Impact of Breeding Hermaphroditic Melon on Early Production and Yield: Case of Snake Melon (Cucumismelo var. flexuosus) and Tibish (C. melo var. tibish)

M. E. Abdelmohsin¹, A. E. El Jack², M. T. Yousif², A. M. El Naim^{1, *}, E. A. Ahmed², M. Pitrat³

¹Department of Crops Sciences, Faculty of Natural Resources and Environmental Studies, University of Kordofan, Elobeid, Sudan

²Faculty of Agricultural Sciences, University of Gezira, Wad Medani, Sudan

³Genetics and Breeding of Fruits and Vegetables, Montfavetcedex, France

Abstract

Melon sex types are monoecious, andromonoecious, gynoecios or hermaphrodite. Monoecious is dominant (AG) and hermaphrodite is double recessive (ag). Hermaphrodite ‘Paul’ accession from India crossed with snake melon (monoecious) and tibish (andromonoecious) to transfer hermaphroditic character into both types of melon. F1 progenies of monoecious with hermaphrodite were monoecious, in F2 segregation ratio of 9 monoecious: 3 and romonoecious: 3 gynoecious: 1 hermaphrodite was obtained. But in cross of and romonoecious with hermaphrodite and romonoecy was dominant in F1 and ratio of 3 and romonoecious: 1hermaphrodite were obtained indicated that hermaphrodite is recessive to andromoney. Production of hermaphrodite melon is possible through successive selfing progenies of the crosses. Hermaphrodite lines obtained from these crosses exceeded their parents in earliness and number of fruits/plant.

Keywords

Hermaphrodite, Monoecious, Melon, Sex Expression, Melon Production

Received:May 17, 2015
Accepted: May 27, 2015
Published online: July 7, 2015

@ 2015 The Authors. Published by American Institute of Science. This Open Access article is under the CC BY-NC license. http://creativecommons.org/licenses/by-nc/4.0/

1. Introduction

Most species in the genus Cucumis are monoecious although some are dioecious[1]. In melon (Cucumismelo L.), 60 to 70% cultivated melons are and romonoecious but wild melons are monoecious. Monoecism is found mainly in the flexuosus, acidulus, chate and momordica groups. And romonoecism is controlled by a single recessive gene [2], which was first named monoecious gene (symbol M) and now is and romonoecious gene (symbol a). Another phenotype, hermaphroditism, has been found in very few accessions. Poole and Grimball [3] studied the genetic control of sex expression in melons and they concluded that there is a recessive gene named gynoecious (symbol g), independent from gene a and the interaction between the two loci is as follows: A Gmonoecious, Aggynoecious, a Gandro monoecious and ag hermaphrodite. In tibish melon two linked genes (A-2 and A-3) were involved in the loci of and romonoecya/A[4]. Production hermaphroditic plants are necessary to use plants of sex types that enable self-pollination. However, many dicotyledonous plants, and in particular Cucurbitaceae, may be monoecious, and romonoecious, gynoecious or hermaphroditic. In cucumber, gynoecious plants were used for hybrid production and the fruits developed without fertilization in term of parth enocarpy, but this phenomenon not reported in melon. The implementation means for producing hermaphroditic is possible by conducting chemical treatments to plants, so plants get the ability for self-pollination. In melon, for example, spraying of ethylene inhibitors such as silver nitrate or silver thiosulfate enhances stamens in female flowers [5], [6]. In melons female or perfect flowers are appeared at the first and second nodes of the lateral branches in one-flowered sometimes two-flowered inflorescences in monoecious or and romonoecious plants, then male inflorescences appeared in the following nodes of the lateral branches and in the main stem in multi-flowered inflorescences, but in hermaphrodite plant flowers appeared in main stem and lateral branches in one-flowered. Breeding hermaphrodite cultivars, with perfect flowers at each node on the main stem and the lateral branches, belonging to cultivate groups of melon where fruits are harvested before maturity, as in flexuosus or tibish groups, could be a possibility to increase earliness and yield. Accordingly, the aims of this study are to study the possibility of transferring hermaphroditic character into snake melon and tibish and its effects in earliness and yield production.

2. Materials and Methods

2.1. Snake Melon

Four genotypes were used in this study; three monoecious cultivars belonging to the flexuosus group, namely, Alimin from Sudan, PI222187 from Afghanistan and Snakemelon from Saudi Arabia and the fourth was the hermaphroditic melon accession Paul from Far-East supplied by B. Kubicki. The F₁ and F₂ progenies between the three monoecious accessions and Paul were produced by hand pollination. The evaluation of 100 plants of each F₂ progeny was conducted on young plants in pots (9 cm in diameter). The hermaphroditic plants were selected, transplanted in the greenhouse and selfed and crossed to their flexuosus parents to produce F₃ and BC F₂ progenies. Twenty three F₃ lines of the cross Alimin´ Paul, seven F₃ lines of the cross PI222187 ´ Paul and thirteen F₃ lines of the cross Snakemelon ´ Paul were grown under field conditions at the University of Gezira farm (Sudan) during the winter season of 2006/2007. The plants were evaluated for sex type; the longest ovary plants of F₃ were selfed to produce F₄.

Bc F2 plants of each cross evaluated for sex type, then hermaphrodite plants selfed four times under field condition of Sudan and greenhouse condition of INRA during seasons 2006/07 and 2007/08 successively.

F4 plants evaluated for sex type under greenhouse condition of INRA season 2007, hermaphroditic plants crossed to their parents to produce Bc F4 progenies.

Bc F4 plants of each cross evaluated for sex type, then hermaphrodite plants selfed twice under field and greenhouse conditions of Sudan and France during 2007/08 successively.

The following BC progenies were evaluated on the bases of agronomic characteristics:

- BC F₄(Alimin´ Paul) ´Alimin

- BC F₄(Snakemelon´ Paul) ´ Snakemelon

- BC Alimin´ F₂(Alimin´ Paul)

These genotypes were tested for agronomic traits, in comparison with their flexuosus parents (Snakemelon, PI 222187, Alimin) and the popular cultivar of snake melon ‘Silka’ in Sudan, in RCBD (Randomized Complete Block Design) with 3 replications.

The parameters collected were:

●    Earliness (days from sowing to the first perfect flower in the plant)

●    Yield (number of fruits/plant, fruit yield/plant).

2.2. Tibish

Three accessions of tibish were used, two and romonoecious accessions collected from Kordofan region during season 2005, namely tibish JabelKordofan 4,Tibishkhurtagat 15 and a monoecious tibish entry collected from Managil area, Gezira State, named as tibishManagil 36, were crossed with hermaphrodite accession ‘Paul’ (hermaphrodite x andromonoecious) in 2007 under greenhouse conditions at INRA, France to produce F1 progenies of each cross.

The F1 of each cross were evaluated in Sudan 2007/2008 under field conditions and selfed to produce F2.

Thirty plants of each F2 progeny were grown in pots under greenhouse conditions at INRA, France in 2008, then evaluated for sex types. The segregated hermaphroditic plants were selected, selfed and crossed with their parent to produce F3 and BC F₂ progenies. Then, F3 and BC F2 plants seeds of each progeny were grown under field conditions of Sudan and the plants were evaluated for sex type.

The following selected genotypes were evaluated for some agronomic characteristics: BcTibishkordofan 4 x F2 (Paul x Tibishkordofan 4), BcTibishkhurtagat 15 x F2 (Paul x Tibishkhurtagat 15) and BcTibishManagil 36 x F2 (Paul x TibishManagil 36).

The lines were planted in the experiment with their tibish cultivars parents in RCBD (Randomized Complete Block Design) with 3 replications.

 Parameters measured were:

●    Earliness (days from sowing to the first perfect flower in the plant)

●    Yield (no. fruits/plant, fruit yield/plant).

2.3. Statistical Analysis

Segregation ratios in F₂ populations were tested for goodness of fit to theoretical ratio with Chi square test, analysis of variance for RCBD was used to analyse the data according to Steele and Torrie[17].

3. Results and Discussion

3.1. Snake Melon

F₁plants of crosses between monoecious lines and hermaphroditic Paul were clearly monoecious, indicating that monoecy was dominant in these accessions. In the F₂ generation, the segregation ratio of 9: 3: 3:1 for monoecious: and romonoecious: gynomonoecious: hermaphrodite was obtained, respectively, as presented in (Table 1). In the F₂ (Alimin´ Paul), the segregations ratio was significant (χ²= 5.928 P< 0.11) but not the same as in the F₂(PI 222187 ´ Paul) and the F₂ (Snakemelon´ Paul). In both progenies, the segregating ratios were less than 5%.In some plants, male flowers with rudimentary ovary were observed (abnormal male) which produced fruit with irregular shape (Fig. 1 and 2). The appearance and sequential of the abnormal male was similar to male flowers in the main stem and lateral branches.

Table 1. The segregating generations of F₂ plants (monoecious x hermaphrodite).

Mono: monoecious, Andro: andromonoeciuos, Gyno: gynoecious, Herm: hermaphrodite

Figure 1. Female (left), hermaphrodite (middle) and abnormal male (right) flowers observed in F₂ progenies.

Figure 2. Shape of the fruit produced from abnormal male flowers.

For agronomic trials Table 2 showed the means of earliness, fruits number and fruits weight of the genotypes of BC in advanced stage of genetic purity for hermaphroditic character compared to their monoecious flexuosus parents. All the lines significantly exceeded the parents in earliness (P < 0.01), lines derived from the cross of (Alimin´Paul) produced flowers earlier (33-43 days) than those derived from the cross of Snakemelon ´ Paul (36-57 days). Among the 22 tested lines, only two lines, namely (BC Alimin´ F₂(Alimin´ Paul) A1G3F1A and (BC F₄ (Alimin´ Paul) ´Alimin) A1Y1A3B, were earlier in flowering than the others (33 days). The latest line flowering was BC F₄ (Snakemelon ´ Paul) ´ Snakemelon A1A1A1B (57 days). The parent Snakemelon was the latest in flowering compared to other parents (66 days). The average of the parents was 53 days whereas the average of the lines was 39 days with reduction in earliness by about 26%.

Figure 3. Harvested fruits of hermaphrodite snake melon.

Fruit number exhibited significant differences among the lines and their parents (P < 0.01). The lines gave greater number of fruit (4-25) than the parents (2-6). The highest number of fruits was obtained from the line (BC Alimin´ F₂(Alimin´ Paul) A1A1A1A, which was 25 fruits. The lowest numbers of fruits were obtained from the parents Snakemelon, PI 222187 and Silka, which were 2 and 3, respectively. The percentage of increase in yield (fruit number) was about 275%.

There were highly significant differences in fruit weight (P <0.01) between the lines and their parents. The heaviest fruits were obtained from the parent Alimin, 1967 g (334 g/ fruit). The highest yield was obtained from the line BC Alimin´ F₂ (Alimin´ Paul) BC1I2 A1A1A1A which was 1817g (74 g/ fruit) and line BC Snakemelon ´ F₂ (Snakemelon ´ Paul) A1H1A2B gave 1700 g (107 g/fruit). The fruits produced varied in weight and in shape, which were elongated, oval and round (Fig. 3).

Table 2. The means of the days to flowering (days), number of fruits and fruit weight (fr.wt) of BC lines and their flexuosus parent.

3.2. Tibish

In all crosses between accessions of and romonoecious Tibish, namely (Tibish Jebel Kordofan 4 and TibishKhurtagat 15), with hermaphrodite Paul, F1 plants were entirely and romonoecious indicating that and romonoecy is dominant to hermaphrodite (Table 3).

F2 plants segregated in and romonoecious to hermaphrodite ratio of 3: 1 (X² =0.04 P < 0.91), indicating that and romonoecious was dominant to hermaphrodite (Table 3). All plants in F3 were hermaphrodite.

For the BC of F2 plants with their parents BC Tibish Khurtagat 15 x F2 (Paul x T. Khurtagat 15) compared to BC Tibish Kordofan 4 x F2 (Paul x T. Kordofan 4), the plants obtained were segregated in the ratio of 3 and romonoecious: 1 hermaphrodite (X² = 4.6 P< 0.006and X² =7.68 P< 0.03, respectively). High numbers of hermaphrodite plants were observed.

Table 3. Segregating generation of sex expression of F2 (hermaphrodite x andromonoecious) and their BC.

For agronomic trials the Table 4 showed the means of earliness, number of fruits and fruit weight of the hermaphroditic lines derived from the crosses among hermaphrodite Paul and and romonoecious Tibish varieties. They showed no significant differences with respect to earliness. They started flowering during 34 to 57 days after sowing while the parents started flowering 40- 42 days after sowing.

Number of fruits/ plant exhibited high significant differences among the lines and their parents (P < 0.01). Line BC T. Khurtagat 15 x F2 (Paul x T. Khurtagat 15 (BC1I2 A1J1A) produced a high number of fruits (19.00), similar to its parent T. Khurtagat 15 whereas line BC T. Khurtagat 15 x F2 (Paul x T. Khurtagat 15 (BC1I2 A1H2C) gave a few number of fruits (2.00).

Table 4.The means of days to flowering (days), number of fruit and fruit weight (fr.wt) of the BC lines and their tibish parents.

Fruit weight showed significant differences between lines and their parents (P < 0.01). Heavier fruits were obtained from the parent T. khurtagat 15 (2334.00 gm), with average fruit weight of 125 gm/fruit. Among the tested lines, line BC T. Khurtagat 15 x F2 (Paul x T. Khurtagat 15 (BC1I2 A1J1B) gave the heaviest fruit weight (1875 gm), with average fruit weight of 133 gm/fruit.The sex types in Cucurbitaceae are monoecious, and romonoecious, gynomonoecious or hermaphrodite. Work on inheritance of flowering types in melon (Cucumismelo L.) done by many researchers [2], [3], [9], [10], [16], [18], [19]. [20], [21][22] indicated that monoecy is dominant to and romonoecy. Dominance is controlled by a single pair of alleles giving monoecious, and romonoecious, gynomonoecious and hermaphrodite in ratios of 9:3:3:1 in F₂ of the cross between monoecious and hermaphrodite plants as we revealed in this study. In tibish, and romonoecious was dominant in F1 indicating that and romonoecy was dominant to monoecy. The ratio of 3 and romonoecious: 1 hermaphrodite was obtained in F2 generation indicating that hermaphroditic was recessive to and romonoecy by a single gene [7], [11]. Recent study indicated that dominance of and romonoecy in tibish may controlled by other two linked genes involved in loci of and romonoecy[4].Cultivar Paul was the hermaphrodite accession used as a source to transfer the hermaphroditic character to flexuosus and tibish. Production of hermaphroditic lines is possible through many selfing generations. Moreover, no difference effectsin source of pollen whether from male or hermaphrodite flowers [12]. Hermaphrodite flowers appeared earlier compared to female flowers in monoeciousor hermaphrodite in and romonoecious plants. Therefore, early production and high number of fruits could be obtained, because hermaphrodite flowers always bear in the main stem and lateral branches instead of the first two nodes of lateral branches as in monoecious or and romonoecious plants. Appearance of male flowers with rudimentary ovary (abnormal male) in F₂ and F₃ generations indicated that these generations probably influenced by environmental conditions or it has modifier genes. These abnormal male flowers produced fruits with irregular shapes [10],[11]. Tendency: Flowers on the main stem of hermaphrodite plants have often a longer peduncle than flowers on the lateral branches (Figure 1). Moreover, abnormal fruits observed on the main stem compared with branches?

Tibish was domesticated recently as a melon type [13], which was different from another intra-specific melon groups[14] and considered as a source of resistance to some major diseases of melon [15]. On the other side, the inheritance of sex expression in tibish was complicated because there are two linked genes were involved in the loci of and romonoecya/A[4]. Moreover, no pleiotropic effects of and romonoeciousgene ‘a’were expected on fruit shape because it will be in dominant form. More studies concerning stability of hermaphrodite and fruit quality are required.

4. Conclusions

Based on the results of this study, the production of hermaphrodite melon will be possible through successive selfing progenies of the crosses. The Hermaphrodite lines from these crosses exceeded their parents in characters of earliness and number of fruits per plant. Thus further research is needed to production of hermaphrodite melon.

References

1. Kirkbride, J. H. 1993. Biosystematics monograph of the genus Cucumis (Cucurbitaceae). Parkway Publishers,Boone, North Carolina, USA.
2. Rosa, J. T. 1928. The inheritance of flower types in Cucumisand Citrullus. Hilgardia 3: 233-250.
3. Poole C. F. and Grimball, P. C. 1939. Inheritance of new sex forms in Cucumismelo L. J Hered 30:21.
4. Abdelmohsin, M.E., ElJack, A.E., Yousif,M.T., Ahmed, E.A., andPitrat, M. 2012. Inheritance of andromonoecy in Tibish.Proceedingsof the X^th EUCARPIA meeting on genetics and breedingCucurbitaceae (eds. Sari, Solmaz and Aras) Antalya (Turkey), October, 15-18^th, 2012.
5. Rudich J, Halev,y A.H. and Kedar, N. 1969. Increase in femaleness of three cucurbits by treatment with Ethrel, an ethylene releasing compound.Planta,86:69–76
6. Stankovic L, Prodanovic S: Silver nitrate effects on sex expression in cucumber.ActaHorticult2002, 579:203-206.
7. Rowe, F.W.E. (1969). A review of the familieHolothuriidae (Holothuroidea: Aspidochirotida). Bulletin of the British Museum of Natural History (Zoology), 18(4): 119-170.
8. Poole C. F. and Grimball, P. C. 1939. Inheritance of new sex forms in Cucumismelo L. J Hered 30:21
9. Kenigsbuch, D. and Cohen, Y. 1987.Downy mildew- powdery mildew and fusarium wilt resistant in muskmelon breeding line PI 124111F. phytoparasitica, 15: 187- 195.
10. Kenigsbuch, D. and Cohen, Y. 1990.The inheritance of gynoecy in muskmelon. Genome 33:317–320.
11. Risser, G. and Rode, J. C. 1979.Iduction par le nitrate d’argent de fleursstaminees chez des plantesgynoïques de melon (Cucumismelo L.)Annules de l’Amelioration de plantes29 : 349) 352
12. Kouonon, L. C., Jacquemart, A. Bi, A. I. Z. Bertin, P. Baudoin, J. and Dje, Y. 2009. Reproductive biology of the andromonoeciousCucumismelosubsp. Agrestis (Cucurbitaceae). Annuls of Botany atwww.aob.oxfordjournals.org.
13. Pitrat, M., Chauvet, M. and Foury, C. 1999. Diversity, history, and production of cultivated cucurbits. ActaHorticulturae 492: 21–28
14. Mohamed, EI, and Pitrat, M. 1999. Tibish, a melon type from Sudan. Cucurbit Genetics Cooperative Report, 22:21–23.
15. Mohamed, E. I. and Yousif, M. T. 2004. Indigenous melons (Cucumismelo L.) in Sudan: A review of their genetic resources and prospects for use as sources of disease and insect resistance. Plant genetic Resources Newsletter 138: 36- 42.
16. Kubicki, B. 1969. Comparative studies on sex determination in cucumber(Cucumissativus L.) andmuskmelon (Cucumismelo L.).Genet Pol,10:167–183.
17. Steele, R. G. and Torrie, J. H. 1980. Principles and procedures of statistics: Abiometrical approach. (2^nded).McGraw-Hill, N Y. pp633.
18. Stepansky, A., Kovalski, I. and Perl-Treves, R. 1999. Intraspecific classification of melons (Cucumismelo L.) in view of their phenotypic and molecular variation. Plant Systematics & Evolution, 217: 313-333
19. Esquinas-Alcazar JT, Gulick, P. J. 1983. Genetic resources of Cucurbitaceae. A global report. IBPGR, Rome , Italy
20. Munger, H.M. and Robinson, R.W. 1991. Nomenclature of Cucumismelo L. Cucurbit Genetics Cooperative Report, 14:43 –44
21. Pitrat, M., Hanelt, P., Hammer, K. 2000. Some comments on interspecificclassification of cultivars of melon. ActaHorticulturae, 510:29–36.
22. Taha, M., Omara, K. and ElJack, A. 2003. Correlation among growth, yield and quality charactersin Cucumismelo L. CGC Rept. 26:9-11.