:: Korean Journal of Breeding Science ::
Journal Search Engine
Search Advanced Search Adode Reader(link)
Download PDF Export Citaion korean bibliography PMC previewer open access
ISSN : 0250-3360(Print)
ISSN : 2287-5174(Online)
Korean Journal of Breeding Science Vol.44 No.4 pp.476-482
DOI : https://doi.org/10.9787/KJBS.2012.44.4.476

Genetic Analysis of Carotenoids Content in Red Pepper (Capsicum annuum L.) Fruit using Complete Diallel Cross

Byoung Cheorl Kang5*, Woo-Moon Lee1, Jung-Ho Kwak1, Su Kim1, Won Byoung Chae1, Dong Kum Park1, Moo Kyoung Yoon1, Do Ham Pae2, Dae-Geun Oh3, Hyo Geun Park4
5Dept. of Plant Science, College of Agriculture and Life Sciences, Seoul National University,
1Dept. of Horticultural Crop Research, National Institute of Horticultural & Herbal Science, RDA, 2Foundation of Agricultural Technology Commercialization & Transfer, 3Korea National College of Agriculture and Fisheries, 4Research and Development Unit, Pepper and Breeding Institute,
Received on September 28, 2012. Revised on November 23, 2012.


Regarding carotenoids content, the genetic basis, heritability and combining ability in six red pepper inbred lines wereinvestigated using full diallel crosses. Both additive and non-additive gene actions govern inheritance of carotenoids content. Themean square of array through variance and covariance analysis (Wr-Vr) was insignificant, which suggest that inbred linesinvolved in diallel cross may have no epistatic effects. The Vr/Wr graph revealed the influence of partial dominant gene actiontowards low carotenoids content and the absence of non-allelic interaction. The H2 component was smaller than the H1 and the[H2/4H1] component was 0.187 less than 0.25, indicating unequal proportion of positive and negative alleles in the parents. Theestimates of broad and narrow sense heritability for carotenoids content were 0.956 and 0.832, respectively. The variance ofgeneral combining ability (GCA) was relatively higher than that of specific combining ability (SCA), which implied that theadditive gene effects were predominant as compared to both dominant and epistatic effects for the accumulation of carotenoidsin this genetic population. The values of GCA of ‘62024L1’ and ‘62067L2’ were higher than those of the other parents. These2 inbred lines, therefore, can be considered as useful breeding materials to enhance fruit carotenoids content in other red peppervarieties.



In Korea, red pepper is used mainly in the form of ground powder and its market value depends on the degree of red color. The color of red pepper powder is important as one of the main quality factors and the good color tone makes some Korean side dishes look more delicious. Their horticultural characteristics were described in Table 1. The fruit color is affected by intricate factors, such as variety, cultivation environment, harvesting time and processing method (Davis et al. 1970, Lee et al. 1973, Mínguez-Mosquera & Hornero-Méndez 1994, Hornero-Méndez et al. 2000). The deep red color of Capsicum pepper fruit comes from carotenoid pigments, mostly synthesized during fruit ripening. The final red color of the mature fruit is developed by the carotenoid pigments such as capsanthin, capsorubin, and capsanthin 5, 6-epoxide, which are contained exclusively in the genus Capsicum (Curl 1962, Lefebvre et al. 1998). Recently, the gene for capsanthin-capsorubin synthase (CCS), which plays a role in conversion of antheraxanthin to capsanthin and violaxanthin to capsorubin, was identified (Throrup et al. 2000, Hur et al. 2001). In order to obtain genetic information on quantitative traits, the technique of diallel mating has been used (Hayman 1954a, b, Jinks 1954, Mather & Jinks 1982). Haymam-Jinks model provided useful information on genetic mechanism of plant traits in early generation like F1 Griffing (1956) theorized general (GCA) and specific combining ability (SCA) in relation with diallel crossing system. In the previous studies on the quantitative genetic analysis for carotenoids content, Om & Pyo (1981) and Kweon et al. (2006) reported broad-sense heritability was high while additive gene effects were low. Also, recessive genes were predominant compared to dominant genes in the parental lines.

Park & Harn (1984) and Park (1986) described that the carotenoids content of F1  showed heterosis and it is evident that there was no significant effect on SCA. On the contrary, Todorov & Manuelyan (1989) reported the inheritance of carotenoids content in Capsicum as partially dominant toward to low carotenoids content. The purpose of this study was to select elite lines for breeding high carotenoids content varieties through clarification of inheritance mode using the analysis of six parental diallel crosses.


Plant material

Eight lines of red pepper were selected from Capsicum germplasm at vegetable research division of NIHHS (National Institute of Horticultural & Herbal Science) based on variation for carotenoids content in mature fruit and crossed among each line with a full diallel fashion in 2004. The following year, 56 hybrids and its parents were planted in the experimental fields on May, 20th, 2005 by triplicate randomized block design. In each replication, there were 20 plants spaced at a distance of 150 cm between rows and 25 cm within rows.


Analysis of carotenoids content

The fruits of red pepper were harvested at fully matured stage and dried at 60℃ for 3 days. Extraction of total carotenoids from ground red pepper except seeds was carried out according to the previously described methods (Mínguez -Mosquera et al. 1992). The total carotenoids content was calculated from

 Where If  is a correction factor for the apparatus.

Statistical analysis

 Hayman’s diallel analysis procedure (1954a) was used to analyze the different type of gene actions such as additive (a), dominance (b), overall maternal (c), and residual reciprocal (d) effects in a diallel table. The variance (Vr)/ covariance (Wr) and the genetic components of variance (D, F, H1 , H2, h2 , and E) were calculated using the procedures given by Hayman (1954b), Mather & Jinks (1982), and Schaff et al. (1987). The analysis of variance for GCA and SCA was carried out according to Griffing (1956) using method III, model I.

For the genetic analysis of diallel cross experiment, a program based on the SAS statistics package(v. 9.1, NC, U.S.A.) was developed referring to the paper of Kim & Bae (2002), “The Guide for Breeding Experiment” (Kim & Choi, 1996), and DIAL98 program which was an upgraded version of the original Ukai’s (1989) software.


The horticultural traits of the eight inbred lines are listed in Table 1. The diallel table for the total carotenoids content of eight inbred lines and their F1 crosses with reciprocals are presented in Table 2.

Table 1. The horticultural traits of 8 parental lines used for the construction of diallel crosses.

Table 2. Total carotenoids content of eight parental lines and their 56 F1 progenies and maternal effect of eight parents.

Analysis of variance in the diallel table for carotenoids contents revealed that both additive (a) and dominant (b) effects were significant (Table 3). The average maternal effect (c) was not significant, while the specific maternal effect (d) was significant. Because no maternal effect is assumed for genetic analysis procedure (Hayman 1954b), we found out two parental lines, ‘62066R8’ (P3) and ‘62019R4’ (P7), which had relatively high maternal effects (Table 2). Therefore, these two lines were eliminated from the diallel table to make a 6 × 6 sub-diallel table. In the analysis of sub-diallel table, additive (a) and dominant effects (b) were significant but the maternal effect became not significant (Table 3).

Table 3. Analysis of variance of 8×8 and 6×6 diallel table for total carotenoids content according to Mather and Jinks (1982).

The b2 and b3 portion of the b item were significant, which exhibited asymmetrical distribution of dominant genes in the parents with the presence of dominant interaction between specific genotypes in both eight and six parental crosses (Table 3). These results were not consistent with the previous report by Park (1986) except for additive effect (a) component. 

The adequacy of additive-dominance model was validated by the analysis of variance for (Wr+Vr) and (Wr-Vr). Wr+Vr between array mean square were significant, which indicates the presence of non-additive gene action. The result of Wr-Vr analysis validated the diallel assumption, since Wr-Vr mean square was non-significant (Table 4). 

Table 4. Analysis of variance for Wr+Vr and Wr-Vr estimates (6×6).

On the basis of simple Hayman-Jinks model, the Vr, Wr graph (Fig. 1A) has a regression coefficient (b) that was not significantly different in unity (slope=0.9834±0.1213). Therefore, the genes controlling the inheritance of carotenoids content appeared to interact cumulatively without independent action of non-allelic genes. This result was different from the existence of epistatic effects within parents by the result of Om & Pyo (1981). As the regression line passed above the point of origin, it proved that the inheritance of carotenoids content was influenced by partial dominance. This finding of partial dominance was also reported by Todorov & Manuelyan (1989). The parents, P8 and P4 occupied positions near the point of origin indicating predominance of dominant genes. Conversely, the parents, P2 and P1 were located at the positions away from the point of origin, indicating relatively excessive recessive genes. The Yr, Wr+Vr graph (Fig. 1 B) also confirmed the above results. The parents, 2 and 1 having relatively high carotenoids were located at the first quadrant demonstrating predominance of recessive genes. 

Fig. 1. Vr, Wr (A) and standardized Yr, Wr+Vr (B) graph of six parents diallel analysis for carotenoids content.

Genetic component of variation

The genetic information estimated from sub-diallel table is presented in Table 5. The additive component value (D) was larger than the genetic components H1and H2, which imply that additive genetic effects were more pronounced than dominance effect in the genetic mechanism controlling this trait. The mean degree of dominance at each locus ([H1/D]1/2) was 0.549 indicating incomplete or partial dominance. This result agrees with the Vr, Wr graph shown in Fig. 1A. The H1 and H2 components were shown to be unequally different distribution of positive and negative genes in the parents. This fact was also supported by H2/4H1 (uv) ratio which was less than 0.25. The positive value of F suggests that the parental lines used in this study have more dominant genes than recessive genes for total carotenoids content. The ratio of dominant and recessive alleles (kd/kr) was more than 1.0, which also supports the above fact. The non-significance of h2 component for this trait indicates the absence of overall dominance effect by heterozygous loci. The correlation coefficient (r) between Wr+Vr and parental value (Yr) was 0.937, suggesting that the action of dominant genes was negative for carotenoids content. The broad and narrow sense heritabilities were relatively high as 0.956 and 0.832, respectively. This is inconsistent with the result of previous study carried out by Om & Pyo (1986) and Kweon et al. in 2006. The difference may be from the genetic materials and the experimental conditions in this study. 

Table 5. Genetic components of variation for carotenoids content in 6 parental diallel crosses.

Combining ability

The analysis of combining ability derived from crossing experiments is given in Table 6. The variances of GCA and SCA were significant for total carotenoids content. GCA variance was relatively higher than SCA indicating predominance of additive gene effect over dominant effect.

Table 6. Analysis of variance for GCA and SCA effect for carotenoids content in six parental diallel crosses.

As shown in Table 7, P1 and P2 (‘62024L1’ and ‘62067L2’) parents showed high GCA effect while P4 and P8 (‘62026R4’ and ‘62027R3’) parents exhibited otherwise. The GCA estimates of ‘62024L1’(P1) and ‘62067L2’(P2) were significantly higher than the other parents, indicating that additive genes play major role in these two lines. This also suggests that these two lines are potential breeding materials to increase carotenoids content in the fruit of red pepper varieties. 

Table.7. GCA effect and SCA effect of each crosses for carotenoids content in six parental diallel crosses.

SCA effects of F1 hybrids from six parental lines varied considerably showing both positive and negative effects (Table 7), which indicate genetic diversity of the parental lines. The P1×P2 and P4×P5 crosses showed relatively high positive SCA effects. These high SCA effects of F1 hybrids suggest the importance of both additive and non-additive gene action for carotenoids content in red pepper fruit.

The genetic variation for carotenoids content in red pepper appeared to be influenced mainly by genes with additive and dominance effect. In most case, additive effect was predominant compared with dominance effect. Based on high narrow sense heritability (Table 5), it is expected to obtain potentially useful lines possessing high levels of carotenoids through selection in early generations. 


This research was supported by Technology Development Program for Agriculture and Forestry (code:308020-5), Ministry for Food, Agriculture, Forestry and Fisheries, Republic of Korea. 


1.Curl AL. 1962. The carotenoids of red bell peppers. J. Agri. Food Chem. 10: 504-509.
2.Davis GH, Mathews S, Kirk JT. 1970. The nature and biosynthesis of the carotenoids of different colour varieties of Capsicum annuum. Phytochem. 9: 797-805.
3.Griffing B. 1956. Concept of general and specific combining ability in relation to diallel crossing system. Aust. J. Biol. Sci. 9: 463-493.
4.Hayman BI. 1954a. The analysis of variance of diallel tables. Biometrics 10: 235-244.
5.Hayman BI. 1954b. The theory and analysis of diallel crosses. Genetics 39: 789-809.
6.Hornero-Méndez D, Gómez-Ladrón de Guevar R, Mínguez -Mosquera MI. 2000. Carotenoid biosynthesis changes in five red pepper (Capsicun annuum L.) cultivars during ripening. cultivar selection for breeding. J. Agric. Food Chem. 48: 3857-3864.
7.Hur JH, Kang BC, Nahm SH, Kim S, Ha KS, Lee MH, Kim BD. 2001. A candidate gene approach identified phytoene synthase as the locus for mature fruit color in red pepper (Capsicum ssp.). Theor. Appl. Genet. 102: 524-530.
8.Jinks JL. 1954. The analysis of continuous variation in a diallel cross of Nicotiana rustica L. Genetics 39: 767-88.
9.Kim KH, Choi HC. 1996. The guide of breeding experiment. Kor. J. Breed. 319-351.
10.Kim SY, Bae JS. 2002. Program development of genetic analysis for diallel cross experiment. The Korean Communications in Statistics 9(3): 675-682.
11.Kweon OY, Kim YK, Yoon JB, Park HG, Yoon WM. 2006. Inheritance of Carotenoids Content in Red Pepper (Capsicum annuum L.) Fruit. Kor. J. Hort. Sci. Technol. 24(2): 123-127.
12.Lee SW, Kim KS, Lee SS, Jo YK. 1973. Studies on fruit characters and chemical components in several varieties of pepper. J. Kor. Soc. Hort. Sci. 13; 27-34.
13.Lefebvre V, Krutz M, Camara B, Palloix A. 1998. The capsanthin-capsorubin synthase gene: a candidate gene for the y locus controlling the red fruits colour in pepper. Plant Mol. 36: 785-789.
14.Mather K, Jinks JL. 1982. Biometrical Genetics. Third edition. Chapman and Hall Ltd. London. 255-291.
15.Mínguez-Mosquera MI, Jaren-Galan M, Garrido-Fernandez J. 1992. Color quality in paprika. J. Agri. Food Chem. 40: 2384-2388.
16.Mínguez-Mosquera MI, Hornero-Méndez D. 1994. Comparative study of the effect of paprika processing on the carotenoids in peppers (Capsicum annuum) of the Bola and Agridulce varieties. J. Agri. Food Chem. 42: 1555-1560.
17.Om YH, Pyo HK. 1981. Studies on quantitative characters in red pepper (Capsicum annuum L.). J. Kor. Soc. Hort. Sci. 22(4): 231-264.
18.Park JB, Harn DW. 1984. Expression of heterosis and combining ability for carotenoids and vitamin C in intraspecific hybrids of pepper. Thesis-The University of Seoul 18: 209-216.
19.Park JB. 1986. Expression of heterosis and combining ability for component characters of quality in interspecific hybrids of pepper. Thesis-The University of Seoul 20: 153-169.
20.Schaff DA, Milliken GA, Clayberg CD. 1987. Computational procedure for a weighted diallel analysis. TAG. 74: 538-541.
21.Throrup TA, Tanyolac B, Livingstone KD, Povosky S, Paran I, Jahn M. 2000. Candidate gene analysis of organ pigmentation loci in Solanaceae. Proc. Natl. Acad. Sci. USA. 97: 11192-11197.
22.Todorov J & Manuelyan H. 1989. Inheritance of colour content of red pepper for grinding. Eucarpia VIIth Meeting on Genetics and Breeding on Capsicum and Eggplant. pp.49-53. Kragujevac, Yugoslavia.
23.Ukai Y. 1989. A microcomputer program DIALL for diallel analysis of quantitative characters. Jpn. J. Breed. 39: 107-109.