macerated tissue solution was filtered through a disposable filter (30 mm CellTrics; Partec), 1.6 mL of staining solution was added, and the resulting cell suspension was taken to the Utah State University Center for Integrated BioSystems (Logan) for fluorescence-based flow cytometry analysis (FACSAria II; Becton, Dickinson and Co., Franklin Lakes, NJ). Several checks with known ploidy levels were included to validate flow cytometry results. Plant Variety Protection certificates available from the GRIN indicate that ‘Joseph’ (PI 601054, PVP no. 8400003) and ‘Nezpurs’ (PI 601053; PVP no. 8400002) are 4x, whereas ‘Bighorn’ (PI 549274, PVP no. 8800064) is 6x (GRIN, 2013). The reported ploidy of ‘Covar’ is 2x (Huff and Palazzo, 1998), whereas Ma (2012) designated ‘Durar’ as 6x and PI 344631 as 4x. To further access methodological repeatability, 15% of the samples were randomly chosen and evaluated at least twice. If a notable inconsistency was observed in any sample examined, the cytometry analysis was repeated until unambiguous results were obtained. If a sample:control ratio equaled 1, 1.5, 2, or 3, the sample was declared as a diploid, triploid, tetraploid, or hexaploid, respectively.