JFS S: Sensory and Food Quality

Effects of Seasonal Variation on Sensory Properties and Total Phenolic Content of 5 Lettuce Cultivars
MARISA L. BUNNING, PATRICIA A. KENDALL, MARTHA B. STONE, FRANK H. STONAKER, AND CECIL STUSHNOFF
ABSTRACT: Butterhead, crisphead, green leaf, red leaf, and romaine types of lettuce (Lactuca sativa L.) are all commonly available in U.S. markets. Sensory properties of lettuce may vary in response to environmental factors that often fluctuate widely throughout the growing season. Bitterness is generally thought to increase in lettuce grown at higher temperatures and may be related to phenolic content. This study evaluated sensory properties and total phenolic content of 5 lettuce cultivars harvested early, midway, and late in the growing season and investigated possible correlations with environmental temperature and light intensity indexes. Thirty panelists rated bitterness, appearance, flavor, texture, and overall acceptability of Crisp and Green (green leaf ), Crispino (crisphead), Green Forest (romaine), Lochness (butterhead), and Vulcan (red leaf ) lettuce. There was considerable variation in sensory ratings among the 5 cultivars (P < 0.005) but few differences within cultivars across the growing season. The crisphead cultivar, Crispino, received higher scores (P < 0.01) for flavor, texture, and overall acceptability and was rated less bitter (P < 0.05) than other cultivars. Total phenolic content varied significantly (P < 0.001) among cultivars with the red leaf cultivar, Vulcan, exhibiting the highest levels. There was no correlation between bitterness and total phenolic content or environmental factors. Differences among lettuce cultivars appear to have a larger impact on sensory and phenolic profiles than environmental variation during the growing season. Keywords: lactuca sativa, lettuce, seasonality, sensory, total phenolics

Introduction

vailability of diverse and colorful types of salad greens has increased in recent years, along with information regarding the positive nutritional attributes of leafy green vegetables. The yearround and widespread intake of lettuce has helped to establish it as the most commonly consumed salad vegetable (USDA-ERS 2007). Darker lettuces, such as romaine, green leaf, and red leaf, are better sources of several nutrients than head lettuce, including vitamin A, niacin, riboflavin, thiamine, calcium, iron, potassium, manganese, selenium, and beta-carotene (USDA-ARS 2008). Lettuce is an important specialty crop in Colorado. Multiple lettuce crops can be harvested during one growing season and environmental conditions may vary considerably among early, mid-, and late season harvests. Regional environmental conditions, particularly higher altitude and light intensity, may further create stresses that shift the pattern of phytochemical synthesis (Liu and others 2007). Exposure to increased radiation and temperature has been shown to amplify the production of phenolic compounds in lettuce, which may boost dietary benefits (Caldwell 2003). Bitterness, an important flavor characteristic of lettuce, has been found to increase with higher growing season temperatures (Simonne and others 2002) and longer growing days (Rubatzky and Yamaguchi 1997). Favorable sensory attributes are essential for consumer acceptance (Pollard and others 2002) and excessive bitterness is usually associated with consumer rejection (Drewnowski

MS 20090763 Submitted 8/8/2009, Accepted 12/9/2009. Authors Bunning, Kendall, and Stone are with Dept. of Food Science & Human Nutrition, and authors Stonaker and Stushnoff are with Dept. of Horticulture and Landscape Architecture, Colorado State Univ., Fort Collins, CO U.S.A. 80523. Direct inquiries to author Bunning (E-mail: mbunning@cahs.colostate.edu).

and Gomez-Carneros 2000). Substantial variation in degree of bitterness has been reported among lettuce cultivars (Simonne and others 2002) and may be attributed to differing quantities of bitter components including the sesquiterpene lactones, lactucin, and lactucopricrin (Price and others 1990). A number of plant-based phenolic compounds are characteristically bitter and, therefore, aversive to consumers (Lesschaeve and Noble 2005); yet, many studies assessing phenolic content of crops do not evaluate taste. Qualitative and quantitative differences in phenolic content have been observed among lettuce cultivars, but little is known about the impact of phenolic compounds on bitterness and other taste attributes. Nicolle and others (2004) reported finding 10-fold more chlorogenic acid in a red leaf cultivar than in a green leaf cultivar as well as quantitative differences in dicaffeoyl tartaric acid among green leafed lettuce. Over 50-fold differences in quercetin content have been reported among lettuce cultivars (Crozier and others 1997). Liu and others (2007) reported that variations in phenolic content of lettuce were influenced by cultivar, type, and pigmentation, with red leaf cultivars exhibiting the highest levels. The evaluation of sensory and chemical properties in leafy vegetables presents a unique set of challenges, including color variability, rapid perishability, and lack of homogeneity (Crozier and others 1997). Using uniform samples of known cultivars and controlling growing, harvest, and postharvest conditions improve product consistency, which is essential for objective analytical evaluation. Few studies have examined the effects of genetics and seasonality on sensory or chemical properties of lettuce. The goal of this research was to assess the effects of seasonal variation in regional growing conditions on sensory properties of 5 selected cultivars of lettuce and to investigate possible correlations between bitterness,
C

S: Sensory & Food Quality

S156

JOURNAL OF FOOD SCIENCEVol. 75, Nr. 3, 2010

2010 Institute of Food Technologists R doi: 10.1111/j.1750-3841.2010.01533.x

Further reproduction without permission is prohibited

Effect of seasonal variation on lettuce . . .

total phenolic content, and 3 environmental indexes (growing degree days [GDD], radiation measured in total Langleys, and number of days with temperatures exceeding 30 C in the 30 d preceding harvest). The 5 selected cultivars represent the types most commonly available to consumers: butterhead, crisphead, green leaf, red leaf, and romaine. ples were evaluated by a consumer panel of 30, recruited from university students, faculty, and staff members. Prospective panelists completed questionnaires concerning time availability, lettuce preferences, and consumption habits. Training was limited to evaluation of bitterness. Panelists were trained to recognize bitterness by tasting 5 solutions of wormwood (Artemisia absinthium L.) extract (Herbal Remedies, Casper, Wyo., U.S.A.) with increasing levels of bitterness, from 0 to 100 L/mL. Lettuce and wormwood are classified in the same taxonomic subfamily, Lactuceae, within the family, Asteraceae, and contain similar bitter substances (Wagner 1977). In preparation for sensory evaluation, outer, damaged, or discolored leaves were removed and the lettuce was thoroughly rinsed with tap water and dried with paper towels. Four bite-size pieces (2 cm 2 cm) from inner and outer leaves were placed in snack-size zipper-sealed bags labeled with 3-digit code numbers which corresponded to labeled score sheets. The order of presentation of the 5 cultivars was completely randomized. Water and unsalted crackers were provided for cleansing the palate between samples. Samples were evaluated under fluorescent lighting by individual panel members midmorning on the day following harvesting. Two entire leaves of each cultivar were used for appearance evaluation. Panelists indicated their perception of flavor, appearance, texture, and overall acceptability on a 168-mm unstructured linear scale positioned directly above 7 categories ranging from very unacceptable to very acceptable. Bitterness was evaluated using a similar 7-category line scale, labeled from extremely mild to extremely bitter. For computation and analysis, the distance in millimeters was measured and values were converted to a 100-point scale. It is generally accepted that a trained panel should not be used for assessment of product acceptability because panelists would no longer behave as na ve consumers (Meilgaard and others 1999; Carpenter and others 2000). The sensory panelists in this study received training on identification of various concentrations of a bitter compound in order to familiarize them with intensity of bitterness (Poli and others 2002) but were not trained in any other sensory elements. Bitterness intensity was analyzed separately from other attributes. Since acceptability of lettuce from harvests throughout the summer months was being evaluated, the same panel was used for all 3 evaluations. Bitter intensity has frequently been measured using scalar techniques (Thorngate 1997). The use of a linear scale allowed panelists freedom to use intermediate points along the scale for perceived intensity of attributes, as opposed to being limited to discrete options (Stone and Sidel 2004). Linear rating scales have been used for sensory evaluation in lettuce (Delaquis and others 2000) and by consumers for evaluation of bitterness in beer (Einstein 1976).

Materials and Methods

Plant material and growing conditions
Five lettuce cultivars with promising production characteristics were selected for evaluation from a group of 25 previously screened for bolting resistance and total phenolic content (Liu 2004). Crisp and Green (green leaf), Crispino (crisphead), Vulcan (red leaf), and Green Forest (romaine) seeds were purchased from Johnnys Selected Seeds (Albion, Maine, U.S.A.) and Lochness (butterhead) seeds were purchased from Vilmorin (La M enitr e, France). Crispino is classified as an iceberg lettuce but forms a looser, less-dense head than the typical head lettuce sold in retail markets. The 5 lettuce cultivars were grown near Fort Collins, Colo., U.S.A., using organic production methods and harvested 70 d after seeding at the end of June, July, and August. Production methods conformed to requirements set by U.S. Department of Agriculture (USDA) for organic certification (USDA-AMS 2006). Field-grown plants were spaced at 8-inch intervals except for romaine transplants which were spaced 10 inches apart. Moisture levels were maintained using drip irrigation with municipal water. All lettuce was harvested early in the morning at market maturity and transported in coolers with ice packs to the laboratory for further processing within 1 h of harvest.

Environmental indexes
Germination, growth rate, and other metabolic processes of lettuce are particularly sensitive to environmental factors (Wien 1997). Maximum and minimum temperatures and total Langleys, a measure of solar radiation, were monitored to determine possible correlations with sensory quality and chemical assay data. Weather data, recorded hourly, were collected from an online site that tracks data at a weather station located within 100 m of the lettuce plot. GDD and number of days above 30 C (86 F) during the 30 d prior to harvest were used as measures of heat exposure. Plant development is dependent on daily accumulation of heat and GDD are used as a heat accumulation index to estimate temperature effects on growth and development during the growing season (Grierson 2002). The amount of heat required for plants to move to the next development stage remains constant for a particular species. For lettuce, the minimum base temperature below which development does not occur is 4.4 C. Daily GDD was calculated by subtracting the base temperature of lettuce (4.4 C) from the average of the daily maximum and minimum temperatures (McMaster and Wilhelm 1997). Daily GDD was totaled over the 30 d preceding harvest to give an accumulated heat index value. Solar radiation data, measured with an Eppley Precision Spectral Pyranometer at the weather station, provides information on the amount of energy striking a surface during a particular time period. Total Langleys are units of radiant flux equal to one calorie per square centimeter (cal/cm2 ). This measure of the total energy received in the ultraviolet, visible, and infrared wavelengths is a much wider band of wavelengths than plants can utilize but can be useful as an indicator of the amount of light available for plant growth (Hart 1988).

Total phenolic analysis

From each planting of the 5 cultivars, 35 g wedge-shaped samples, that included inner and outer leaves, were collected from 3 heads of lettuce and placed in a Genesis Freeze-Dryer (Virtis Inc., Gardiner, N.Y., U.S.A.) at 40 C, 10 C, +18 C, +28 C for 24 h each. Dried samples were ground into powder using an Osterizer 10 Speed Blender (Oster Inc., Boca Raton, Fla., U.S.A.) and sieved with a Nr 20 Tyler sieve (WS Tyler Inc., Mentor, Ohio, U.S.A.). The powdered samples were stored in 25 mL plastic centrifuge tubes sealed with a screw cap to prevent uptake of moisture and stored at 20 C. Four hundred mg of each sample was extracted in 10 mL Sensory analysis of 80% acetone, vortexed, and rotated in the dark for 2 h. After Study protocol was reviewed and approved by the Colorado State centrifugation, aliquots were transferred and vacuum-centrifuged. Univ. Human Research Committee. Freshly prepared lettuce sam- Dried samples were stored at 80 C.
Vol. 75, Nr. 3, 2010JOURNAL OF FOOD SCIENCE S157

S: Sensory & Food Quality

Effect of seasonal variation on lettuce . . .

The FolinCiocalteu assay, a widely accepted spectrophotometric method based on a color reaction (Singleton and Rossi 1965; Spanos and Wrolstad 1990), was used to quantify total phenolic content of dried lettuce tissue extracts. Thirty-five L of diluted sample was pipetted in microplate wells in triplicate. Using a multichannel pipette, 150 L of 0.2 M FolinCiocalteu reagent (Sigma Aldrich, St. Louis, Mo., U.S.A.) and 115 L of 7.5% Na2 CO3 (Fisher Scientific, Pittsburgh, Pa., U.S.A.) were added to all wells. The plate was incubated at 45 C for 30 min, cooled to room temperature for 1 h, and read at 765 nm using a spectrophotometer (Spectra Max Plus, Molecular Devices Corp., Sunnyvale, Calif., U.S.A.). Phenolic content was estimated by derivation from a standard curve based upon gallic acid and expressed as mg/g dry tissue weight. The presence of ascorbic acid in vegetable extracts can interfere with the FolinCiocalteu assay (Prior and others 2005). Following the methods of Esparza-Rivera and others (2006) samples were analyzed using high-performance liquid chromatography (HPLC) to verify low levels of Vitamin C after vacuum centrifugation. the kinds most frequently purchased. The number of consumers panelists necessary for sensory acceptability studies varies in the literature (Stone and Sidel 2004; Hough and others 2006) but generally, more than 30 panelists are used to evaluate a food product. In this study however, panelists were frequent consumers of lettuce and the same panelists were used for all evaluations across the growing season. All panelists had received training with reference samples in order to identify bitterness intensity.

Effect of seasonal variation on sensory attributes

For each of the 5 lettuce types evaluated, sensory scores for appearance and texture did not vary (P > 0.05) among early, midseason, and late harvest dates (Table 2). Flavor and overall acceptability scores also did not vary by harvest date with 2 exceptions. Green Forest (romaine) lettuce harvested in July (77.1) scored higher (P < 0.05) in flavor acceptability than Green Forest lettuce harvested in June (64.9) and Crisp and Green (green leaf) lettuce harvested in August (79.5) scored higher (P < 0.05) in overall acceptability than Crisp and Green lettuce harvested in July (68.9). In this evaluation, higher ratings for bitterness equated to extremely bitter. Three of the 5 cultivars displayed differences in bitterness score by harvest date (Table 3). The bitterness scores of Crispino (crisphead) lettuce decreased (P < 0.05) during the course of the growing season while Lochness (butterhead) increased (P < 0.05) and Crisp and Green (green leaf) lettuce exhibited higher scores (P < 0.05) in July than August. In contrast, Vulcan (red leaf) and Green Forest (romaine) did not vary (P > 0.05) in bitterness ratings across the growing season and most ratings for these 2 cultivars were in the neutral category, neither bitter nor mild. Although bitterness is thought to increase with higher growing season temperatures (Simonne and others 2002), only Crisp and Green (green leaf) was classified somewhat bitter by the panel and rated higher (P < 0.05) in bitterness in the July growing period which had the highest average temperatures. The study of bitter taste can be problematic: the threshold for detecting bitterness is much lower than other tastes, bitterness acuity is highly variable, and bitterness is associated with a great number of structurally diverse compounds (Thorngate 1997).

Statistical analyses
Mixed model analysis of variance was performed with panelist and panelist harvest treated as random variables and harvest, cultivar, and harvest cultivar as fixed effects. Statistical analysis of the data was performed with Fishers protected least significant difference test and Pearson correlation analysis (P < 0.05) using SAS Proc Mixed (Statistical Analysis Systems Inc., Cary, N.C., U.S.A., version 9.1).

Results and Discussion

Environmental variation

Lettuce harvests took place late in the months of June, July, and August, with data collected on environmental parameters during the 30 d prior to harvest. Variations in temperature and solar radiation during the 3 growing periods are shown in Table 1. The daily median temperature always exceeded 4.4 C, the base temperature of lettuce. There was considerable variation in temperature and solar radiation among the 3 growing periods. The early growing period (June) had the lowest average minimum and maximum temperatures and the widest temperature variation while the mid- Effect of cultivar on sensory attributes season (July) had the highest average minimum and maximum Differences were observed among cultivars for all sensory attemperatures, highest GDD sum, and most days above 30 C. The tributes (P < 0.005) indicating genetic composition plays a major mean daily and 30-day sum total of solar radiation measured in role in sensory of these lettuce cultivars. Langleys was highest in the late growing period (August). Appearance. Mean appearance scores of the 5 cultivars, averaged over the 3 harvest times (P < 0.0001), ranged from 78.6 Sensory panelists for Lochness (butterhead) to 88.3 for Vulcan (red leaf) (Table 4). The sensory panel consisted of Colorado State Univ. faculty (8), In this study, the red and dark green leaf cultivars (Vulcan and staff (12), and students (10); 53% were above the age of 40 and 70% Crisp and Green) scored in the very acceptable range and signifwere female. According to the preevaluation questionnaire, 60% of icantly higher (P < 0.05) in appearance than the romaine, butterpanelists consumed lettuce 3 or more times per week; only 13.3% head, and crisphead cultivars. Appearance is particularly important ate lettuce less than once a week. Romaine was the lettuce type for lettuce since it is the primary attribute consumers use in makmost preferred by the panel while romaine and mixed greens were ing the decision to purchase (Allende and others 2004). Visual
Table 1 --- Environmental data for lettuce plots recorded over the 30 d preceding each harvest. Temperature ( C) Harvest Nr 1 2 3
a Heat b

S: Sensory & Food Quality

Maximum Mean 23.2 28.2 26.5 Range 11.4 to 34.7 14.8 to 36.3 17.2 to 34.4 Mean 8.2 11.6 9.3

Minimum Range 1.7 to 14.2 7.5 to 17 4.6 to 14.5

GDDa Sum 339.5 464.4 404.7 Nr Daysb > 30 C 2 13 6

Solar Radiation Total Langleys Mean 343.8 482.7 486 Sum 10313 14482 14581

Date 6/29/2004 7/27/2004 8/29/2004

accumulation indicator: Growing Degree Day (GDD) = ([Tmax + Tmin] 2) 4.4 C. Heat stress indicator: number of days with maximum temperatures exceeding 30 C.

S158

JOURNAL OF FOOD SCIENCEVol. 75, Nr. 3, 2010

Effect of seasonal variation on lettuce . . .

imperfections, both type and quantity, can influence acceptance by consumers. Flavor. Across cultivars, the overall mean flavor rating for Crispino (crisphead) was 79.6, which was higher (P < 0.01) than the mean flavor ratings for the other 4 cultivars (Table 4). Still, flavor scores for all cultivars were in the acceptable range; except for the romaine cultivar, Green Forest, which was classified as somewhat acceptable. Perception of flavor involves a number of factors and has a major role in determining the acceptability of foods. In foods that are consumed raw, such as lettuce, flavor is due to substances that are either naturally present or formed immediately by chemical reactions (Carpenter and others 2000). The key components of lettuce flavor are believed to be sweetness and bitterness (Delaquis and others 2000). Texture. Mean texture acceptability scores varied (P < 0.001) across cultivars from 72.7 for Green Forest (romaine) to 81.4 for Crispino (crisphead), but all were in the acceptable range (Table 4). Texture of fresh vegetables is primarily determined by cell wall structure and turgor pressure (Waldron and others 2003) and is particularly important in raw produce because consumers associate it with freshness and wholesomeness (Allende and others

Table 2 --- Sensory quality and acceptability of 5 lettuce cultivars across growing season. Harvest Month Attribute Appearance Cultivar Crispino Crisp and Green Vulcan Green Forest Lochness Crispino Crisp and Green Vulcan Green Forest Lochness Texture Crispino Crisp and Green Vulcan Green Forest Lochness O. Acceptability Crispino Crisp and Green Vulcan Green Forest Lochness Crisphead Green Leaf Red Leaf Romaine Butterhead Crisphead Green Leaf Red Leaf Romaine Butterhead Type Crisphead Green Leaf Red Leaf Romaine Butterhead Crisphead Green Leaf Red Leaf Romaine Butterhead June 81.5 2.5 a 85.2 2.5 a 88.1 2.6 a 79.6 2.5 a 76.3 2.5 a
a

July 82.6 2.6a 90.0 2.6a 86.4 2.6a 82.4 2.6a 80.2 2.6a 78.0 3.4a 67.4 3.4a 75.8 3.3a 77.1 3.4a 70.4 3.3a 78.1 2.9a 75.6 2.9a 75.7 2.8a 76.6 2.9a 76.2 2.9a 82.6 3.4a 68.9 3.4b 75.7 3.3a 77.4 3.5a 68.9 3.4a

August 78.3 2.7a 88.0 2.7a 90.5 2.7a 77.7 2.7a 79.4 2.7a 81.5 3.8a 76.8 3.8a 68.7 3.8a 70.5 3.8ab 72.6 3.8a 83.9 3.1a 78.9 3.1a 73.4 3.1a 69.5 3.1a 75.2 3.1a 80.3 3.8a 79.5 3.8a 72.3 3.8a 68.9 3.8a 69.9 3.8a

Flavor

79.2 3.4a 74.5 3.3a 70.2 3.4a 64.9 3.4b 72.3 3.3a 82.3 2.9a 79.1 2.8a 72.4 2.8a 72.0 2.9a 73.9 2.8a 75.9 3.4a 71.5 3.3ab 72.0 3.3a 68.6 3.4a 69.9 3.3a

a Data expressed as means SEM. Values were measured on a 168-mm line and converted to a 100-point scale, 0 to 14.6 = Very Unacceptable, 14.7 to 29.0 = Unacceptable, 29.1 to 43.4 = Somewhat Unacceptable, 43.5 to 57.8 = Neither Acceptable nor Unacceptable, 56.9 to 71.2 = Somewhat Acceptable, 71.3 to 85.6 = Acceptable, 85.7 to 100 = Very Acceptable. Means within rows followed by different letters designate signicant differences (P < 0.05).

Table 3 --- Bitterness ratings of 5 cultivars of lettuce grown early, midway, and late in the growing season.a Bitterness Rating Cultivar Crispino Crisp and Green Vulcan Green Forest Lochness Type Crisphead Green Leaf Red Leaf Romaine Butterhead June 46.0 4.0b 50.0 3.9ab 50.2 3.9a 48.3 4.0a 44.4 3.9a July 39.6 4.0ab 58.8 4.0b 47.8 3.9a 40.9 4.0a 49.6 4.0ab August 29.7 4.3a 44.8 4.3a 52.9 4.3a 47.5 4.3a 56.1 4.3b Mean Score 38.5 2.81 51.2 2.82 50.3 2.72 45.6 2.82 50.1 2.82

a Data expressed as means SEM. Scored on a 168-mm categorical line scale and converted to a 100-point scale, 0 to 14.6 = Extremely Mild, 14.7 to 29.0 = Very Mild, 29.1 to 43.4 = Somewhat Mild, 43.5 to 57.8 = Neither Bitter nor Mild, 56.9 to 71.2 = Somewhat Bitter, 71.3 to 85.6 = Very Bitter, 85.7 to 100 = Extremely Bitter. Means within rows followed by different letters designate signicant differences (P < 0.05). Means within mean score column followed by different numbers designate signicant differences (P < 0.05).

Table 4 --- Mean ratings for sensory quality and acceptability of 5 lettuce cultivars.a Cultivar Sensory Attribute Appearance Flavor Texture Overall Acceptability Crispino Crisphead 80.8 1.7b 79.6 2.2a 81.4 1.8a 79.6 1.7a Crisp and Green Green Leaf 87.7 1.8a 72.9 2.2b 77.8 1.8ab 73.3 1.7b Vulcan Red Leaf 88.3 1.7a 71.6 2.2b 73.8 1.8bc 73.4 1.7b Green Forest Romaine 79.9 1.7b 70.9 2.2b 72.7 1.8c 71.7 1.7b Lochness Butterhead 78.6 1.7b 71.7 2.2b 75.1 1.8bc 69.5 1.7b

P Value
<0.0001 0.0030 0.0006 0.0013

a Data expressed as means SEM. Values were measured on a 168-mm line and converted to a 100-point scale, 0 to 14.6 = Very Unacceptable; 14.7 to 29.0 = Unacceptable; 29.1 to 43.4 = Somewhat Unacceptable; 43.5 to 57.8 = Neither Acceptable nor Unacceptable; 56.9 to 71.2 = Somewhat Acceptable; 71.3 to 85.6 = Acceptable; 85.7 to 100 = Very Acceptable. Means within rows followed by different letters designate signicant differences (P < 0.05). P value = calculated probability value, Pr < F.

Vol. 75, Nr. 3, 2010JOURNAL OF FOOD SCIENCE

S159

S: Sensory & Food Quality

Effect of seasonal variation on lettuce . . .

2004). Texture is a critical feature because fresh lettuce is expected to have a crispy, crunchy texture despite having a composition that is approximately 95% water. Perception of a crunchy texture is a complex concept which includes sounds, fracture characteristics, density, and geometry (Fillion and Kilcast 2002). Overall acceptability. Crispino (crisphead) scored higher (P < 0.02) than the other cultivars in overall acceptability (Table 4). Mean overall acceptability scores ranged from 69.5 for Lochness (butterhead) to 79.6 for Crispino. The rating for Lochness placed it in the somewhat acceptable category while the ratings for the other 4 cultivars were in the acceptable range. Bitterness. Similar to the flavor and overall acceptability results, mean bitterness scores (all harvest dates) were significantly lower (P < 0.05) for Crispino than for the other 4 cultivars evaluated (Table 3). Crispino was rated somewhat mild while the other 4 cultivars were rated in the neutral category, neither bitter nor mild. Trends in fresh produce consumption are strongly influenced by consumer perception (Barrios and Costell 2004). Many factors, including product characteristics and quality perception, influence the consumers decision to purchase (Waldron and others 2003), so it is important that produce meets the standard expectation of the consumer. All samples in this study were scored at the higher end of the acceptability scale, suggesting that lettuce quality was not negatively impacted by environmental variations. July harvests. For ease of comparison with bitterness scores, total phenolic content was expressed as milligrams of gallic acid equivalents (GAE) per gram dry weight (gdw). Mean total phenolic content expressed as GAE per gram fresh weight (gfw) varied among cultivars as follows: Vulcan (211.48) > Crisp and Green (158.36) > Green Forest (95.07) > Lochness (93.11) > Crispino (66.77). Mean total phenolic content of the 5 cultivars for all harvest dates varied (P < 0.0001) with Vulcan (red leaf) exhibiting the highest level of total phenolic content (46.9 mg GAE/gdw) and Crispino (crisphead) the lowest level (13.9 mg GAE/gdw). Increases in phenolic content did not parallel bitterness scores (Figure 2) and for all cultivars, correlations between phenolic content and bitterness score were not significant (P > 0.05). Correlations were also measured between total phenolic content and the 3 environmental indexes (GDD, solar radiation (total Langleys), and number of days with temperatures exceeding 30 C (86 F), but no significant trends were observed (P > 0.05). Vitamin C levels of freeze-dried samples ranged from 6.9 mg/100 g fresh weight in Nevada to 13.6 mg/100 gfw in Green Forest but did not vary significantly by cultivar (P = 0.089).

Conclusions
ensory panel and analytical evaluations of these 5 lettuce cultivars demonstrated that genetic constitution had a greater impact on sensory qualities and total phenolic content than seasonal variation. Appearance, flavor, texture, overall acceptability, degree of bitterness, and total phenolic content varied greatly among lettuce cultivars of different types and pigmentation patterns but, with few exceptions, did not vary within cultivars across the growing season. Vulcan, a red leaf cultivar, exhibited the highest levels (P < 0.001) of total phenolic content. Red and green leaf lettuce cultivars received the highest scores for appearance, but Crispino, an iceberg type of crisphead lettuce, was rated highest in flavor, texture, and overall acceptability and lowest in bitterness and total phenolic content. This was somewhat surprising since iceberg lettuce received the lowest preference ranking in the prequestionnaire. This particular subtype, however, is leafier and less dense than commercially grown iceberg lettuce. Sensory characteristics of fresh lettuce are important variables to producers because of

Correlation coefficients
A positive linear relationship existed between overall acceptability and flavor, with a correlation coefficient of R2 = 0.73 and overall acceptability and texture with a coefficient of R2 = 0.64. In contrast, overall acceptability and appearance were poorly correlated (R2 = 0.31). These results agree with those of Esparza-Rivera (2005) and suggest that consumer panelists primarily based their ratings for overall acceptability on flavor and texture rather than on how the lettuce looked.

mg GAE/gdw

S: Sensory & Food Quality

Effect of growing season and cultivar on total phenolic content

As with sensory analysis, analytical evaluations revealed considerable variation in total phenolic content across the 5 cultivars, but variations within cultivars across the growing season were only seen for Vulcan and Crispino (Figure 1). Total phenolic content was higher (P < 0.005) in Vulcan (red leaf) and lower (P < 0.05) in Crispino (crisphead) harvested in August compared to June and

Total Phenolic Content and Bitterness

Bitterness Score 70 Total Phenolics

Total Phenolic Content

70 60 50 40 29-Jun 27-Jul 29-Aug

60 50 40 30 20

30

10
20

0
10 0 Crisp & Green Green Leaf Vulcan Red Leaf Green Forest Romaine Lochness Butterhead Crispino Crisphead

Crispino Crisphead

Green Forest Romaine

Lochness Butterhead

Crisp & Green Green Leaf

Vulcan Red Leaf

Lettuce Cultivar and Type

Figure 2 --- Total phenolic content and bitterness scores of 5 lettuce cultivars. Bitterness scores were evaluated on Figure 1 --- Total phenolic content of 5 cultivars of let- a 168-mm line and converted to a 100-point scale. Total tuce, measured in milligrams of gallic acid equivalents phenolic values were expressed as milligrams of gallic per gram dry weight, grown early, midway, and late in acid equivalents (GAE) per gram dry weight lettuce. Error the growing season. bars indicate SEM (standard error of the mean).
Lettuce Cultivar & Type

S160

JOURNAL OF FOOD SCIENCEVol. 75, Nr. 3, 2010

Effect of seasonal variation on lettuce . . .

their impact on consumer acceptance. Although there was variation in sensory ratings among the 5 cultivars, mean scores of all attributes remained within the acceptable range, indicating that various types of lettuce with acceptable sensory qualities can be grown in this region during summer months. There was not a significant correlation between bitterness and total phenolic content or the 3 environmental indexes monitored in this study. Evaluation of sensory properties and quantification of phenolic content in lettuce can be used to identify specific cultivars that exhibit superior attributes and may improve market competitiveness of various types and particular cultivars of lettuce.
Lesschaeve I, Noble AC. 2005. Polyphenols: factors influencing their sensory properties and their effects on food and beverage preferences. Am J Clin Nutr 81(1):330S5S. Liu XF. 2004. Total phenolic content and DPPH radical scavenging ability of lettuce (Lactuca sativa L.) grown in Colorado [MSc thesis]. Fort Collins, Colo.: Colorado State Univ. 55 p. Liu XF, Ardo S, Bunning M, Parry J, Zhou KQ, Stushnoff C, Stonaker FH, Yu LL, Kendall P. 2007. Total phenolic content and DPPH radical scavenging activity of lettuce (Lactuca sativa L.) grown in Colorado. LWT Food Sci Tech 40(3):5527. McMaster GS, Wilhelm WW. 1997. Growing degree-days: one equation, two interpretations. Agric For Meteorol 87(4):291300. Meilgaard M, Civille GV, Carr BT. 1999. Sensory evaluation techniques. 3rd ed. Boca Raton, Fla.: CRC Press. 387 p. Nicolle C, Carnat A, Fraisse D, Lamaison JL, Rock E, Michel H, Amouroux P, Remesy C. 2004. Characterisation and variation of antioxidant micronutrients in lettuce (Lactuca sativa folium). J Sci Food Agric 84(15):20619. Poli F, Sacchetti G, Tosi B, Fogagnolo M, Chillemi G, Lazzarin R, Bruni A. 2002. Variation in the content of the main guaianolides and sugars in Cichorium intybus var. Rosso di Chioggia selections during cultivation. Food Chem 76(2):13947. Pollard J, Kirk SFL, Cade JE. 2002. Factors affecting food choice in relation to fruit and vegetable intake: a review. Nutr Res Rev 15(2):37387. Price KR., Dupont MS, Shepherd R, Chan HWS, Fenwick GR. 1990. Relationship between the chemical and sensory properties of exotic salad crops - colored lettuce (Lactuca-Sativa) and chicory (Cichorium-Intybus). J Sci Food Agric 53(2):18592. Prior RL Wu XL., Schaich K. 2005. Standardized methods for the determination of antioxidant capacity and phenolics in foods and dietary supplements. J Agric Food Chem 53(10):4290302. Rubatzky VE, Yamaguchi M. 1997. World vegetables: principles, production, and nutritive values. 2nd ed. New York, N.Y.: Chapman & Hall. 843 p. Simonne A, Simonne E, Eitenmiller R, Coker CH. 2002. Bitterness and composition of lettuce varieties grown in the southeastern United States. HortTechnology 12(4):7216. Singleton VL, Rossi JA. 1965. Colorimetry of total phenolics with phosphmolybdicphosphotungstic acid reagents. Am J Enol Vitic 16:14458. Spanos GA, Wrolstad RE. 1990. Influence of processing and storage on the phenolic composition of thompson seedless grape juice. J Agric Food Chem 38(7):156571. Stone H, Sidel JL. 2004. Sensory evaluation practices. 3rd ed. San Diego, Calif.: Elsevier Adademic Press. 377 p. Thorngate JH. 1997. Factors Affecting the perception of bitterness: a review. In: Roy G, editor. Modifying bitterness. Lancaster, Pa.: Technomic Pub. Co. 332 p. USDA-AMS. 2006. United States Department of Agriculture-Agricultural Marketing Service. National Organic Program Standards. Available from: http:// www.ams.usda.gov/nop/NOP/standards.html. Accessed Nov 20, 2008. USDA-ERS. 2007. United States Department of Agriculture-Economics Research Service. U.S. Lettuce Statistics 2007. Available from: http://usda.mannlib. cornell.edu/MannUsda/viewDocumentInfo.do?documentID=1576. Accessed Nov 20, 2008. USDA-ARS. 2008. United States Department of Agriculture-Agricultural Marketing Service National Nutrient Database for Standard Reference, Release 21: Nutrient Data Laboratory. Available from: http://www.nal. usda.gov/fnic/foodcomp/search/. Accessed Nov 20, 2008. Wagner H. 1977. Pharmaceutical and economic uses of the compositae. In: Heywood VH, Harborne JB, Turner BL, editors. The biology and chemistry of the compositae. San Francisco, Calif.: Academic Press. p 41136. Wien HC. 1997. Lettuce. In: Wien HC, editor. The physiology of vegetable crops. New York: CAB Intl. p 479509. Waldron K, Parker M, Smith A. 2003. Plant cell walls and food quality. Comp Rev Food Sci Food Safety 2:10119.

Acknowledgments
Funding was provided by the Colorado State Univ. Agricultural Experiment Station.

References
Allende A, Aguayo E, Artes F. 2004. Microbial and sensory quality of commercial fresh processed red lettuce throughout the production chain and shelf life. Int J Food Microbiol 91(2):10917. Barrios E, Costell E. 2004. Review: use of methods of research into consumers opinions and attitudes in food research. Food Sci Tech Int 10(6):35971. Caldwell CR. 2003. Alkylperoxyl radical scavenging activity of red leaf lettuce (Lactuca sativa L.) phenolics. J Agric Food Chem 51(16):458995. Carpenter RP, Lyon DH, Hasdell TA. 2000. Guidelines for sensory analysis in food product development and quality control. 2nd ed. Gaithersburg, Md.: Aspen Publishers. 210 p. Crozier A, Lean MEJ, McDonald MS, Black C. 1997. Quantitative analysis of the flavonoid content of commercial tomatoes, onions, lettuce, and celery. J Agric Food Chem 45(3):5905. Delaquis PJ, Stewart S, Cliff M, Toivonen PM, Moyls AL. 2000. Sensory quality of readyto-eat lettuce washed in warm, chlorinated water. J Food Qual 23(6):55363. Drewnowski A, Gomez-Carneros C. 2000. Bitter taste, phytonutrients, and the consumer: a review. Am J Clin Nutr 72(6):142435. Einstein MA. 1976. Use of linear rating scales for evaluation of beer flavor by consumers. J Food Sci 41(2):3835. Esparza-Rivera J. 2005. Effects of ascorbic acid applied by two hydrocooling methods on physical, chemical and sensory attributes of green leaf lettuce stored at 5 C [PhD diss.]. Fort Collins, Colo.: Colorado State University. 140 p. Esparza-Rivera JR, Stone MB, Stushnoff C, Pilon-Smits E, Kendall PA. 2006. Effects of ascorbic acid applied by two hydrocooling methods on physical and chemical properties of green leaf lettuce stored at 5 degrees C. J Food Sci 71(3):S2705. Fillion L, Kilcast D. 2002. Consumer perception of crispness and crunchiness in fruits and vegetables. Food Qual Prefer 13(1):239. Grierson W. 2002. Role of temperature in the physiology of crop plants: pre- and postharvest. In: Pessarakli M, editor. Handbook of plant and crop physiology. New York: Marcel Dekker, Inc. p 1333. Hart JW. 1988. Light and plant growth. Topics in plant physiology; 1. Boston, Mass.: Unwin Hyman. 204 p. Hough G, Wakeling I, Mucci A, Chambers E, Gallardo I, Alves L. 2006. Number of consumers necessary for sensory acceptability tests. Food Qual Prefer 17(6):5226.

Vol. 75, Nr. 3, 2010JOURNAL OF FOOD SCIENCE

S161

S: Sensory & Food Quality