Introduction
Pummelo (Citrus grandis L. Osbeck) is one of the important tropical fruit crops under the genus Citrus and is considered to be one of the three true citrus species along with C. reticulata and C. medica (Federici et al., Reference Federici, Fang, Scora and Roose1998). Its role in citrus breeding highlights the importance of preserving and utilizing ancient cultivars to meet contemporary agricultural challenges (Nicolosi et al., Reference Nicolosi, Deng, Gentile, La Malfa, Continella and Tribulato2000; Barkley et al., Reference Barkley, Roose, Krueger and Federici2006). It is the largest citrus fruit in the Rutaceae family (2n = 18) and is the ancestor of numerous citrus fruits, including tangelos and grapefruit (Scora, Reference Scora1975; Velasco and Licciardello, Reference Velasco and Licciardello2014). Citrus fruits are an integral part of the human diet, renowned for being free from sodium and cholesterol while offering substantial amounts of vitamin C, folic acid, potassium, flavonoids, coumarins, pectin and dietary fibres (Singh et al., Reference Singh, Singh, Singh, Parthasarathy and Vinoth2015b). In India, pummelo is known by various vernacular names such as chakotra in South India, chakkota in Karnataka, pampanas in Maharashtra and mahtabi in North India.
Pummelo can be easily identified due to its distinctive morphology, such as large leaves with winged petioles, highly fragrant flowers and bigger-sized fruit. Pummelo fruit is now gaining popularity in India due to its high nutritional value, antioxidant properties, medicinal uses and increased tolerance to insects and diseases as compared to other citrus fruits (Mokbel and Hashinaga, Reference Mokbel and Hashinaga2006). Fresh fruit can be consumed directly or else it can be used in a variety of recipes to make jam, jelly, cakes and beverages. Pummelo has a thick rind which makes it simple to handle and there is potential for its development as an export crop in India (Wen et al., Reference Wen, Cai, Wang, Tan and Lan2010). Some local types have a thin rind that is simple to peel which suits the domestic market. In Vietnam, pummelo flowers are used for perfume, while the wood serves as tool handles and firewood. The leaves, flowers, fruits and seeds are also used in herbal medicine for cough, fever and gastrointestinal disorders (Hossain et al., Reference Hossain, Disha and Rahim2018).
Pummelo is believed to have originated in Southeast Asia (Scora, Reference Scora1975). It is grown as a backyard garden crop all over India, whereas the maximum diversity was found in the North-East region, Bihar and Bengal (Engels, Reference Engels2001; Malik et al., Reference Malik, Chaudhury, Dhariwal and Kalia2006; Hazarika et al., Reference Hazarika, Lalthlamuani, Lalchhanmawia, Lalrinfeli and Nautiyal2016). Devanahalli pummelo, cultivated near Bengaluru's Kempegowda International Airport, is renowned for its exceptional quality. It was awarded Geographical Indication (GI) status in 2009 and is valued for its distinct taste, large size, thick rind and juicy pink to red flesh with a sweet, mildly tangy flavour and pleasant aroma (Singh et al., Reference Singh, Singh, Singh, Parthasarathy and Vinoth2015b). Traditionally grown by the ‘Thigala’ community, only a few of these valuable plants remain today, preserved by a handful of progressive farmers in the Devanahalli region (Anon., 2009).
Pummelo is propagated through seeds or by vegetative methods (Chomchalow et al., Reference Chomchalow, Chapman and Griffee2004; Koti et al., Reference Koti, Swamy, Rao, Jayappa and Ugalat2023). It being monoembryonic and self-incompatible, lacks nucellar seedlings like other citrus fruits (Uzun and Yesiloglu, Reference Uzun and Yesiloglu2012). Seedlings are primarily of zygotic origin and exhibit wide variation due to natural hybridization and outcrossing, resulting in differing quality across various regions in India. Significant disparities exist in their physical characteristics and the genetic variation of this fruit has not yet been fully documented. Therefore, clonal selection is a key method for genetic improvement and the selection of new genotypes (Singh et al., Reference Singh, Singh, Nath, Parthasarathy, Sthapit and Vinoth2015a).
The Devanahalli pummelo is on the verge of extinction due to urbanization and changes in cultivation practices (Malik et al., Reference Malik, Kumar, Singh, Dhariwal and Chaudhury2013), resulting in a decrease in genotypic variation and a decline in overall variability as unwanted phenotypic characteristics are eliminated (Wen et al., Reference Wen, Cai, Wang, Tan and Lan2010; Kaur et al., Reference Kaur, Malik, Choudhary, Rohini, Choudhury and Kumar2019). This situation underscores the urgent need to identify high-yielding, superior-quality pummelo genotypes. Characterizing Citrus genetic resources is essential for assessing and quantifying existing diversity within species and cultivars, analysing their relationships and facilitating efficient germplasm management and also improves the production as well as yield of the country. Surveying is an effective method for selecting these genotypes. Estimating genetic diversity aids in identifying key genotypes, monitoring genetic erosion, selecting species for breeding and developing in-situ and ex-situ conservation strategies (Kaur et al., Reference Kaur, Malik, Choudhary, Rohini, Choudhury and Kumar2019). Despite pummelo having vast diversity in the Devanahalli region, farmers are not receiving high-quality planting materials or fruits. Therefore, a survey was conducted to identify elite genotypes in this region. The identified superior genotypes will be provided to the farming community to enhance the cultivation and yield of trees. Further, these can be utilized in future citrus breeding programmes for improvement.
With this in mind, the present study was conducted to evaluate and characterize 40 pummelo genotypes grown in homesteads across Devanahalli and the Bengaluru Rural region. This research aimed to: (1) investigate the diversity of pummelo in the Devanahalli and Bengaluru Rural regions using morphological, biochemical and sensory evaluation; (2) examine the correlation among various morphological and biochemical traits; and (3) determine suitable genotypes for fresh consumption, nutraceutical and industrial applications. The findings of this study will contribute to improving pummelo production by identifying the most suitable genotypes and informing conservation strategies.
Materials and methods
Experimental material
The survey was carried out in Devanahalli and Bengaluru Rural districts of Karnataka, India to know the extent of diversity in the naturally existing pummelo population and identify elite genotypes with horticultural traits. The pummelo trees are found growing in home gardens, abandoned lands as well as in cultivable fields in the Devanahalli region. The preliminary survey and identification of pummelo genotypes were done randomly and after the initial survey, 40 elite promising genotypes were selected for the study based on their growth, flowering behaviour, fruiting behaviour and attractiveness from June 2021 to November 2022 (S4).
Morphological characterization
In this study, the morphological characteristics of three fruits from each surveyed pummelo genotype were analysed and observations were recorded for 11 quantitative traits of fruit following the guidelines of the International Plant Genetic Resources Institute (IPGRI, 1999) for testing citrus plants for distinctiveness, uniformity and stability. Measurements of the fruit's physical characteristics were conducted according to standards. The weight of the fruit, pulp and seeds was measured using an electronic digital balance (CWS 6KE-scale tech [0.1 g to 3 kg]) and fruit volume by the water displacement method (cm3).
Biochemical characterization
For biochemical characterization, fully mature fruits from 40 selected genotypes were harvested at the proper stage. Biochemical analysis for 10 quality traits was performed with three replicates for each genotype. Initially, the fruit pulp samples were extracted to measure several biochemical parameters. Total soluble solids (TSS) was estimated by using a hand refractometer (RHB-32ATC-Brix refractometer). The ascorbic acid content of fruit pulp was estimated using the 2, 6-dichlorophenol indophenol titration method given by Rao and Deshpande (Reference Rao and Deshpande2006). The titrable acidity, reducing and non-reducing sugars were measured following the standard method (AOAC, 2006).
The total sugar content was determined by following the procedure of Sadasivam and Manickam (Reference Sadasivam and Manickam1992). The Folin–Ciocalteu assay is used to estimate total phenolic content (TPC) using the spectrophotometric method (Sun et al., Reference Sun, Powers and Tang2007). The total antioxidant activity was estimated by using the FRAP (Ferric Reducing Antioxidant Power) method (Benzie and Strain, Reference Benzie and Strain1996). Total flavonoid content (TFC) was estimated as per the method of Singleton et al. (Reference Singleton, Orthofer and Lamuela-Raventos1999) using catechin as a standard. For consumer preferences and market demands the sensory evaluation was conducted to assess texture, colour, flavour, taste and overall acceptability on a 9-point hedonic scale following the guidelines established by Jellinick (Reference Jellinick1985).
Statistical analysis
The analysis of biometrical and biochemical traits of pummelo genotypes was conducted using R software version 4.0.4 (R Core Team, 2024). Variability among traits was assessed through phenotypic and genotypic coefficients of variation (GCV and PCV), while Spearman correlation coefficients evaluated the relationships between traits. Principal component analysis (PCA) was employed to investigate the relationships among genotypes, using standardized principal component (PC) scores derived from a correlation matrix based on mean values. Before performing cluster analysis, each trait was normalized using Z-scores to ensure consistent scaling. Following this, hierarchical cluster analysis was performed using the ‘hclust’ function and Ward's method to group genotypes according to their normalized traits (R Core Team, 2024).
Results
Morphological and biochemical characterization
The per se performance of the fruit traits of surveyed 40 pummelo genotypes is presented in online Supplementary Table S1. The estimates of variability for different metric attributes, i.e. range, coefficient of variation, heritability, genetic advance and genetic mean are represented in Table 1. The fruit length, diameter and diameter of the axis ranged from 9.45 to 18.87 cm, 8.70 to 17.38 cm and 0.69 to 5.32 cm respectively, with the highest fruit length and diameter in genotype HSR-1. Similarly, fruit volume ranged from 417.33 to 3528.33 ml with the highest value observed in genotype DNM-1 and fruit weight ranged from 344.03 to 1746.23 g highest value observed in DNM-1. Pulp weight and peel weight ranged from 194.39 to 1073.07 g and 133.41 to 1109.21 g respectively with maximum pulp weight observed in DSR-2 and peel weight in CHES-1. Total seed weight ranged from 0.33 to 29.99 g and the lowest seed weight was observed in genotype DNM-1. The number of segments and seeds per fruit ranged from 8.67 to 17.00 and 5.00 to 91.67, respectively, and the maximum number of segments observed in genotype DNM-1 and seeds per fruit in genotype DSP-8. Rind thickness ranged from 0.60 to 3.49 cm with the lowest thickness in genotype DVP-4.
Table 1. Estimate of variability for different metric attributes in 40 pummelo genotypes
![](https://static.cambridge.org/binary/version/id/urn:cambridge.org:id:binary:20250107180858650-0634:S1479262124000674:S1479262124000674_tab1.png?pub-status=live)
TSS serve as a key indicator of fruit taste and quality, ranging from 7.63 to 11.07°Brix, with the highest value observed in genotype DSP-17. The acidity of the fruit, which also influences consumer acceptability, is generally preferred to be lower in genotypes. Significant variation in acidity (0.73–2.07%) was observed among the genotypes, with the lowest value recorded in BKN-1. Genotypes DVP-1 and DVP-2 recorded good taste, as expressed by the TSS and TA ratio (9.59 and 9.72, respectively), while few of the genotypes were unpalatable, with values lower than 6. Total sugar and reducing sugar content varied from 2.47 to 7.46% and 1.18 to 4.00%, respectively, with the highest total sugars observed in genotype DAN-1 and reducing sugar in genotype RHREC-1.
TPC, TFC and antioxidant activity showed significant variation among the genotypes. The TFC ranged from 39.26 to 137.78 mg CE/g FW with the highest TFC observed in DSP-6. While the TPC ranged from 55.08 to 166.17 mg GAE/100 g FW with the highest value observed in DSP-16. The antioxidant activity was measured by FRAP assay (mg AEAC/100 g), it varied between 1.22 (RHREC-2) and 7.06 (DSP-16).
Phenotypic and genotypic coefficient of variation
Variability plays a significant role in selecting desirable genotypes, which is assessed by determining the GCV and PCV (Table 1). Respective values of the higher PCV and GCV were estimated for total seed weight, peel weight, the diameter of the axis, fruit volume, the total number of seeds per fruit, pulp weight, fruit weight, total antioxidant activity, non-reducing sugars, TFC, rind thickness, TPC, reducing sugars, total sugar, titratable acidity and ascorbic acid content. While other traits such as fruit length, fruit diameter, TSS/TA and the number of segments exhibited moderate range and TSS showed a low level of PCV and GCV, respectively.
Heritability estimates in a broad sense and genetic gain
The heritability and genetic gain percentages for various fruit traits are presented in Table 1. The heritability of different characters varied from 90.20 to 99.70% which indicates very high heritability and genetic advance ranged from 0.49 to 1368.54. However, the genetic gains were noted as low (16.08%) to high (137.77%). The analysis reveals that traits such as TPC, TFC, ascorbic acid content and TSS exhibit extremely high heritability and genetic gain, indicating strong genetic control and high potential for improvement through selection. Traits such as total antioxidant activity, the diameter of the axis, fruit volume and fruit weight also have high heritability but slightly lower genetic gains, suggesting that while they can be effectively improved, it may be at a slower rate. On the other hand, traits such as rind thickness, the total number of seeds/fruit, pulp weight, peel weight and total sugar content display high heritability but moderate genetic gains indicating a more controlled potential for improvement. Finally, TSS/TA and the number of segments have the lowest heritability and genetic gains, suggesting these traits are less influenced by genetics and may require more complex breeding strategies for significant enhancement.
Correlation analysis
The correlation analysis revealed several significant associations between morphological and biochemical traits (Fig. 1). Notably fruit weight was positively associated with pulp weight (0.86), peel weight (0.83), total number of seeds (0.45), TSS (0.45) and reducing sugar (0.43), whereas negative associations with ascorbic acid content and TPC. Pulp weight displayed positive correlations with all fruit size-related traits as well as TSS (0.46). Conversely, negative associations were observed with TPC (−0.45) and TFC (−0.42).
![](https://static.cambridge.org/binary/version/id/urn:cambridge.org:id:binary:20250107180858650-0634:S1479262124000674:S1479262124000674_fig1.png?pub-status=live)
Figure 1. Map of linear correlations between morphological and biochemical variables. The size and colour intensity of the circles indicate the magnitude of the correlation.
TSS was positively correlated with fruit weight (0.45), pulp weight (0.46) and reducing sugar (0.37), while negatively correlated with ascorbic acid content (−0.44). TSS/TA ratio was weakly correlated with TSS (0.32) and negatively correlated with titrable acidity (−0.78) and non-reducing sugar (−0.50). Vitamin C content was positively correlated with TPC (0.62) and TFC (0.62), whereas negatively correlated with fruit length (−0.42). Total sugar showed positive associations with titratable acidity (0.72). TPC (0.66) and TFC (0.59) were positively correlated with their antioxidant capacity.
Principal component analysis and clustering
PCA was conducted on 21 traits of the 40 selected pummelo genotypes to enhance the interpretability of the quantitative data. The results presented in online Supplementary Table S2 represent the eigenvectors for the principal components PC1–PC21. The results revealed that the first nine coordinates largely accounted for the variation among the genotype (92.05%) and among these first four coordinates had higher variation (70.99%).
The first principal component (PC1) accounted for 30.07% of the total variation and explained most of the variation, showing high loading values for several key traits such as fruit weight (0.37), fruit volume (0.36), pulp weight (0.35), fruit length (0.31), pulp weight (0.27) and fruit diameter (0.27). PC2 explained 18.42% of the total variation and accounted for the variation in TSS/TA (0.37), total number of seeds per fruit (0.29), seed weight (0.27) and number of segments (0.25). PC3 (14.50%) mainly represented TSS/TA (0.19), fruit diameter (0.06) and pulp weight (0.06) while by PC4 (8.00%) variation had high loading values for TSS (0.46), seed weight (0.28), number of seeds per fruit (0.25) and reducing sugar (0.24) (online Supplementary Table S3).
To compare genotypes across a range of traits and identify the most suitable genotype for the breeding programme, a biplot analysis of the first two principal components (PC1 and PC2) was performed using R (Fig. 2). This analysis projected the variables onto the factor planes, highlighting the diversity among individual genotypes. For example, the genotypes DVP-1, COHB-2, HSR-1 and DNM-1 had higher values for commercially important traits such as fruit length, diameter, volume, weight, pulp weight and TSS traits in the first quadrant of the biplot (+, +). The genotypes CHES-1, DAK-1, HSR-5 and HSR-2 had higher values for the sugar-related traits such as reducing sugar and total sugars in the second quadrant (−, +) and the fourth quadrant has traits such as TPC, TFC and vitamin C content in the pulp which included the genotypes DSP-4, DSP-5, DSP-9, DSP-17 and GKVK-1.
![](https://static.cambridge.org/binary/version/id/urn:cambridge.org:id:binary:20250107180858650-0634:S1479262124000674:S1479262124000674_fig2.png?pub-status=live)
Figure 2. Two-dimensional biplot for PC1 and PC2 (explaining 48.5% of the total variance) based on morphological and biochemical characteristics of 40 pummelo genotypes.
By using the morpho-biochemical data a dendrogram was generated through the hclust function in R and data revealed that 40 pummelo genotypes could be divided into two distinct groups: group I and group II (Fig. 3). Group I is mainly composed of genotypes which were collected from the Devanahalli taluka and Soppahalli Horticulture farm while group II is composed of both the Bengaluru Rural and Devanahalli regions. This dendrogram effectively illustrated the variability of the analysed characters among the genotypes. Notably, it indicated that genotypes DSP-4, DSP-5, DSP-6, DSP-8 and DSP-12 which had higher amounts of vitamin C, TPC and TFC were clustered in group I. The genotypes, such as DVP-1, HSR-1, BKN-1 and DNM-1 are found to have higher fruit morphological traits such as fruit length, diameter, weight, pulp weight, peel weight, number of seeds and TSS were clustered in group II. These findings aligned with the results obtained from the PCA analysis.
![](https://static.cambridge.org/binary/version/id/urn:cambridge.org:id:binary:20250107180858650-0634:S1479262124000674:S1479262124000674_fig3.png?pub-status=live)
Figure 3. Hierarchal clustering of 40 pummelo genotypes based on morphological and biochemical.
Sensory analysis
Sensory properties are key to a product's quality, making sensory analysis vital for food quality assessment (Manay and Swamy, Reference Manay and Swamy2002) and data about the sensory evaluation are presented in Table 2. Texture is a key physical attribute that is perceived through visual, tactile and oral sensations. Genotype DVP-2 had the highest mean texture score of 8.5, while DSP-8 had the lowest mean score of 6.9. Colour significantly influences the overall quality assessment of fruits and juices. The mean scores for fruit colour varied between 6.9 and 8.4, with genotypes DVP-2 and DVP-4 achieving the highest mean score of 8.4. Flavour imparts an identifiable character to food and is a central attribute in sensory evaluation. The mean scores for flavour ranged from 6.6 to 8.3 with the highest score recorded in genotypes DVP-1, DVP-2 and DSR-1.
Table 2. Mean scores obtained for sensory analysis of fresh fruits of different pummelo genotypes
![](https://static.cambridge.org/binary/version/id/urn:cambridge.org:id:binary:20250107180858650-0634:S1479262124000674:S1479262124000674_tab2.png?pub-status=live)
Taste is an aesthetic appreciation in the mouth and the most important sensory characteristic for determining the acceptability of fruits and juices. Genotypes DVP-2 and DVP-4 scored the highest for taste with a mean score of 8.3. The overall acceptability of the fruit pulp was assessed through sensory analysis, with genotype DVP-2 receiving the maximum mean score of 8.4 (online Supplementary Fig. S1). This was followed by DVP-4 and DSP-19, each with a score of 8.3.
Discussion
This study primarily aimed to identify key traits through morphological and biochemical analysis to enhance our understanding of genetic diversity, population structure and the relationships among and within the genotypes. The present study examined 40 pummelo genotypes from Devanahalli and Bengaluru Rural areas, revealing significant genetic variations based on 21 morphological and biochemical traits. Key traits that influenced fruit quality and differentiated among the populations are fruit length, diameter, fruit weight, pulp weight, peel weight, seed weight, taste (TSS: TA), TSS, acidity, TPCs and TFC.
Several studies have examined the diversity of Indian pummelo (Singh et al., Reference Singh, Singh, Nath, Parthasarathy, Sthapit and Vinoth2015a, Reference Singh, Singh, Singh, Parthasarathy and Vinoth2015b; Hazarika et al., Reference Hazarika, Lalthlamuani, Lalchhanmawia, Lalrinfeli and Nautiyal2016; Kaur et al., Reference Kaur, Malik, Choudhary, Rohini, Choudhury and Kumar2019). This study aims to analyse the genetic diversity of Devanahalli pummelo, a GI-tagged variety from the region. Few varieties such as DVP-1, HSR-1, DNM-1 and COHB-2 showed higher fruit weight (1746.23 g) and volume (3528.33 ml) compared to previous studies indicating that Devanahalli pummelos are significantly larger (Hazarika et al., Reference Hazarika, Lalthlamuani, Lalchhanmawia, Lalrinfeli and Nautiyal2016; Haque et al., Reference Haque, Hossain, Rahim, Alam, Elhakem, Alqurashi, Althaqafi, Aloufi and Sami2022).
Among the various physico-chemical characters, total seed weight, peel weight, diameter of the axis, fruit volume, fruit weight, pulp weight and peel weight show high PCV and GCV values, indicating substantial genetic variability. Suggesting strong genetic bases and being less influenced by the environment, making them good candidates for selection in breeding programmes (Mishra et al., Reference Mishra, Ram and Kumar2015, Ahsan et al., Reference Ahsan, Majidano, Bhutto, Soomro, Panhwar, Channa and Sial2015). Fruit length, fruit diameter, TSS/TA ratio, number of segments and TSS show moderate to low PCV and GCV. These traits are influenced more by environmental factors or have less genetic diversity within the studied population, making them less reliable for selection based solely on these metrics (Sekhon and Sharma, Reference Sekhon and Sharma2019). The close values of PCV and GCV observed for most traits suggest that the observed variability is largely due to genetic differences (Singh et al., Reference Singh, Raju, Prasad and Bharadwaj2008, Nayak et al., Reference Nayak, Singh and Srivastav2013).
However, genetic variation alone does not fully explain the extent of inheritable differences. To better understand this, we estimated broad-sense heritability for all traits. The traits analysed exhibited very high heritability, ranging from 90.20 to 99.70%. Traits such as fruit weight, pulp weight and peel weight, TPC, TFC, ascorbic acid content and TSS had higher heritability indicating that these characteristics are predominantly governed by genetics and amenable to change through selection as phenotypic expression (Angami et al., Reference Angami, Kalita, Touthang, Makdoh, Bagra, Singh and Lungmuana2022). Similarly high heritability values have been reported for average fruit weight, juice content and acidity in sweet oranges (Rajae et al., Reference Rajae, Najat, Mohamed, Tarik, Ennacir, Adnane, Kawtar and Hamid2019).
Genetic gain values, which indicate the expected improvement from selection are recorded higher for TPC, suggesting a strong potential for rapid improvement of the traits through selection. Total antioxidant activity, diameter of the axis, fruit volume and fruit weight, though have very high heritability show slightly lower genetic gains, indicating a slower but still effective potential for improvement (Al-Tabbal and Al-Fraihat, Reference Al-Tabbal and Al-Fraihat2012; Rattanpal et al., Reference Rattanpal, Uppal, Gupta and Gill2022). Traits such as rind thickness, total number of seeds per fruit, pulp weight, peel weight and total sugar content exhibit high heritability but moderate genetic gains, suggesting that these traits can be improved through selection but may be more gradual (Dudhe et al., Reference Dudhe, Mulpuri, Meena, Ajjanavara, Kodeboyina and Adala2019). Angami et al. (Reference Angami, Kalita, Touthang, Makdoh, Bagra, Singh and Lungmuana2022) reported that the number of seeds per fruit, fruit weight and peel thickness exhibited the highest GCV and PCV, indicating strong genetic control and making them highly effective targets for selection. Genetic correlations between various fruit quality traits were weak except few of the traits such as TFC, TPC and non-reducing sugars. The findings indicated that there is a positive correlation between pulp weight with traits such as fruit diameter, length, axis diameter, fruit weight, total sugar, reducing sugar and TSS: TA. So, the selection of higher pulped fruits would lead to the selection of the above-mentioned traits (Ahmed et al., Reference Ahmed, Rattanpal and Singh2019; Kaur et al., Reference Kaur, Malik, Choudhary, Rohini, Choudhury and Kumar2019; Haque et al., Reference Haque, Hossain, Rahim, Alam, Elhakem, Alqurashi, Althaqafi, Aloufi and Sami2022).
Cluster analysis showed that most of pummelo's seedling belongs to different sources of origin which indicates that differences are due to factors beyond just geographical separation. It also suggests that genotypes from the same geographic location can have different genetic compositions or vice-versa (Majumder et al., Reference Majumder, Hassan, Rahim and Kabir2013). Though the first three PCs account only for about 62.99% of the variability, selection pressure may not be enough to get the desired type genotypes (Singh et al., Reference Singh, Singh, Nath, Parthasarathy, Sthapit and Vinoth2015a, Reference Singh, Singh, Singh, Parthasarathy and Vinoth2015b).
Ascorbic acid and titrable acidity content were higher than in previous studies conducted by researchers in pummelo (Singh et al., Reference Singh, Singh, Nath, Parthasarathy, Sthapit and Vinoth2015a; Hazarika et al., Reference Hazarika, Lalthlamuani, Lalchhanmawia, Lalrinfeli and Nautiyal2016; Angami et al., Reference Angami, Kalita, Touthang, Makdoh, Bagra, Singh and Lungmuana2022). Ascorbic acid was positively correlated with TPC and TFC in the present study. DSP series genotypes have higher content than other genotypes, making this trait a crucial selection phenomenon as these genotypes have higher nutraceutical value (Chaturvedi et al., Reference Chaturvedi, Kumara, Sane, Singh, Kumar and Tripathi2023). TSS is vital for determining fruit sweetness and consumer preference (Kumar et al., Reference Kumar, Lamers, Singh, Ladaniya and Sthapit2015; Hazarika et al., Reference Hazarika, Lalthlamuani, Lalchhanmawia, Lalrinfeli and Nautiyal2016). The positive correlation between TSS, fruit size and pulp weight makes TSS a key marker for selecting high-quality, flavourful varieties. Additionally, reducing sugar content, linked to fruit weight, pulp weight and acidity, is useful for identifying elite genotypes.
Pummelo fruit pulp is rich in various health-boosting secondary metabolites such as polyphenols and flavonoids, which are renowned for their antioxidant properties; because of these traits, popularity of pummelo is growing in India right now (Kumar et al., Reference Kumar, Lamers, Singh, Ladaniya and Sthapit2015). The DSP series genotypes showed higher TPC and TFC compared to other genotypes. DSP-5, DSP-6, DSP-8 and DSP-16 showed higher TPC and TFC levels (Zhang et al., Reference Zhang, Duan, Zang, Huang and Liu2011; Nayak et al., Reference Nayak, Sethi, Sharma, Dubey and Bhowmik2020). The other genotypes were similar to those found in citrus species (Pichaiyongvongdee et al., Reference Pichaiyongvongdee, Rattanapun and Haruenkit2014; Kumar et al., Reference Kumar, Sharma, Dubey, Sharma, Sethi, Mishra, Mathur, Vittal, Shivran and Sharma2024). TFC, TAA and TPC also have higher genetic gain and advance which can be used in further selection of the elite genotypes.
DVP-1 and DVP-2 had the highest TSS/TA ratio, indicating superior taste and potential consumer appeal. Genotypes DVP-2, DVP-4, DSP-19 and DVP-1 showed the highest overall fruit pulp acceptability, making them promising candidates to be used in breeding programmes or for further evaluation. The sensory scores were higher than those reported for pummelo by Gaikwad et al. (Reference Gaikwad, Haldavanekar and Jadhav2019) and for mandarin by Lego et al. (Reference Lego, Upadhyay, Bhutia and Sharma2023). This suggests that Devanahalli pummelo's balanced sweetness and acidity make it superior in terms of taste.
Buyers of table fruits, including pummelo, prefer fruits with high weight, size, pulp content and favourable quality traits such as minimal seeds, high juice yield, vitamin C, TSS and a balanced sugar-to-acid ratio (Hazarika et al., Reference Hazarika, Lalthlamuani, Lalchhanmawia, Lalrinfeli and Nautiyal2016). In breeding, these traits are prioritized. Based on the study, DVP-1 and DVP-2 are superior genotypes among those surveyed in the Devanahalli region. Further studies are needed for the selection of increased TSS and reduced rind thickness genotypes, leading to improved pummelo varieties for growers. These chosen genotypes could serve as crucial breeding material for creating improved varieties.
Future research should expand the genetic pool by exploring diverse pummelo populations in Devanahalli and surrounding regions to enhance variability and uncover elite genotypes. Modern techniques such as marker-assisted selection and genomic analysis can accelerate breeding for fruit quality, disease resistance and adaptability and can be used to differentiate the genotype. With urbanization threatening genetic resources in regions like Devanahalli, both in-situ and ex-situ conservation efforts are crucial. Barcoding for the elite genotypes has to be done in response to conserving the superior genotypes. Investigating pummelo's antioxidant properties can also boost its use in nutraceuticals, increasing market demand and promoting sustainable farming. Ongoing efforts will drive the development of resilient, high-quality citrus varieties for future agricultural needs.
Conclusion
This comprehensive study on the genetic diversity of 40 pummelo genotypes from the Devanahalli region of Karnataka, India, reveals significant morphological and biochemical variation, offering valuable insights for future breeding programmes. Our results demonstrate substantial genetic variability across key fruit traits, such as fruit size, pulp weight, peel weight and TSS, suggesting strong genetic control over these characteristics. High heritability and genetic gain estimates for certain traits, such as total phenol and flavonoid content, indicate a strong potential for improvement through targeted selection. Conversely, traits such as the TSS/TA ratio and number of segments, which exhibit lower heritability, may require more complex breeding strategies. Correlation analysis further highlights the potential for simultaneous enhancement of fruit size, quality and nutritional value in pummelo breeding programmes. The clustering and PCA analyses underscore the genetic diversity within the pummelo genotypes, dividing them into two distinct groups and identifying genotypes with superior traits for commercial and nutritional purposes. This emphasizes the importance of sensory attributes, with certain genotypes showing high consumer acceptability based on taste, texture and flavour. These findings not only enhance our understanding of pummelo genetic diversity but also lay the groundwork for developing citrus varieties with improved fruit quality, disease resistance and adaptability.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/S1479262124000674
Acknowledgements
The authors express their gratitude to the pummelo growers of the Devanahalli region and the Horticulture Department of Karnataka for providing the experimental material (fruits) for this study. They also thank the Department of Fruit Science, College of Horticulture Bengaluru, UHS, Bagalkot, for their technical support throughout the study.