Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Genetic Diversity, Population Structure, and Heritability of Fruit Traits in Capsicum annuum

  • Rachel P. Naegele,

    Affiliation USDA, Agricultural Research Service, San Joaquin Valley Agricultural Sciences Center, 9611 South Riverbend Avenue, Parlier, CA, United States of America

  • Jenna Mitchell,

    Affiliation Department of Plant, Soil and Microbial Sciences, Michigan State University East Lansing, MI 48824, United States of America

  • Mary K. Hausbeck

    hausbec1@msu.edu

    Affiliation Department of Plant, Soil and Microbial Sciences, Michigan State University East Lansing, MI 48824, United States of America

Genetic Diversity, Population Structure, and Heritability of Fruit Traits in Capsicum annuum

  • Rachel P. Naegele, 
  • Jenna Mitchell, 
  • Mary K. Hausbeck
PLOS
x

Abstract

Cultivated pepper (Capsicum annuum) is a phenotypically diverse species grown throughout the world. Wild and landrace peppers are typically small-fruited and pungent, but contain many important traits such as insect and disease resistance. Cultivated peppers vary dramatically in size, shape, pungency, and color, and often lack resistance traits. Fruit characteristics (e.g. shape and pericarp thickness) are major determinants for cultivar selection, and their association with disease susceptibility can reduce breeding efficacy. This study evaluated a diverse collection of peppers for mature fruit phenotypic traits, correlation among fruit traits and Phytophthora fruit rot resistance, genetic diversity, population structure, and trait broad sense heritability. Significant differences within all fruit phenotype categories were detected among pepper lines. Fruit from Europe had the thickest pericarp, and fruit from Ecuador had the thinnest. For fruit shape index, fruit from Africa had the highest index, while fruit from Europe had the lowest. Five genetic clusters were detected in the pepper population and were significantly associated with fruit thickness, end shape, and fruit shape index. The genetic differentiation between clusters ranged from little to very great differentiation when grouped by the predefined categories. Broad sense heritability for fruit traits ranged from 0.56 (shoulder height) to 0.98 (pericarp thickness). When correlations among fruit phenotypes and fruit disease were evaluated, fruit shape index was negatively correlated with pericarp thickness, and positively correlated with fruit perimeter. Pepper fruit pericarp, perimeter, and width had a slight positive correlation with Phytophthora fruit rot, whereas fruit shape index had a slight negative correlation.

Introduction

Peppers (Capsicum annuum) are an important spice and vegetable crop grown in the U.S. and worldwide. In 2014, the U.S. imported $1.6 billion and produced over $834 million of bell and chile peppers [ERS, 2014]. According to the FAO, in 2013 nearly 200 million and 33 million tonnes of green and dry peppers, respectively, were produced worldwide. These numbers include chile, bell, and specialty-type peppers. In the U.S., bell peppers account for $618 million of the pepper market, with chile peppers making up an additional $216 million [NASS, 2014]. Specialty peppers, including cheese-type peppers and those with diverse shape, color or flavor, are a relatively small component of the market. While the U.S. grows predominantly thick walled bell-type peppers, pepper fruit shape can vary greatly [1]. Fruit shape and pericarp or fruit thickness are two of the most important characteristics in deciding a pepper cultivar's regional success. Bell and cheese (sweet pimento style) type peppers are often mild or non pungent with thick flesh. Bell peppers have a blocky, lobed appearance, while cheese peppers are lobed and squat or flat. Pungent peppers, including jalapeno, habanero, serrano, poblano, shishito, and thai peppers, can vary greatly in size, shape, pungency level, color, and flesh thickness [2].

Fruit shape has been extensively studied in the Solanaceae including tomato, pepper and eggplant [1,311]. In tomato, quantitative trait loci (QTL) contributing to fruit shape have been identified on chromosomes 2, 3, 7, 8, and 10 [1,3,12,13]. In tomato, fruit shape is primarily determined by allelic variation in the Sun, Ovate, Fasciated (FAS), and Locule Number (LC) genes [14]. Rodriguez et al., demonstrated that up to 71% of the specific shape variation could be explained by individual alleles of these genes in a diverse collection of 368 wild and cultivated tomatoes [14]. When QTLs from tomato and pepper were compared, fruit weight was highly co-localized between species, and a single fruit shape QTL was co-localized suggesting conserved elements are contributing to one, if not both, of the traits [1,4].

In pepper, previous studies have evaluated the heritability and effect of QTL associated with fruit horticultural characteristics [2,4,911,15,16]. Multiple QTLs have been detected on chromosomes 1–4, 8, 10 and 11 for fruit length, width, and the fruit shape ratio (length:width) [4,9,15,17]. Two major fruit QTLs, designated fs3.1 (fruit shape) and fs10.1 (fruit elongation), were mapped in a BC4F2 population segregating for fruit shape to chromosomes 3 and 10, respectively [9]. These QTLs explained 67 and 44% of the variation for fruit shape and elongation, respectively, observed in the population. Most recently, Vilarinho et al, evaluated the inheritance of fruit traits in relation to pericarp shape, color thickness and total soluble solids [18]. Based on segregation ratios, they determined that the round shape trait was controlled by a single gene. In a serrano by jalapeno recombinant inbred line F8 population, Naegele et al. identified five QTLs contributing to fruit shape and one QTL for pericarp thickness on chromosomes 1,2,4,10 and 3, respectively, explaining 4 to 26% of the variation [15]. Tsaballa et al., evaluated the expression of a gene with sequence similarity to the tomato gene Ovate and found significant differences between a round and elongated pepper cultivar [16].

In 2012, another QTL analysis determined that fruit mass, length, diameter, shape ratio, and flesh thickness were controlled by two dominant genes with heritability ranging from 38–88% [10]. When evaluating a pepper germplasm collection from the Caribbean, fruit width was highly heritable, and fruit weight and width were positively correlated, consistent with the QTL analysis by Chaim et al [2,9]. In another mapping study, it was estimated that the heritability of fruit shape and flesh thickness were both 80% [11]. The INRA described the phenotype of over 1,300 pepper accessions in their collection for 12 fruit traits; shape and color were diverse among the domesticated species, while wild species typically had small, elongated fruit [19]. Despite the number of studies evaluating fruit shape in pepper, a limitation to all was the use of subjective visual (elongate, triangular, square, heart, etc.) or manual (length/width ratio) measurements to classify fruit shape. Objective and accurate measurements of fruit will aid in our understanding of the factors of controlling fruit traits. In tomato, improved phenotyping software has been developed, allowing for more objective accurate measurements of fruit characteristics [6,20]. This software has already been successfully implemented in related species [5,6].

While fruit shape is one of the most important considerations for a cultivar, disease resistance is also necessary. Due to breeding bottlenecks, cultivated varieties often do not have resistance to many diseases. Frequently, resistance is identified in small-fruited wild species and incorporated into larger-fruited commercial cultivars [21,22]. Negative horticultural traits may also be transferred along with the positive traits such as disease resistance through linkage drag or as pleiotropic effects. Recently, in tomato, a study demonstrated that undesirable effects on maturity, fruit size, yield and plant architecture were linked to resistance to the late blight pathogen (Phytophthora infestans) [23]. In pepper, an overlap between fruit characteristics and disease resistance was identified for a single isolate of P. capsici, a devastating pathogen that incites fruit, foliar, and root rot [15]. In eggplant, fruit shape was positively correlated with disease susceptibility to P. capsici in a germplasm population [24]. In kiwi, negative correlations between resistance to the bacterial pathogen Pseudomonas syringae pv. actinidiae and number of fruit per vine suggested that resistance could result in reduced yield [25]. When transferring disease resistance into commercial cultivars, it is important to identify potential correlations, linkage drag, and pleiotropic effects.

Understanding the heritability, correlation, and diversity of fruit traits is essential for the efficient utilization of pepper germplasm. The objectives of this study were to i) determine fruit horticultural characteristics using the Tomato Analyzer (TA) software, ii) determine population structure associated with fruit traits of interest, iii) associate fruit shape categories with TA values, iv) determine the broad sense heritability for each fruit trait, and v) identify correlations among fruit traits and disease resistance to Phytophthora capsici.

Materials and Methods

One hundred sixteen peppers (Capsicum annuum), 114 of which had been previously evaluated for Phytophthora fruit rot resistance, were used in this study (Table 1) [26,27]. Twenty seeds from each line were planted into a 72-cell tray (Hummert Intl.) filled with a soilless-based mix (Suremix, Growers Products Inc. Galesburg, MI) in a polyethylene greenhouse at Michigan State University's Horticulture Research and Teaching Farm (Holt, MI). Seedlings were transferred to 1 L black plastic pots (Hummert Intl.) filled with the same soilless-based mix and grown to maturity. Mature fruit were harvested from each plant, bulked by line, and returned to the lab for evaluation.

Clean mature pepper fruit were sliced longitudinally, placed face down on an Epson Perfection V30 scanner (Epson America, Long Beach, CA), and scanned. Using the Tomato Analyzer (TA) software v3.0, fruit perimeter, area, width at mid height, max width, height at mid width, max height, shoulder height and fruit shape index external 1 were determined as described [6,13,20]. Fruit shape categories Circular (smaller values indicate more circular), Rectangular (ratio of the area of the shape containing the fruit to the area of the rectangle contained by the fruit), Ellipsoid (smaller values indicate fruit is more ellipsoid), Ovoid and Obovoid were calculated by TA. When the software was unable to accurately identify the outline of a fruit shape, or proximal or distal ends, points were adjusted manually. Fruit end shape (pointed or blunt) was assessed visually for each line. Fruit shape (Long, Ellipsoid, Rectangular, Oxheart, Heart, Round, Flat) was assessed visually and categorized using the designations described by Rodriguez et al. [14]. Additionally, fruit shape categories (Elongate, Oblate, Round, Conic, Campanulate, Bell, Mixed) for 79 accessions that had previously been characterized by Bosland, were also included [27]. Fruit pericarp thickness was measured using a hand caliper on each side of a longitudinal slice and averaged for each fruit.

Data were analyzed in the software SAS v9.3 (SAS Cary, NC) using the PROC MIXED function. Significant differences were detected using ANOVA and separated using LSD (P = 0.05). For fruit shape index, perimeter, and area, data were natural log transformed to fulfill assumptions of normality. Correlations were detected using Pearson's Correlation coefficient (r) at P = 0.05 among fruit traits and disease. Only lines for which complete TA data and disease data were available were used for correlation analyses. Disease data from a previous study were used for lesion area at three and five days post inoculation (dpi) [26]. Only the first two reps (for a total of 10 peppers) were used for fruit characteristics and disease correlations. Broad sense heritability for each trait was estimated using the mean squares implemented within the formula described by Fehr [28]. Confidence intervals were calculated according to Knapp et al. [29].

Previously, twenty-three simple sequence repeat (SSR) markers were evaluated for the population [26]. For the subset of pepper lines evaluated in this study, genetic structure of the population was evaluated in the software STRUCTURE v3.4 [30] with a burn in of 300,000 and a MCMC of 500,000 with correlated allele frequencies [31]. To test the putative number of populations, K values of 1–15 were evaluated with three independent runs. Lambda was estimated at 0.55 and the value of K was reported to be five according to the methods by Evanno et al [32] implemented in STRUCTURE Harvester [33]. The significance of Wright's FST, a measure of the genetic differentiation among sub populations, was determined using PowerMarker v3.25 [34] with 1,000 permutations. Differentiation was defined according to Hartl and Clark [35]. Population structure was sorted by predefined categories (pericarp thickness, fruit shape, and end shape) using the Population Sorting Tool [24]. Lines were considered to belong to a cluster if they had a membership (Q) ≥ 60% in that cluster. For categorical analyses in STRUCTURE, pepper lines were grouped based on a pericarp thickness of <0.05, 0.05 to 0.10, 0.11 to 0.15, 0.16 to 0.20, 0.21 to 0.30, or ≥0.30 cm. Only categories represented by three or more individuals with unmixed fruit were included in population structure and geographic-level ANOVA analyses.

Results and Discussion

Since their initial domestication in Mexico, peppers have been under strong selection for fruit shapes and size [36]. While wild relatives and landrace peppers are frequently small and highly pungent, domesticated pepper fruit have an endless array of phenotypic diversity [2,19]. For most countries and markets, there are distinct regional preferences for the type of pepper consumed. These regional preferences have contributed to strong phenotypic diversity among market classes [37]. In this study, fruit traits varied in the population, and significant differences were detected among lines for each of the phenotypic traits evaluated (Fig 1, Table 1, S1 Table). The mean pericarp thickness of the population was 0.14 ± 0.002 cm. The lines with the thinnest pericarp were PIs 267730, 593495, and 102883 (0.01 cm). The line with the thickest pericarp was PI 432802 (0.51 cm). When grouped by continent, the pericarp thickness of fruit from Europe was the highest (0.22), while fruit from South America had the thinnest pericarp (0.09). When grouped by country, fruit from Serbia were the thickest (0.25) while fruit from Ecuador were the thinnest (0.06 cm) (Table 2). Many of the accessions from South America were wild or landrace individuals, and had thin fruit compared to the cultivated fruit from Europe, which were more than twice as thick on average. The variation in pericarp thickness was also detected among countries.

thumbnail
Fig 1. Mature pepper fruit phenotypic diversity in size, shape, end shape, and pericarp thickness of a worldwide collection.

https://doi.org/10.1371/journal.pone.0156969.g001

thumbnail
Table 2. Fruit thickness, perimeter, area and fruit shape compared among countries and continents.

https://doi.org/10.1371/journal.pone.0156969.t002

The line with the smallest perimeter was PI 267730 (2.74 cm) and the line with the largest perimeter was PI 640791 (25.53 cm). The mean perimeter for the population was 11.39 ± 0.20 cm. Fruit from Italy had the largest perimeter (14.56 cm) and fruit from Brazil had the smallest (7.84 cm). Fruit with the largest perimeter (12.55 cm) came from Europe; fruit from South America had the smaller perimeter (8.52 cm). The population mean for fruit area was 6.71 ± 7.68 cm2; the smallest was 0.50 cm2 (PI 593493) and the largest was 23.26 cm2 (PI 645520). Turkey (9.31 cm2) had fruit with the largest area, whereas fruit from Taiwan were the smallest (2.35 cm2). Fruit from Europe had the greatest area (6.91 cm2) and fruit from South America had the smallest (3.85 cm2). PI 645520 (0.64) and PI 511879 (4.37) had the lowest and highest fruit shape index, respectively. The population mean for fruit shape index was 2.19 ± 1.12. Fruit from Africa (2.40) and Taiwan (3.70) had the largest fruit shape index, while fruit from Europe (1.54) and China (1.55) had the lowest. The smallest maximum width and height for the population was 0.63 cm (PI 593495) and 0.88 cm (PI 267730), respectively. The largest maximum width and height for the population was 7.27 cm (PI 645520) and 9.96 cm (PI 640791), respectively. The population means for maximum width and height was 2.13 ± 1.27 cm and 4.11 ± 2.41 cm, respectively. Broad sense heritability was high (>0.90) for most fruit traits evaluated (Table 3). Pericarp had the highest heritability in the population (0.98). Fruit shape index 1 and width at mid height also had high heritability (0.96) in the population. The lowest heritability was observed for shoulder height (0.56). Previous studies have shown that heritability of fruit shape (length to width ratio) and pericarp thickness are high in peppers [2,911]. Consistent with previous research, this pepper population had high heritability (>0.90) for most of the traits evaluated. The traits with lowest heritability in the population were shoulder height (0.56) and fruit shape triangle (0.84) suggesting these attributes are more subject to environmental variation.

The software STRUCTURE detected 5 genetic clusters (Ln = -3,526.3). The genetic differentiation between clusters was moderate to very great (FST = 0.06–0.16). Clusters did not perfectly differentiate fruit shape or pericarp thickness categories. However, certain clusters were more frequently associated with a particular category (Fig 2) than others. When grouped by pericarp thickness, genetic diversity and polymorphism information content (PIC) were moderate among groups (Table 4). The highest PIC and genetic diversity were in fruit from the 0.05–0.10 (PIC = 0.40, GD = 0.44) and 0.16–0.20 (PIC = 0.40, GD = 0.45) categories. When grouped by pericarp thickness, cluster 4 (dark blue) was less frequently found in peppers with a pericarp <0.05, 0.11 to 0.15, and 0.21 to 0.30. Cluster 2 (yellow) was less frequently associated with peppers with a pericarp <0.05 or ≥0.30. Little differentiation (FST = 0 to 0.05) was detected between peppers with a pericarp thickness of <0.05 or 0.05 to 0.10 and peppers with a pericarp thickness of 0.16 to 0.20, or peppers with a pericarp thickness of 0.16 to 0.20 and peppers with a pericarp thickness ≥0.30 (Table 5). Moderate differentiation (FST = 0.05 to 0.15) was detected between peppers with a pericarp thickness of 0.05 to 0.10 and peppers with a pericarp thickness ≥0.30. These data, combined with pericarp differences among continents, suggest that the differentiation is a result of pericarp thickness and not just a pleiotropic difference between wild and cultivated lines.

thumbnail
Fig 2. Population structure of pepper (Capsicum annuum) grouped by pericarp thickness categories.

Individuals are represented by their proportionate membership (0 to 1) in cluster 1 (purple), cluster 2 (light yellow), cluster 3 (sky blue), cluster 4 (steel blue), or cluster 5 (orchid). A white space and black tick marks separate subgroups of individuals.

https://doi.org/10.1371/journal.pone.0156969.g002

thumbnail
Table 5. Genetic differentiation of pepper lines when grouped by pericarp thickness (cm).

https://doi.org/10.1371/journal.pone.0156969.t005

For pepper fruit shape, moderate to very great differentiation was detected among many of the predefined categories using the descriptors designated by Rodriguez et al [14]. Flat peppers were very greatly differentiated from long and rectangular peppers, but not significantly differentiated from oxheart-shaped peppers (Table 6). Oxheart-shaped and rectangular peppers had little differentiation from long peppers. Cluster 3 (light blue) was not detected in the heart shape category. Clusters 1 (dark purple) and 5 (orchid) were not detected in the flat or oxheart categories (Fig 3). Cluster 5 was also not detected in the rectangular category. The round pepper category was not represented by three or more individuals, and comparisons could not be made with remaining fruit shape categories. For the Bosland shape descriptors, no significant differentiation was detected among categories. End shape (pointed or blunt) had little differentiation (0.0001) among the subpopulations. Only cluster 5 (orchid) was underrepresented in blunt individuals (Fig 4). When grouped by country, clusters did not perfectly coincide with categories (Fig 5).

thumbnail
Fig 3. Population structure of pepper (Capsicum annuum) grouped by fruit shape categories described by Rodriguez et al [14].

Only categories represented by more than four individuals are included. Individuals are represented by their proportionate membership (0 to 1) in cluster 1 (purple), cluster 2 (light yellow), cluster 3 (sky blue), cluster 4 (steel blue), or cluster 5 (orchid). A white space and black tick marks separate subgroups of individuals.

https://doi.org/10.1371/journal.pone.0156969.g003

thumbnail
Fig 4. Population structure of pepper (Capsicum annuum) grouped by fruit end shape categories.

Individuals are represented by their proportionate membership (0 to 1) in cluster 1 (purple), cluster 2 (light yellow), cluster 3 (sky blue), cluster 4 (steel blue), or cluster 5 (orchid). A white space and black tick marks separate subgroups of individuals.

https://doi.org/10.1371/journal.pone.0156969.g004

thumbnail
Fig 5. Population structure of pepper (Capsicum annuum) grouped by country of origin.

Only countries represented by more than four individuals are included. Individuals are represented by their proportionate membership (0 to 1) in cluster 1 (purple), cluster 2 (light yellow), cluster 3 (sky blue), cluster 4 (steel blue), or cluster 5 (orchid). A white space and black tick marks separate subgroups of individuals.

https://doi.org/10.1371/journal.pone.0156969.g005

thumbnail
Table 6. Genetic differentiation of pepper lines when grouped by fruit shape categoriesA.

https://doi.org/10.1371/journal.pone.0156969.t006

When fruit shape index parameters from TA were compared to visual fruit shape categories, variation was evident. For heart-shaped fruit, the average fruit shape index was 1.28 and ranged from 1.01 to 1.44. The average maximum width was 1.32 cm and ranged from 1.16 to 1.46 cm and the average maximum height was 1.65 cm with a range of 1.17 to 2.11 cm. For long fruit, the average fruit shape index was 3.06 and ranged from 1.75 to 4.29. The maximum width for long fruit was 1.84 cm with a range of 0.68 to 3.57 cm. The average maximum height was 1.65 cm with a range of 1.17 to 2.11 cm. For rectangular fruit, the average fruit shape index was 1.86 and ranged from 0.70 to 3.45. The average maximum width was 2.31 cm with a range of 0.63 to 4.22 cm, and the average maximum height was 3.78 cm with a range of 0.88 to 6.82 cm. Smaller fruit with a higher fruit shape index were frequently found in South America. North America wasn't significantly different, and could be the result of including small-fruited breeding lines. Population structure also supported differences among categories with certain clusters being more frequently associated with some categories compared to others. For example, cluster 1 (dark purple) was more frequently associated with thinner pericarp, and long or rectangular shaped peppers (Fig 3). This cluster was also more frequently associated with fruit from Mexico and the USA, consistent with similarities in fruit shape index and size. Combining these data with metabolomic or disease data could provide useful tools for germplasm selection [26,38].

The combination of markers and fruit shape categories used in this study was not sufficient for separating the population structure of fruit shape. When fruit shape was designated using the terms described by Bosland, no significant differentiation was detected among categories. When fruit shape was grouped by categories described by Rodriguez [14,39], significant differentiation was detected between five of the category combinations. However, the Rodriguez categories were not sufficient to perfectly differentiate each shape using these markers. Fruit shape index, maximum width, and maximum height values from TA were associated with a range of values for each of the categorical descriptors of shape. For heart and oxheart-shaped peppers, the fruit shape index ranges were small and may be predictive of actual fruit shape. For long and rectangular peppers, however, the fruit shape index ranges were broad suggesting that further division of shape categories will be needed.

Combining subjective definitions such as those employed by the INRA with objective TA measurements may improve separation of shape categories in pepper [19,37]. Using a controlled and accurate categorical definition of fruit shape in pepper will improve the classification and delineation of shape categories, which in turn can improve our ability to determine genetic components controlling shape. Further refinement of fruit shape categories, their associations to fruit shape alleles such as caOvate and fruit shape index values are needed to further our understanding of fruit shape in pepper.

The fruit shape index was positively correlated with fruit perimeter (r = 0.2267, P<0.0001), and height (midpoint (r = 0.4979, P< 0.0001) and maximum (r = 0.4822, P<0.0001)), but negatively correlated with fruit pericarp (r = -0.3587, P<0.0001) and width (midpoint (r = -0.4626, P<0.0001) and maximum (r = -0.3915, P<0.0001)). Fruit shape index is measured as the ratio of fruit length to width, and previously studies have shown positive and negative correlations with fruit length and width, respectively [40]. Similarly pericarp thickness was negatively correlated with fruit width, consistent with results from Dwivedi et al [41], while fruit shape index was negatively associated similar to Rao et al [42]. Fruit shape and flesh thickness are important considerations for cultivar classification (bell, cheese, jalapeno, habanero, serrano, poblano, shishito, and thai). Linkage between fruit shape characteristics could affect the speed at which, traits such as flavor compounds could be integrated from chili-type peppers into the sweet bell and cheese-type peppers.

Fruit shape triangle (the ratio of the width at the upper position to the width at the lower position) was positively correlated only with pericarp thickness (r = 0.1233, P = 0.0002). Tomato Analyzer fruit shape identifiers (Elliptical, Circular, Rectangular, Obovoid and Ovoid) varied in correlation and significance with remaining fruit categories (S2 Table). Shoulder height was positively correlated with pericarp thickness (r = 0.1541, P<0.0001), perimeter (r = 0.0999, P = 0.0019), width (midpoint (r = 0.1344, P<0.0001) and maximum (r = 0.1316, P<0.0001).) The TA shape designations (Circular, Rectangular, and Ellipsoid) were significantly associated with most of the remaining traits evaluated. Pericarp thickness was negatively correlated for TA Circular (r = -0.2899, P < 0.0001) and positively correlated with Rectangular (r = 0.1347, P <0.0001) categories, indicating that pericarp was thicker for more rectangular peppers and less thick for circular peppers.

Perimeter was positively correlated with TA calculated categories Ellipsoid (r = 0.4086, P< 0.0001) and Circular (r = 0.3203, P<0.0001) categories, with moderate r values. Fruit shape index was positively correlated with the TA Circular (r = 0.8434, P< 0.0001) and Ellipsoid (r = 0.3052, P < 0.0001) and negatively correlated with TA Rectangular (r = -0.2225, P< 0.0001). Based on correlations among fruit traits, only fruit shape index, pericarp thickness, fruit shape triangle, and shoulder height were used for disease-fruit trait correlations.

Previously, a small, yet significant, isolate-specific correlation between fruit shape and disease susceptibility to P. capsici in a small mapping population was identified [40]. However, no other fruit traits were correlated with disease susceptibility. In a study by Biles et al, cuticle thickness, but not pericarp thickness was associated with disease resistance [43]. In this study, susceptibility to Phytophthora fruit rot was significantly positively correlated with pericarp thickness for both isolates evaluated at three and five dpi (Table 7). Fruit shape was negatively correlated with isolate OP97 at five dpi (r = -0.1618, P<0.0001), and isolate 12889 at three (r = -0.0684, P = 0.0362) and five (r = -0.1221, P = 0.0002) dpi. Correlations were significant, albeit weak, suggesting that fruit shape and thickness may be linked to disease susceptibility in some genetic backgrounds. Fruit shape triangle was weakly positively associated with susceptibility to isolate 12889 at five dpi (r = 0.0682, P = 0.0368). Shoulder height was positively associated with susceptibility to isolate 12889 at three (r = 0.0953, P = 0.0035) and five (r = 0.0847, P = 0.0094) dpi. Fruit shape triangle and shoulder height were not significantly correlated with susceptibility to isolate OP97. These correlations were both weaker and isolate specific, suggesting that breeders will be able to separate the traits with minimal effort. Understanding the broad sense heritability and potential correlations of fruit traits can greatly reduce the time to develop a commercially acceptable cultivar. In the Solanaceae, wild relatives are an important source of important horticultural traits such as abiotic and biotic resistance [21,22,44,45]. When these traits are incorporated into commercial backgrounds, deleterious or undesirable characteristics must be removed through repeated backcrossing to a commercial parent. This can take numerous generations depending on the trait, its ease of phenotyping, heritability, and any available molecular markers. In some instances, these traits may also be negatively linked with favorable traits such as yield or disease susceptibility [23,25].

thumbnail
Table 7. Pearson’s correlation coefficient (r) between pepper fruit traits and disease resistance to Phytophthora capsici.

https://doi.org/10.1371/journal.pone.0156969.t007

Conclusion

Previously, fruit shape was negatively correlated with disease resistance to a single isolate, but no correlation was detected with pericarp thickness in a pepper mapping population [15]. However in this study, disease susceptibility was positively correlated with increased pericarp thickness for both isolates at three and five dpi. Based on these results, thin fruit were more resistant to Phytophthora capsici across the collection. Fruit shape was also negatively correlated with disease susceptibility to both isolates at 5 dpi, consistent with the previous study [15]. Fruit perimeter was positively associated with disease susceptibility for isolate 12889 at three and five dpi, but not OP97. Similar results with fruit shape triangle and shoulder height suggest that isolate-specific correlations may also confound breeding for fruit traits.

These data suggest that peppers with thicker flesh, similar to those seen in North America and Europe, tend to be more susceptible to P. capsici. While this does not directly translate to a reduction in yield, it indicates that breeding thick-fruited bell peppers with the preferred size and shape and sufficient Phytophthora fruit rot resistance may be a challenge. Negative and positive correlations among fruit horticultural traits and disease resistance traits can complicate the breeding process. However, using controlled fruit characteristic vocabularies, and understanding the correlations among fruit traits and disease resistance will be essential for continued crop improvement.

Supporting Information

S1 Table. Pepper fruit measurements for width at midpoint (width at mid), height at midpoint (height at mid), fruit shape triangle, and ellipsoid in cm.

https://doi.org/10.1371/journal.pone.0156969.s001

(XLSX)

S2 Table. Pearson Correlation Coefficients for fruit traits.

https://doi.org/10.1371/journal.pone.0156969.s002

(XLSX)

Acknowledgments

The authors would like to acknowledge the technical assistance of S. Boyle and A. Tomlinson, and the statistical assistance of J. Fry at Michigan State University. Mention of trade names or commercial products in this publication is solely for the purpose of providing specific information and does not imply recommendation or endorsement by the U.S. Department of Agriculture. USDA is an equal opportunity provider and employer.

Author Contributions

Conceived and designed the experiments: RN. Performed the experiments: RN JM. Analyzed the data: RN. Contributed reagents/materials/analysis tools: MH. Wrote the paper: RN MH JM.

References

  1. 1. Paran I, Knaap E van der. Genetic and molecular regulation of fruit and plant domestication traits in tomato and pepper. J Exp Bot. 2007;58: 3841–3852. pmid:18037678
  2. 2. Bharath SM, Cilas C, Umaharan P. Fruit trait variation in a caribbean germplasm collection of aromatic hot peppers (Capsicum chinense Jacq.). Hortscience. 2013;48: 531–538.
  3. 3. Ku HM, Doganlar S, Chen KY, Tanksley SD. The genetic basis of pear-shaped tomato fruit. Theor Appl Genet. 1999;99: 844–850.
  4. 4. Zygier S, Chaim AB, Efrati A, Kaluzky G, Borovsky Y, Paran I. QTLs mapping for fruit size and shape in chromosomes 2 and 4 in pepper and a comparison of the pepper QTL map with that of tomato. Theor Appl Genet. 2005;111: 437–445. pmid:15983758
  5. 5. Hurtado M, Vilanova S, Plazas M, Gramazio P, Herraiz FJ, Andújar I, et al. Phenomics of fruit shape in eggplant (Solanum melongena L.) using Tomato Analyzer software. Scientia Horticulturae. 2013;164: 625–632.
  6. 6. Brewer MT, Lang L, Fujimura K, Dujmovic N, Gray S, van der Knaap E. Development of a controlled vocabulary and software application to analyze fruit shape variation in tomato and other plant species. Plant Physiol. 2006;141: 15–25. pmid:16684933
  7. 7. Nunome T, Ishiguro K, Yoshida T, Hirai M. Mapping of fruit shape and color development traits in eggplant (Solanum melongena L.) based on RAPD and AFLP markers. Breeding Science. 2001;51: 19–26.
  8. 8. Qiao J, Chen Y, Wang L, Liu F, Zhang Y, Lian Y. QTL analysis for fruit shape in eggplant based on genetic linkage map. Acta Horticulturae Sinica. 2012;39: 1115–1122.
  9. 9. Chaim AB, Paran I, Grube RC, Jahn M, Wijk R van, Peleman J. QTL mapping of fruit-related traits in pepper (Capsicum annuum). Theor Appl Genet. 2001;102: 1016–1028.
  10. 10. Xue-jun C, Rong F, Kun-hua Z, Nan-sheng M, Chang-lin H. Major gene plus polygene inheritance analysis of fruit traits in pepper. Xibei Zhiwu Xuebao. 2012;32: 246–251.
  11. 11. ShuGen Z, ZhongRen J, YongPing X, ChunLing L. Genetic analysis of fruit traits based on a double haploid population from a hybrid of pepper (Capsicum annuum L.). Acta Horticulturae Sinica. 2008;35: 515–520.
  12. 12. Xiao H, Radovich C, Welty N, Hsu J, Li D, Meulia T, et al. Integration of tomato reproductive developmental landmarks and expression profiles, and the effect of SUN on fruit shape. BMC Plant Biology. 2009;9: 49. pmid:19422692
  13. 13. Brewer MT, Moyseenko JB, Monforte AJ, Knaap E van der. Morphological variation in tomato: a comprehensive study of quantitative trait loci controlling fruit shape and development. J Exp Bot. 2007;58: 1339–1349. pmid:17283371
  14. 14. Rodríguez GR, Muños S, Anderson C, Sim S-C, Michel A, Causse M, et al. Distribution of SUN, OVATE, LC, and FAS in the tomato germplasm and the relationship to fruit shape diversity. Plant Physiol. 2011;156: 275–285. pmid:21441384
  15. 15. Naegele R, Hill T, Ashrafi H, Reyes Chin-Wo S, Van Deynze A, Hausbeck MK. QTL mapping of fruit rot resistance to the plant pathogen Phytophthora capsici Leonian in a recombinant inbred line Capsicum annuum L. population. Phytopathology. 2013;
  16. 16. Tsaballa A, Pasentsis K, Darzentas N, Tsaftaris AS. Multiple evidence for the role of an Ovate-like gene in determining fruit shape in pepper. BMC Plant Biol. 2011;11: 46. pmid:21401913
  17. 17. Ben Chaim A, Borovsky Y, Rao GU, Tanyolac B, Paran I. fs3.1: a major fruit shape QTL conserved in Capsicum. Genome. 2003;46: 1–9. pmid:12669791
  18. 18. Vilarinho LBO, Silva DJH da, Greene A, Salazar KD, Alves C, Eveleth M, et al. Inheritance of fruit traits in Capsicum annuum: heirloom cultivars as sources of quality parameters relating to pericarp shape, color, thickness, and total soluble solids. J Amer Soc Hort Sci. 2015;140: 597–604.
  19. 19. Sage-Palloix AM, Jourdan F, Phaly T, Nemouchi G, Lefebvre V, Palloix A. Analysis of diversity in pepper genetic resources: distribution of horticultural and resistance traits in the INRA pepper germplasm. Niemirowicz-Szezytt K, editor. 2007;
  20. 20. Gonzalo MJ, Brewer MT, Anderson C, Sullivan D, Gray S, Knaap E van der. Tomato fruit shape analysis using morphometric and morphology attributes implemented in tomato analyzer software program. J Amer Soc Hort Sci. 2009;134: 77–87.
  21. 21. Ashrafi H, Kinkade M, Foolad MR. A new genetic linkage map of tomato based on a Solanum lycopersicum x S. pimpinellifolium RIL population displaying locations of candidate pathogen response genes. Genome. 2009;52: 935–956. pmid:19935918
  22. 22. Kearney B, Staskawicz BJ. Widespread distribution and fitness contribution of Xanthomonas campestris avirulence gene avrBs2. Nature. 1990;346: 385–386. pmid:2374611
  23. 23. Haggard JE, Johnson EB, Clair DAS. Linkage relationships among multiple QTL for horticultural traits and late blight (P. infestans) resistance on chromosome 5 introgressed from wild tomato Solanum habrochaites. G3. 2013;3: 2131–2146. pmid:24122052
  24. 24. Naegele RP, Boyle S, Quesada-Ocampo LM, Hausbeck MK. Genetic diversity, population structure, and resistance to Phytophthora capsici of a worldwide collection of eggplant germplasm. PLoS ONE. 2014;9: e95930. pmid:24819601
  25. 25. Cheng C-H. Inheritance of resistance to Pseudomonas syringae pv. actinidiae and genetic correlations with fruit characters in a diploid Actinidia chinensis (kiwifruit) population. Euphytica. 2014;198: 305–315.
  26. 26. Naegele RP, Tomlinson AJ, Hausbeck MK. Evaluation of a diverse, worldwide collection of wild, cultivated, and landrace pepper (Capsicum annuum) for resistance to Phytophthora fruit rot, genetic diversity, and population structure. Phytopathology. 2014;105: 110–118.
  27. 27. USDA, ARS, National Genetic Resources Program. Germplasm Resources Information Network—(GRIN). [Internet]. [cited 1 Apr 2015]. Available: http://www.ars-grin.gov.4/cgi-bin/npgs/html/eval.pl?490665
  28. 28. Fehr WR. Principles of cultivar development. New York, USA: Macmillan Publishing Company; 1987.
  29. 29. Knapp SJ, Stroup WW, Ross WM. Exact confidence intervals for heritability on a progeny mean basis. Crop S. 1985;25: 192–194.
  30. 30. Pritchard JK, Stephens M, Donnelly P. Inference of population structure using multilocus genotype data. Genetics. 2000;155: 945–959. pmid:10835412
  31. 31. Falush D, Stephens M, Pritchard JK. Inference of population structure using multilocus genotype data: linked loci and correlated allele frequencies. Genetics. 2003;164: 1567–1587. pmid:12930761
  32. 32. Evanno G, Regnaut S, Goudet J. Detecting the number of clusters of individuals using the software structure: a simulation study. Molecular Ecology. 2005;14: 2611–2620. pmid:15969739
  33. 33. Earl DA, vonHoldt BM. STRUCTURE HARVESTER: a website and program for visualizing STRUCTURE output and implementing the Evanno method. Conservation Genet Resour. 2011;4: 359–361.
  34. 34. Liu K, Muse SV. PowerMarker: an integrated analysis environment for genetic marker analysis. Bioinformatics. 2005;21: 2128–2129. pmid:15705655
  35. 35. Hartl DL, Clark AG. Principles of population genetics. 4th ed. Sunderland, Massachusetts: Sinauer Associates, Inc. Publishers; 2007.
  36. 36. Kraft KH, Brown CH, Nabhan GP, Luedeling E, Ruiz J de JL, d’Eeckenbrugge GC, et al. Multiple lines of evidence for the origin of domesticated chili pepper, Capsicum annuum, in Mexico. PNAS. 2014;111: 6165–6170. pmid:24753581
  37. 37. Nicolaï M, Cantet M, Lefebvre V, Sage-Palloix A-M, Palloix A. Genotyping a large collection of pepper (Capsicum spp.) with SSR loci brings new evidence for the wild origin of cultivated C. annuum and the structuring of genetic diversity by human selection of cultivar types. Genet Resour Crop Evol. 2013;60: 2375–2390.
  38. 38. Wahyuni Y, Ballester A-R, Tikunov Y, Vos RCH de, Pelgrom KTB, Maharijaya A, et al. Metabolomics and molecular marker analysis to explore pepper (Capsicum sp.) biodiversity. Metabolomics. 2013;9: 130–144. pmid:23335867
  39. 39. Rodriguez JM, Berke T, Engle L, Nienhuis J. Variation among and within Capsicum species revealed by RAPD markers. Theor Appl Genet. 1999;99: 147–156.
  40. 40. Naegele RP, Hausbeck MK. Evaluation of pepper fruit for resistance to Phytophthora capsici in a recombinant inbred line population, and the correlation with fruit shape. Plant Disease. 2014;98: 885–890.
  41. 41. Dwivedi Neeraj, Kumar Rajesh, Paliwal Rajneesh, Kumar Uttam, Kumar Sanjeet, Singh Major, et al. QTL mapping for important horticultural traits in pepper (Capsicum annuum L.). Journal of Plant Biochemistry and Biotechnology. 2015;24: 154–160.
  42. 42. Rao GU, Chaim Ben A, Borovsky Y, Paran I. Mapping of yield-related QTLs in pepper in an interspecific cross of Capsicum annuum and C. frutescens. Theoretical and Applied Genetics. 2003;106: 1457–1466. pmid:12750789
  43. 43. Biles CL. Relationship of Phytophthora fruit rot to fruit maturation and cuticle thickness of New Mexican-type peppers. Phytopathology. 1993;83: 607.
  44. 44. Moreau P, Thoquet P, Olivier J, Laterrot H, Grimsley N. Genetic mapping of Ph-2, a single locus controlling partial resistance to Phytophthora infestans in tomato. MPMI. 1998;11: 259–269.
  45. 45. Song J, Bradeen JM, Naess SK, Raasch JA, Wielgus SM, Haberlach GT, et al. Gene RB cloned from Solanum bulbocastanum confers broad spectrum resistance to potato late blight. PNAS. 2003;100: 9128–9133. pmid:12872003