Crossability studies of inter-specific hybridization between varieties of Sweet orange(Citrus sinensis Osbeck)  and Mandarin (Citrus reticulata Blanco)

Introduction

Citrus is one of the most significant and predominant fruit crops grown worldwide, it belongs to family Rutaceae. Citrus fruits are good source of macro, micronutrients and antioxidants compounds (Ting 1980 and Liu et al 2012), it is has anticancer and anti-inflammatory properties (Ma et al 2020). In India, citrus covers an area of 10.5 lakh ha and is the third most important fruit crop with an annual production of about 14.0 million tons (Anonymous, 2020a). In Punjab, citrus is the main leading fruit among other fruit crops and mainly grown in arid irrigated and sub mountainous regions on an area of57289 hawith an annual production of 1.28 million tons (Anonymous 2020b). Among the citrus, most commercially important species grown in India are the mandarins (Citrus reticulata Blanco) followed by, sweet orange (Citrus sinensis Osbeck) and acid lime (Citrus aurantifolia Swingle). However, citriculture in North India is mainly confined to a single variety i.e. Kinnow mandarin with major commercial significance because no other recommended variety is at par in yield and quality for the region. Thus, there is need to increase the availability of new varieties with higher yield and improvedquality traits in the citrus industry. Till nowseveral commercial varieties have been developed through traditional breeding methods such as inter and intra-specific hybridization, mutagenesis, and clonal selection in citrus (Caruso et al 2020).However, sexual breeding in citrus is not always feasible due to several obstacles such as sexual incompatibility, polyembryony, sterility (Talon and Gmitter 2008). Sexual incompatibility  is common in citrus and cross incompatibilities betweenspecies pose a serious constrains in the inter-specific hybridization because of pre and post zygotic barriers. Barrett (1985) also stated that compatible crosses exhibit  post zygotic barriers such as low and non- viable seed yield, low recovery of hybrids from these crosses, high  mortality rate of  hybrid seedlings, poor field survival etc. Therefore, high cross compatibility between parents involved in crossing is important and it determines the success of controlled hybridization programme. It is utmost important to verify  the compatibility between the species involved in crossing. Inter-specific hybridization provides an opportunity to introgress traits from both the parents and recombines their genomes. Therefore, this investigation was conducted to develop new hybrids with improved horticultural traits such as fruit size, easy peel, juice yield and seedlessness in sweet orange. Further to gain more information on cross compatibility in inter-specific crosses.

Materials and methods

Parent material and hybridization

Three mandarin varieties viz., Mukaku Kishu, W.Murcott and Clementine (male parent) and  two sweet orange varieties viz., Mosambi and Jaffa (female parents) were used to generate hybrids. This investigation was carried out at Punjab Agricultural University, Regional Research Station, Abohar, Punjab India during 2019 and 2020. Varieties were planted at 6 x 6 m spacing. Anthers from unopened flowers from male parent were collected, allowed to dehisce on silica gel, and stored at 4⁰C until required. Emasculation of flower on female parent was done before opening of petal and immediately dusted with the camel hair brush. Controlled cross pollination was done on 17^th to 29^th March 2019 and 15^th to 30^th March 2020. In each cross, data were recorded on initial fruit set (%), fruit drop (%), fruit retention (%) seed content per fruits, fruit qualities in terms of Total soluable solids, acidity, juice content,  and were expressed in percentage.

Statistical Analysis:  The data were subjected to an analysis of variance (ANOVA) andSAS (Statistical Analysis of Variance) version 9.4 software. All data were performed  in triplicates and presented as means. Statistical significance was determined at p≤0.05.

Results and discussion

Fruit set and fruit drop percent

Fruit set percent in both the year (2019 and 2020) followed similar trend and significantly maximum mean fruit set  (49.75 %) was observed in Mosambi × Mukaku Kishucross  and minimum in W.Murcott (31.56 %) fruit set in Mosambi was maximum when Mukaku Kishuused as pollen parent

 (49.75 %) and minimum fruit set (39.15 %) was observed when crossed with W. Murcott. However, fruit set in Jaffa possess maximum fruit set (35.37 %) when W. Murcott used as pollen parent and exhibited minimum fruit set when Mukaku Kishu used as pollen parent. Result indicates that fruit set was influenced by different pollinators for both sweet orange varieties (Table 1). Overall, Jaffa × Clementine differed significantly from all other crosses. Similar variation in fruit set percentage were reported by Hossein and Rabbani (2011) in lemon genotypes which ranged between (20 to 56%) and also influenced the seed setting in different crosses.  Our results are in accord with the findings of Chao (2005) investigated the  effect of different pollinators on fruit set in mandarins, and reported that cross combination between Nules × Afourer mandarins observed maximum 39.62 % fruit set in 2002 and in similar cross very low fruit set of 4.92 % was recorded in 2003. It was reported that due to low temperature in 2003, the flowers in Nules Clementine and  Afourer could not open properly for crossing therefore resulted in low fruit set. Atawia et al (2016) showed that Shaddock (Citrus maxima L.) pummelo, had maximum fruit set 51.17 % when Balady orange  used as pollen parent followed by 48.30 % when Succary orange was used as pollen parent and 47.00 % fruit set was observed when Balady mandarin was used as pollen parent. While minimum fruit set 37.00 % was observed when Marsh grapefruit was used as pollen parent.Singh et al (2017) reported maximum mean fruit set percentage (15.49) in W. Murcott x Kinnow cross. Jahromi et al (2019) also obtained variation and recorded highest percentage of fruit set (73.75 %) in Clementine when crossed with sweet lime and fruit set was minimum (69.25%) when Lisbon lemon used as pollen parent. All these reports indicated to suggest that percent fruit set depend on the genotype and environment.

            In citrus, two successive waves of fruit drop occurs viz., June drop and Pre- harvest drop. The results revealed that overall maximum (91.64 %) average fruit drop in June was recorded in Jaffa × Mukaku Kishu, and minimum in Mosambi × Mukaku Kishu(77.70%) (Table 1). Similar trend was followed for both the year of investigation in Mosambi.However in case of Jaffa, different trend was observed during both the years. In 2019, June drop was recorded maximum (92.94 %) particularly in Jaffa ×   Kishuand minimum in Jaffa × W.Murcott (83.38 %) and in 2020 minimum (80.35 %) was observed in Jaffa × Clementine. Mosambi had maximum June drop when Clementine used as pollen parent and had minimum fruit drop when Mukaku Kishu was used as pollen parent. Whereas, Jaffa had maximum fruit drop when Kishu was used as pollen parent and minimum fruit drop was observed when Clementine used as pollen parent. This may be due to the fact that citrus possess an internal self-regulatory adjusting mechanism that adjusts fruit loads to enable supply of metabolites to the trees (Goldschmidt and Monselise 1977). June drop in citrus may also occurs due to depletion of carbohydrates caused by growing fruitlets leading to rise in ABA and ethylene content ultimately leading to abscission of fruit (Domingo et al 2007 and Gomez-Cadenas et al  2000).Present results were in concord with the findings of Singh et al (2017), who also observed variation and maximum June drop percent of 40.21 % in Daisy x Kinnow cross and  minimum fruit drop of 31.87 % in Kinnow x W.Murcott.

The effect of pollinators on the pre-harvest fruit drop percent in different inter-specific crosses are shown in (Table 1). The result revealed that maximum pre harvest fruit drop was recorded in Mosambi when W. Murcott pollen was used and minimum fruit drop percentage was observed when Clementine used as pollen parent. Whereas, in Jaffa maximum fruit drop was observed when Mukaku Kishu used as pollen parent and minimum fruit drop was observed when W.Murcott  used as pollen parent and minimum fruit drop was observed when W. Murcott was used as pollen parent. The mean pre-harvest drop was maximum (23.66 %) in Jaffa × Mukaku Kishu and minimum in Mosambi × Clementine (16.52 %). However, maximum fruit drop percentage pattern was different for both the year of investigation. In 2019, maximum (24.82 %) pre harvest fruit drop percentage was recorded in Jaffa ×Mukaku Kishu cross and in 2020 pre harvest fruit drop was maximum (27.77 %) in Mosambi × W. Murcott cross. Jawanda et al (1972) also reported varying percentage of fruit drop in Mosambi (57.2), Pineapple (47.6), Hamlin (48.5), Jaffa, Blood Red (48.6) and Valencia (34.5). The result of present investigation coincides with the findings of Singh et al (2017), who recorded minimum fruit drop (31.87%) in Kinnow x W. Murcott cross and maximum in  Daisy x Kinnow cross (40.21) fruit drop.

Fruit retention percent

Fruit retention percentageon fruit set basis was significantly influenced by pollinators (Table 2). Fruit retention in Mosambi and Jaffa was influenced differently by different pollen parents and maximum retention (22.15 %) was recorded with Mukaku Kishu and minimum (8.66 %) with Clementine pollens in case of Mosambi whereas, in case of Jaffa maximum fruit set (18.11 %) was recorded with W. Murcott pollens and minimum fruit set (6.51 %) was recorded with Mukaku Kishu same trend was followed during both the years (2019 and 2020). Overall fruit retention was recorded higher in Mosambi as compared to Jaffa.

Fruit retention per cross

On the basis of number of crosses made, overall percent fruit retention was ranged in between 2.08 to 9.20 and individually in Mosambi it ranged in between 3.96 to 9.20 and in Jaffa it ranged in between 2.08 to 4.02 (Table 2).Similar trend was followed during both the year of investigation (2019 and 2020). Results also indicated that highest fruit retention per cross in Mosambi was observed when crossed with Mukaku Kishu pollens and lowest was recorded with Clementine pollens. Whereas, in Jaffa fruit retention per cross was highest when cross with Clementine and lowest fruit retention was observed when Mukaku Kishu used as pollen parent. The less fruit retention may be due to the fact that, generally in sweet orange cultivars only less than 1 % fruits reach to maturity from 2.5 lakh flowers per plant (Domingo et al 2007). And fruit retention percent depend on the compatibility between parents involve in crossing as well as on environmental conditions. Similarly, Hossein and Rabbani (2011) also reported higher fruit set percent (20 to 56 %), retention and seed setting in cross-pollination of lemon genotypes.

Seed characters

In general average number of normal seeds recorded higher in Mosambi as compare to Jaffa with all the three male parents and maximum mean number of normal seeds per fruit (18.08) were observedinMosambi × W. Murcott and  minimum in Mosambi × Mukaku Kishu (14.58) (Table 3). Similarly, Jaffa also exhibit significant maximum normal seed with W.Murcott and minimum 7.13 with Mukaku Kishu. The trend in both the year followed the same pattern.  Similarly, Singh et al (2017) evaluated number of seeds per fruit obtained from different intra-specific crosses in mandarin and reported variation among different crosses and maximum normal seeds per fruit (21.21) were recorded  in fruits from Kinnow x Daisy cross which was at par with mean normal seed number fruit fruit (18.01) in Kinnow x W. Murcott cross. And significantly lower mean normal seed number per fruit (12.48) was recorded in Daisy x Kinnow cross. The data on number of rudimentary seeds per fruit clearly indicated significant difference in different inter-specific cross. The maximum mean rudimentary seeds per fruit (7.05) were recorded in cross involvingMosambi × Mukaku Kishu and minimum in Jaffa × W.Murcott (3.11) cross. Similar trend was followed for two consecutive year of investigation (2019 and 2020). 

Studies showed different degree of compatibility among different crosses, and some combinations recorded significantly higher seed number (Table 3). Results showed a significant differences  in total seed per fruit. The maximum mean number of total seed per fruit were recorded (24.20) inMosambi × W. Murcott cross and minimum in Jaffa × Mukaku Kishu (10.79). Similar trend for total number of seeds per fruit was observed  in both the year of investigation (2019 and 2020).  Result also revealed that maximum seeds per fruit in Mosambi and Jaffa were observed when crossed with W. Murcott as male parent. Results suggest that high compatibility of Mosambi and Jaffa with W. Murcott. Similarly Vithanage (1991), Ellendale tangor gave the largest and multiseeded fruits when cross pollinated with Murcott and Emperor. Similarly, Abknar (1996)  also proved the differential effect of different pollinator and observed that Moro blood orange is the best pollinator in Clementine mandarin for reducing the number of seeds, whereas clustered lemon is the best pollinator for increasing percent fruit set in Clementine mandarin. Similarly, Chao (2005) also observed different number of seeds per fruit with different pollen parents in Mandarins crosses. And reported that cross combination between Nules × Afourer mandarins observed maximum 25.36 % seeds per fruit and cross between Nules × Taho Gold recorded minimum seeds per fruit (1.71). While Afourer × Nules cross observed maximum seeds per fruit (12.00) and minimum seeds per fruit (10.77) was observed in Afourer × Nules cross. Our results are in agreement with the findings of Atawia et al (2016), who obtained different number of seeds per fruit with different  pollinizers in Shaddock (Citrus maxima L.) pummelo. Maximum seeds per fruit (110.33) was observed when crossed with  Balady mandarin followed by (97.85) with Balady orange, (79.50) and minimum seeds per fruit was obtained with Marsh grapefruit (62.34). Yildiz and Kaplankiran (2017) also found different effects of pollinators on number of seeds per fruit in Robinson mandarins, and maximum seed per fruit (27.81)was observed in Robinson mandarin × Valencia Late orange and minimum seeds per fruit (8.18) was observed in Robinson mandarin × Midknight Valencia.

Physical characters of fruit

Maximum mean fruit length (68.87 mm, diameter (67.55 mm) and weight (167.50 g) was observed in Mosambi when crossed with W.Murcott at par with Clementine and minimum was noticed when crossed with Mukaku Kishu during both the year. Similar, trend was also observed in case of Jaffa. In general fruit size and weight of Jaffa was lesser than the Mosambiwith all the crosses.Our results are in agreement with the findings of Atawia et al (2016), who obtained different fruit size with different pollinizers in Shaddock (Citrus maxima L.) pummelo. Maximum fruit length (13.47cm) was observed when crossed with Balady mandarin followed by (12.93cm) with Marsh grapefruit and minimum fruit length (11.30 cm) was obtained with Balady orange. Yildiz and Kaplankiran (2017) also found different effects of pollinators on fruit weight  in Robinson mandarins, and maximum fruit weight (137.49g) was observed in Robinson mandarin × Rhode Red Valencia and minimum fruit weight  (116.49 g) was observed in Robinson mandarin × Midknight Valencia. No significant difference innumber of segments per fruit was observed in Mosambi and Jaffa during both the year of investigation (Table 4 and Fig 1).

            A significant result was recorded in different crosses and maximum mean peel weight (42.72 g) and peel thickness (3.00mm) and peel percent (22.20 %) was recorded in Mosambi when cross with W. Murcott which was statistically at par with Clementine and minimum was recorded when crossed with Mukaku Kishu (Table 5). Similar, trend was also observed in case of Jaffa. The variation in peel weight and peel thickness in all the crosses for two season could be due to environment and genotype interaction.

Quality characters

A non significant difference in Mosambi and Jaffa was observed due to pollinators on fruit quality characteristics (Table 6). However, the maximum mean TSS (9.65 ⁰ B) was recorded in Mosambi × Clementine cross and minimum in Jaffa × W.Murcott (8.65⁰ B). Similar trend was followed in both the year of investigation (2019 and 2020). Results discussed above is supported by the findings of Kitat et al (1994) and Wallace et al (2002).  Maximum mean acidity percent (0.60 %) was recorded in  Jaffa × Mukaku Kishu and minimum in Mosambi × Mukaku Kishu(0.53 %). Result indicated that type of pollen parent had no effect on acidity percent of fruits in both the varieties. Likewise, Papadakis et al (2009) reported that cross pollination of Nova with either SRA63 or Marisol had no effect on fruit quality characters such as total acids in the juice.

A significant variation in juice percent was recorded in fruits obtained from different inter-specific crosses. Maximum mean juice (46.80 %) was noticed in Mosambi × Mukaku Kishu which was statistically at par with Jaffa × Mukaku Kishu (45.28 %), followed by Mosambi × Clementine (42.68 %) and Jaffa × Clementine (42.40 %) followed by Mosambi × W.Murcott (41.93 %) and minimum in Jaffa × W.Murcott (40.87 %). Whereas, maximum mean rag (41.45 %) was recorded in Jaffa × Clementine followed by Jaffa × Mukaku Kishu (38.05%) and minimum in Mosambi × Mukaku Kishu(33.35 %). Similar trend in rag percent was recorded in all the crosses during both the years. Similarly, Yildiz and Kaplankiran (2017) also found different effects of pollinators on number of juice content in fruits of Fremont mandarins and maximum  juice percent (51.60 %) was observed in Fremont × Rhode Red Valencia and minimum juice percent (48.01 %) was observed when cross with Midknight Valencia.

Conclusions

High sexual compatibility in inter-specific crosses of sweet orange and mandarin was observedusing former as maternal parent.Wide variation in terms of fruit set and seed content was observed in all the crosses.Highest fruit set was observed  in Mosambi when crossed with Mukaku Kishu whereas, in Jaffa, highest fruit set was recorded when crossed with W.Murcott. June drop wasrecorded minimum in Mosambi with Mukaku Kishu whereas, in Jaffa, fruit drop  was minimum with Clementine used as pollen parent. Maximum mean fruit retention was observed in Mosambi withMukaku Kishu whereas with Jaffa it resulted into minimum retention. Maximum total seeds per fruit inMosambi was observed with W.Murcottcross. Therefore, Mukaku Kishu and W.Murcott mandarins could be use as suitable pollenizer for Mosambi and Jaffa, respectively. Because of highest fruit set and fruit retention and more fruit weight.  However, W.Murcott resulted in higher number of seeds per fruit, thus reduced the quality of fruits. Result indicatedthat Mukaku Kishu was sexually incompatible with Jaffa and had very low fruit setand fruit retention percent. The valuable information gained from this study may help the citrus breeders to select suitable cross combinations and will contribute in enhancement of genetic base, which may play a major role in scion breeding programme of sweet orange.

Funding: No external funding was received to conduct this research study.

Acknowledgments: The authors are grateful to the Dr J CBakhshi, Regional Research Station, Punjab Agricultural University, Abohar, India for providing material  to carry out thisResearch study.

Conflicts of Interest: The authors declare no conflict of interest.

Copyright:

a) the work described has not been published before (except in the form of an abstract or as part of thesis)

b) this manuscript is not under consideration for publication elsewhere;

c) when the manuscript is accepted for publication, the authors agree to automatic transfer of the copyright to the Electronic Journal of Plant Breeding and Indian Society of Plant Breeders.

1. References
2. Abknar, A., 1996. Effect of pollen of different cultivars of Citrus on somequantitative and qualitative of Clementine mandarin. Master’s Degree Thesis inHorticulture. School of Agriculture. University of Tehran, Iran.
3. Anonymous, 2020a. FAOSTAT.  Food and agriculture organization corporate statistical database. http://faostat.fao.org/site/567/default.
4. Anonymous,2020b. Indian Horticulture Database. National Horticulture Board, Gurgaon (Haryana), New Delhi. http://www.nhb.gov.in.
5. Atawia,  A. R., 2016. Abd EL-Latif  F M, E L-Badawy H ,  Abo-Aziz A B, Abou Rayya  M S M,  Baiea M H M, Abdelkhalek A Effect of various pollination treatments on yield  characteristics and fruit quality of shaddock fruits. Int J Sci Eng Res 7, 1–9.
6. Barrett, H. C. 1985. Hybridization of Citrus and related genera. Fruit Var J  39, 11-16.
7. Caruso,  M., Smith, M. W., Froelicher, Y., Russo, G., and Gmitter, F. G., 2020. Traditional breeding, in The Genus Citrus, 1st edition. Eds. M. Talon, M. Caruso and F. G. Gmitter (Cambridge, UK: Elsevier), 129–148.
8. Chao, C., 2005 Pollination study of the mandarins and the effect on seediness and fruit size: implications for seedless mandarin production.
9. Domingo, J. Iglesias, Manuel, Cercos, J. M., 2007.Colmenero-Flores, Miguel A. Naranjo, Gabino Ríos, Esther Carrera, Omar Ruiz-Rivero, Ignacio Lliso, Raphael Morillon, Francisco R. Tadeo and Manue Physiology of citrus fruiting. Talon. Braz. J. Plant Physiol., 19, 333-362.
10. Domingues, Eta, and Tulmann, N. A., 1999. Influence of pollination and floral morphology on fruit setting in sweet orange varieties. Scientia Agricola, 56, 163-170.
11. Goldschmidt, E. E., Monselise, S. P. 1977. Physiological assumptions toward the development of a citrus fruiting model. Proc. Int. Soc. Citrus 2, 668-672.
12. Gomez-Cadenas, A., Mehouachi, J, Tadeo, F. R, Primo-Millo, E, Talon, M., 2000. Hormonal regulation of fruitlet abscission induced by carbohydrate shortage in citrus. Planta 210, 636-643.
13. Hossein, MI. and MG. Rabbani. 2011. Study on the cross compatibility of some lemon genotypes (Citrus limon L.). Bangladesh Journal of Agricultural Resources, 36:241-246.
14. Jahromia, H. A., Zareib, A., and Mohammadkhania, A., 2019. Analysis the effects of pollen grain sources on the fruits set and their characteristics of ‘Clementine’ mandarin using microscopic and molecular approaches Scientia Horticulturae 249,  347–354.
15. Liu, Y., Heying, E., and Tanumihardjo,  S. A., 2012. History, global distribution, and nutritional importance of citrus fruits. Compr Rev Food Sci  Food Saf 11, 530–545.
16. Ma, G., Zhang, L., Sugiura, M., and Kato, M., 2020. Citrus and health, in The Genus Citrus, 1st edition. Eds. M.,. Talon, M. Caruso and F. G. Gmitter (Cambridge, UK: Elsevier), 495–511.
17. Papadakis,  I. E., Protopapadakis,  E. E., Therios, I. N., 2009. Yield and fruit quality of ‘Nova’ hybrid [Citrus clementina hort. ex Tanaka (Citrus reticulata Blanco C. paradisi Macfad)] and two Clementine varieties (C. clementina hort. ex Tanaka) as affected by self- and cross-pollination. Scientia Horticulturae, 121, 38–41.
18. Singh, G., Aulakh,  P. S. and Rattanpal, H. S.  2017. Fruit setting patterns and germination of F1 seeds in some intra-specific crosses of mandarin (Citrus reticulata blanco) .Applied Biological Research 19, 264-269.
19. Soost, R. K., and Cameron, J. W., 1975. Citrus. In Advances in Fruit Breeding (J. Janick and J.N. Moore, eds.). Purdue Univ. Press. West Lafayette, Indiana. pp. 507-540.
20. Sweety, G. S., Rana, and Reddy,  G. C. 2018. Effect of growth regulators on yield and quality of sweet orange  (Citrus sinensis L. Osbeck.) cv. Jaffa. Int J Curr Microbiol App Sci, 2963-2967.
21. Talon, M., and Gmitter, F. G. 2008. Citrus genomics. Int J  Plant Genomics 1– 17.
22. Ting, S. V. 1980.  Nutrients and Nutrition of Citrus Fruits, in Citrus nutrition and quality. Eds. S. NagyJohn and A. Attaway (Washington, DC. USA: American chemical society), 3-24.
23. Yildiz, and Kaplankiran,2017. The Effect of Cross-Pollination on fruit Set and quality in ‘Robinson’ and ‘Fremont’ Mandarins. Ege Univ. Ziraat Fak. Derg., 54,107-112.