Morphology, Heterosis, and Fertility of Novel CMS-Based Solanum melongena × S. aethiopicum Hybrids

Krommydas, Konstantinos; Mavromatis, Athanasios; Bletsos, Fotios; Roupakias, Demetrios

doi:10.3390/agronomy16030306

Open AccessArticle

Morphology, Heterosis, and Fertility of Novel CMS-Based Solanum melongena × S. aethiopicum Hybrids

¹

Institute of Plant Breeding and Genetic Resources, Hellenic Agricultural Organization-Dimitra (ELGO-Dimitra), Thermi, GR-57001 Thessaloniki, Greece

²

Laboratory of Genetics & Plant Breeding, Faculty of Agriculture, Aristotle University of Thessaloniki, GR-54124 Thessaloniki, Greece

^*

Author to whom correspondence should be addressed.

Agronomy 2026, 16(3), 306; https://doi.org/10.3390/agronomy16030306

Submission received: 31 December 2025 / Revised: 16 January 2026 / Accepted: 22 January 2026 / Published: 26 January 2026

(This article belongs to the Special Issue Genetics and Breeding of Field Crops in the 21st Century—2nd Edition)

Download

Browse Figures

Versions Notes

Abstract

Although cytoplasmic male sterility (CMS) is well established in eggplant, CMS-based interspecific hybrids with allied species have not yet been reported or studied. In this study, five previously developed CMS-based interspecific F₁ hybrids between eggplant and Solanum aethiopicum Group Aculeatum (=S. integrifolium) and Group Gilo (=S. gilo), together with their parental lines, were morphologically evaluated for 67 seedling, vegetative, floral, and fruit traits, and their heterosis for vegetative growth was studied. Male fertility was assessed based on anther morphology and pollen viability, while female fertility was evaluated through backcrosses to both parents. The hybrids exhibited predominantly intermediate phenotypes and clustered distinctly from parental lines as confirmed by principal component analysis. Remarkable heterosis was observed for most growth-related traits, indicating favorable nuclear–cytoplasmic interactions despite the use of CMS eggplant lines as maternal parents. All hybrids showed complete male sterility, characterized by non-viable pollen and pronounced anther homeotic alterations, the latter indicating CMS-related effects on male fertility. Female fertility was severely reduced, likely due to meiotic irregularities, as evidenced by the failure of most attempted backcrosses. However, successful recovery of BC₁ progeny after backcrossing one CMS-based F₁ hybrid to S. gilo demonstrates partial reproductive compatibility and provides a genetic bridge for CMS introgression into S. gilo. These results indicate that CMS systems are suitable for eggplant interspecific crosses aimed at vigorous rootstock production and CMS cytoplasm introgression into allied germplasm.

Keywords:

eggplant; cytoplasmic male sterility; wide hybridization; rootstock breeding; heterosis; Solanum gilo; Solanum integrifolium; plant genetic resources

Graphical Abstract

1. Introduction

Eggplant (Solanum melongena L.) is an economically important crop of the Solanaceae family, with worldwide cultivation, especially in Asia and in the Mediterranean region. According to FAO statistics, in 2023, eggplant was cultivated on 1,922,783 hectares worldwide, with a total production of 60,793,941 t [1]. Modern agricultural practices and changing climate conditions pose new challenges in eggplant cultivation. Therefore, the development of genotypes better adapted to low-input farming and resistant to various stresses is crucial for the crop’s sustainability.

Wild and cultivated Solanum species, as well as interspecific hybrids with eggplant, are valuable genetic resources for introgression breeding programs aiming at improving biotic [2,3,4,5,6,7,8] and abiotic stress tolerance [9,10,11,12,13], nutrient use efficiency [14], and fruit quality [15,16,17,18]. Eggplant wide crosses have also been used for the development of grafting rootstocks [19], cytoplasmic male sterility (CMS) systems [20], and mapping populations [21,22,23,24].

Eggplant interspecific hybrids are promising grafting rootstocks, combining various resistances and hybrid vigor with more uniform germination, good grafting compatibility [25,26,27], offering greater practicality for the grafting procedure than the commonly used wild Solanum rootstocks, which possess some disadvantages including irregular germination, slow seedling growth, inconsistent grafting compatibility and unclear status of breeder’s rights [28,29,30,31,32]. However, large-scale interspecific hybrid production is constrained due to the need for manual emasculation in predominantly hermaphroditic Solanum species, making hybrid seed production a highly specialized, labor-intensive, costly procedure.

Utilization of CMS eggplant lines as female parents in crosses naturally prevents self-pollination due to anther non-dehiscence [33] or lack of pollen production [34], offering a cost-effective hybridization method, while ensuring seed genetic purity. Despite the availability of eggplant CMS lines for almost 40 years [35], their application in interspecific hybrid production has not yet been reported.

In a previous work, we produced alloplasmic CMS lines of three Greek eggplant cultivars carrying the cytoplasm of S. violaceum [36]. The CMS eggplant lines used in this study were derived from three popular Greek eggplant cultivars with distinct agronomic and morphological characteristics: ‘Langada’ is mid-early, vigorous, high-yielding with large, cylindrical purple fruits; ‘Emi’ is early and vigorous, with large, oval purple fruits; whereas ‘Tsakoniki’ is early, more compact, with medium-sized violet fruits with white stripes.

Preliminary crosses of these CMS lines to prickly S. integrifolium (=S. aethiopicum L. Group Aculeatum) and cultivated S. gilo (=S. aethiopicum L. Group Gilo) produced the respective CMS-based interspecific hybrids (Figure 1), suggesting a convenient and practical approach for large-scale interspecific hybrids. In addition, since S. aethiopicum groups are important African vegetables [37], transferring CMS into S. integrifolium and S. gilo through cytoplasm substitution would offer new breeding opportunities for these crops.

Both S. integrifolium and S. gilo are valuable for rootstock and introgression breeding, due to their resistance against various stresses, including dominant Fusarium wilt resistance [7] and good crossability with eggplant [38,39,40]. Their interspecific hybrids with eggplant exhibit vegetative vigor, good grafting compatibility, and tolerance to nematodes [25]. These characteristics make these two groups good candidates for interspecific rootstock production.

Before recommending CMS-based interspecific hybrids as rootstocks or genetic materials for breeding purposes, some questions need to be addressed. First, it is important to determine whether the nuclear–cytoplasmic composition of these hybrids (Figure 1) affects their vegetative vigor. Second, the identification of the most heterotic parental combinations is critical for selecting crossing parents for rootstock production. Third, the potential presence of fertility restorer genes (Rf-genes) inherited from S. integrifolium and S. gilo in their CMS-based hybrids would be of practical value and needs to be investigated. Finally, the possibility of transferring the CMS inducing cytoplasm from CMS eggplant in S. integrifolium and S. gilo should also be evaluated.

The objectives of this study were to (i) characterize the phenotype and evaluate the vigor of CMS-based interspecific hybrids between CMS eggplant and S. integrifolium and S. gilo, (ii) assess hybrid heterosis and identify the most heterotic parental combinations, (iii) evaluate male and female fertility to detect potential Rf-genes, and (iv) develop backcross progenies for introgression breeding and transferring the CMS cytoplasm into the S. aethiopicum groups.

2. Materials and Methods

2.1. Plant Material

Five CMS-based interspecific hybrids were previously produced using the scheme presented in Figure 1. The female parents of the hybrids were the CMS lines of Greek cultivars ‘Langada’ (cmsL), ‘Emi’ (cmsE), and ‘Tsakoniki’ (cmsT) carrying the S. violaceum cytoplasm [36]. The male parents were two S. aethiopicum groups: the wild Aculeatum (S. integrifolium; SI) and the cultivated Gilo (S. gilo; SG). In the present study, the evaluated hybrids were cmsL × SI, cmsE × SI, cmsT × SI, cmsE × SG, and cmsT × SG, as the cross cmsL × SG was unsuccessful. Parental genotypes were also grown as controls. The parental plant materials were provided by the Institute of Plant Breeding and Genetic Resources (IPBGR) of ELGO-DEMETER (Thermi, Thessaloniki, Greece).

2.2. Experimental Design, Site, and Climatic Conditions

The experiment was conducted from May to September in an open field at IPBGR (latitude: 40.54° N, longitude: 22.99° E, altitude: 2 m). Climatic conditions are shown in Figure A1. Seeds of the plant material were sown in commercial peat-based substrate, and seedlings were raised for 40 days in an unheated greenhouse. Plants were transplanted in the experimental field in May, arranged according to a completely randomized experimental design, with 10 plants per genotype. The experimental area covered approximately 85 m², including border corridors. Standard eggplant cultivation practices were applied throughout the experiment.

2.3. Phenotypic Characterization

The phenotypic characterization of the plant material was assessed at seedling, vegetative, flowering, and fruit-setting stages using CPVO TP-117/1 eggplant descriptors [41], with modifications when necessary (e.g., addition of fruit color and shape classes for S. integrifolium and S. gilo). Additional descriptors were defined when suitable ones were not available (Table A1). Quantitative traits were measured using the appropriate methods and organs (ruler, caliper, weighing scale, and manual counting). Digital imaging methods for leaf area estimation and leaf blade measurements are given in detail in Appendix A.1 and Appendix A.2.

2.3.1. Seedling Traits

The evaluated seedling traits included hypocotyl coloration, cotyledon length, width, and length-to-width ratio, presence of leaf prickles, and color and intensity of anthocyanins in midrib and prickles.

2.3.2. Plant Traits

Plant traits were examined at anthesis (early June) and included plant growth habit, shoot coloration and anthocyanin intensity, shoot prickle color, and type of inflorescence. Quantitative measurements were made for estimated leaf area (ELA), plant height, number of shoots and leaves, internode length, prickles per internode, flowers per inflorescence, and number of fruits per infructescence. ELA was determined using a digital imaging approach implemented in ImageJ software (version 1.50), described in detail in Appendix A.1.

2.3.3. Leaf Traits

Leaf measurements included midrib coloration, leaf dimensions, prickles number, margin sinuation, and number of lobes. Petiole length and diameter, as well as the number of petiole prickles, were also recorded. Leaf blade dimensions were also obtained using an ImageJ-based digital scanning procedure (Appendix A.2).

2.3.4. Flower Traits

For the characterization of the flower, the following traits were recorded: corolla color, presence of corolla stripes, corolla diameter, stigma color, and the number of anthers, petals, sepals, and stigma lobes.

2.3.5. Fruit Traits

Fruit traits were evaluated on fully developed fruits at the harvest maturity and physiological maturity stages for fruit shape, ribbing intensity, pistil scar depression depth and size, fruit color, presence of surface patterns, flesh color, and flesh texture. Quantitative traits included fruit weight, length, diameter, peduncle length, fruit calyx prickles, and number of locules.

2.4. Evaluation of Heterosis

To evaluate the relative performance of the hybrids in comparison to their parental controls, mid-parent heterosis (MPH) and heterobeltiosis (better-parent heterosis; BPH) were calculated for growth-related traits following the equations given in Fehr [42]:

M P H (%) = \frac{(F_{1} - M P)}{M P} \times 100

B P H (%) = \frac{(F_{1} - B P)}{B P} \times 100

where F₁ = hybrid mean, MP = mid-parent mean, and BP = better-parent mean.

2.5. Male Fertility Assessment

Male fertility was evaluated using pollen staining, pollen release tests, and anther morphology, as described in a previous study [36]. For the staining method, anthers from 10 freshly opened flowers per genotype were smeared in 2% acetocarmine solution, 100 pollen grains per flower were counted with three replications, and the percentage of viable pollen grains was scored. Pollen release ability was assessed by tapping anthers over a black surface and examining pore dehiscence with a stereomicroscope. Anther morphology was visually examined both in the field and in the laboratory.

2.6. Female Fertility Assessment

Female fertility was determined by the percentage of successful backcrosses to their respective eggplant cultivars and allied species. For each combination, 30 backcrosses were made using the first four flowers of the plants (excluding the first) during June. Ten artificial self-pollinations per parental material were used as controls. The resulting fruits were harvested at physiological maturity, and the percentage of seeded fruits was calculated.

2.7. Statistical Analysis

Analysis of variance for morphological traits and pollen viability was conducted using SPSS v.17.0 (IBM Corp., Armonk, NY, USA and statistically significantly different means were separated using Tukey’s HSD test (p ≤ 0.05). Significance of mid- and better-parent heterosis was tested using a t-test (p ≤ 0.05) against mid- and better-parent values. Principal component analysis (PCA) was carried out to explore relationships among traits and among genotypes (XLSTAT, Lumivero, Denver, CO, USA).

3. Results

3.1. Morphological Characterization

3.1.1. Seedling

Seedling morphological traits are shown in Table 1. In general, the genotypes were distinguished by the presence of anthocyanins and prickles in various organs, while each genotype had a uniform phenotype, and no segregation was observed for the evaluated traits (Figure 2). All eggplant cultivars and the interspecific hybrids had purple hypocotyls, while hypocotyls of S. integrifolium and S. gilo were green. Additionally, leaf midrib anthocyanins were present in eggplant, S. integrifolium and hybrids, but were absent in S. gilo. Seedlings of S. integrifolium and its hybrids were also characterized by leaf prickles, whereas S. gilo, its hybrids, and eggplant lacked prickles.

Interestingly, anthocyanin formation in leaf midrib and prickles of S. integrifolium resulted in vinaceous coloration and was light-dependent, i.e., shaded organs did not form anthocyanins, contrasting to the consistent typical purple coloration observed in eggplant and hybrid genotypes (Table 1, Figure 2). Solanum integrifolium had weaker prickle anthocyanin intensity than its hybrids.

Cotyledon dimensions varied significantly among the plant material (Table A2). S. integrifolium and S. gilo had the shortest cotyledons (2.2 and 2.4 cm, respectively), while eggplant cultivars (2.9–3.4) and interspecific hybrids (2.8–3.4) had the longest (Table 1). Cotyledon width ranged from 0.9 cm in S. gilo to 1.2 cm in S. integrifolium. The genotypes were also differentiated by the cotyledon length-to-width ratio (Table 1). Eggplant ‘Langada’ had the highest ratio (3.66) while S. integrifolium had the lowest (1.98), with intermediate values recorded in the hybrids.

3.1.2. Plant

The overall appearance of the interspecific hybrids was intermediate between their respective parental controls (Figure 3). However, hybrid genotypes exhibited superior performance for most growth-related traits (Table 2). At the end of the experiment, representative plants were uprooted, air-dried, and their root systems were visually inspected, revealing a more extensive root system in the hybrids.

Plant growth habit was erect in eggplant cultivars ‘Langada’ and ‘Tsakoniki’, and was semi-erect in ‘Emi’, S. integrifolium, and S. gilo (Figure 3, Table 2), while the hybrids of S. integrifolium and S. gilo were semi-erect and spreading, respectively. Shoot coloration was vinaceous in S. integrifolium, green in S. gilo, and typical purple in eggplant and the interspecific hybrids. Moreover, the hybrids of S. integrifolium had very strong shoot anthocyanin intensity, in contrast to the other genotypes, which had medium intensity.

The estimated leaf area of the hybrids was more than double compared to their respective controls (2709.7–2984.4 cm² vs. 942.4–1388.8 cm²). Also, hybrid plants were significantly taller (101.6–133.6 cm) than the controls (70.0–95.2 cm), while S. integrifolium hybrids were significantly taller than those of S. gilo. The interspecific hybrids also produced more shoots (22.5–41.8) and leaves (76.8–107.0) compared to the parental genotypes (6.8–11.4 and 26.0–42.6, respectively).

The interspecific hybrids had the longest internodes (12.3–14.3 cm), followed by eggplant (11.0–13.0 cm), while S. integrifolium and S. gilo had significantly shortest internodes (9.0 and 8.9 cm, respectively). Both S. integrifolium and its hybrids had internode prickles, but the latter had significantly fewer prickles per internode (1.6–2.8) than the former (4.3). Internode prickle density was highest in S. integrifolium (0.48 prickles cm⁻¹) and moderate in its hybrids (0.17–0.20 prickles cm⁻¹). The color of these prickles was similar to the shoot color, with vinaceous prickles in S. integrifolium and purple in the hybrids.

Inflorescence traits also varied significantly among genotypes (Table 2). Both S. integrifolium and its hybrids had compound inflorescences, with the former having significantly more flowers (9.6) than the latter (5.0–7.6). It is worth noting that while S. gilo and eggplant formed simple inflorescences with few flowers (1.3–2.1), their interspecific hybrids had compound inflorescences with significantly more flowers (5.0–6.3). It was also observed that in September, both F₁(cmsE × SI) and F₁(cmsE × SG) set many parthenocarpic fruits, averaging 3.7 and 2.5 fruits per infructescence, respectively. The number of fruits per infructescence in the parental genotypes ranged between 1.0 and 1.6.

3.1.3. Leaf

Leaf morphological traits also varied among the plant material (Table 3). The leaf midrib was vinaceous in S. integrifolium, green in S. gilo, and purple in eggplant and hybrid materials. Eggplant and hybrid genotypes had significantly longer leaves (22.4–26.6 cm) than S. integrifolium and S. gilo (16.2–16.8 cm). In addition, the hybrids had the widest leaves (18.7–20.7 cm), exceeding both parents (14.9–15.8 cm). The length/width ratio was highest in eggplant (1.42–1.78), lowest in S. integrifolium and S. gilo (1.07 and 1.04, respectively), and intermediate in the hybrids (1.21–1.29). The interspecific hybrids also had the largest leaf blade area (283.9–330.9 cm²), significantly surpassing both parental controls (168.3–259.8 cm²). The leaves of S. integrifolium, S. gilo, and their hybrids were strongly sinuated, contrasting with the very weakly sinuated eggplant leaves, while the number of leaf lobes did not differ significantly between the genotypes.

Solanum integrifolium and its hybrids were characterized by numerous prickles on both leaf surfaces (Table 3). On the other hand, eggplant cultivars sparsely formed very small prickles, and S. gilo and its hybrids completely lacked prickles. S. integrifolium averaged 11.0 prickles on the upper and 17.5 on the lower surface, significantly more than its respective hybrids (7.8–9.0 and 7.2–9.7, respectively) and eggplant (0.75–4.8 and 0.2–1.3, respectively). Notably, S. integrifolium produced more prickles on the lower leaf surface, contrary to eggplant, while the hybrids had an equal distribution. Leaf prickle density was significantly higher in S. integrifolium (0.06 prickles/cm² on the upper and 0.10 on the lower surface than in the interspecific hybrids (0.02–0.03).

Leaf petiole was longer in hybrids (8.7–12.1 cm), intermediate in eggplant (4.8–9.5 cm) and shorter in S. integrifolium and S. gilo (5.3–5.4 cm), while the smallest petiole diameter was recorded in S. integrifolium and S. gilo (0.55–0.62 cm) and the largest in eggplant and hybrids (0.70–0.88 cm). The highest number of petiole prickles was recorded in the interspecific hybrids (3.0–4.2), followed by S. integrifolium (2.7) and eggplant (0.40–0.83). Petiole prickles were denser in S. integrifolium (0.50 prickles per cm), while its hybrids displayed intermediate values (0.28–0.45).

3.1.4. Flower

Flower morphology of the interspecific hybrids was generally intermediate between the respective eggplant cultivars and the wild species (Figure 4, Table 4). Eggplant flowers were purple with broad radial stripes of a lighter hue, contrasting with the entirely white flowers of S. integrifolium and S. gilo. The flowers of the hybrids were white with radial, narrow purple stripes, indicating the presence of anthocyanins. Eggplant had by far the largest corolla diameter (5.2–5.5 cm), while the interspecific hybrids (2.7–3.1 cm) were closer to S. integrifolium and S. gilo (2.3 cm).

In addition, the plant material exhibited differences in the stigma coloration, which was light orange in S. integrifolium and S. gilo, green in eggplant cultivars ‘Emi’ and ‘Langada’, and light green in ‘Tsakoniki’ and the hybrids. The number of petals, sepals, and anthers was slightly lower in hybrids (5.4–6.2) than in eggplant cultivars (6.2–7.5) and similar to S. integrifolium and S. gilo (6.1–6.4), with significant differences observed only between cv. ‘Emi’ and F₁(cmsT × SG). Moreover, S. integrifolium and S. gilo had fewer stigma lobes (3.3–3.5) compared to eggplant (4.4–4.8) and their hybrids (4.0–4.6).

3.1.5. Fruit

The genotypes examined displayed differences in fruit morphology and fruit-setting behavior (Figure 5, Table 5). The eggplant cultivars, as well as S. integrifolium and S. gilo, produced seeded fruits following both natural and artificial self-pollination. In contrast, the interspecific hybrids did not set any fruit until September, when abundant parthenocarpic fruit formation was observed in F₁(cmsE × SI) and F₁(cmsE × SG). The external appearance of these parthenocarpic fruits more closely resembled that of S. integrifolium and S. gilo rather than eggplant, although they exhibited several unique traits (Figure 5).

Solanum integrifolium and S. gilo had fruits of similar morphology (Figure 5, Table 5). Those of S. integrifolium were flattened with very strong ribbing, an indented apex, a very large pistil scar, and a very deep indentation. At harvest maturity, the fruit skin was green and very glossy, with weak, sparse stripes, turning red at physiological maturity. The fruit’s flesh was white with a gelatinous-to-compact texture. Solanum gilo had flattened fruits, with strong ribbing and an indented-to-flattened apex. The pistil scar was also very large with a deep indentation. External fruit features resembled S. integrifolium except for medium glossiness, while its flesh was orange with a gelatinous texture. Fruits of eggplant cv. ‘Emi’ were obovate and lacked ribs, with a flattened apex. The pistil scar was medium-sized with an absent or very shallow indentation. At harvest maturity, fruit color was purple without stripes and with medium to strong glossiness, turning brown at physiological maturity. The fruit flesh was white with a compact texture.

Both CMS-based hybrids produced flattened to globular fruits, with medium ribs and an indented-to-flattened apex. The pistil scar was very small with a shallow indentation, distinguishing the hybrids from their parents. At harvest maturity, the fruit skin was green, turning to orange at physiological maturity. Fruit stripes were present and more pronounced than in S. integrifolium and S. gilo, with medium density. The flesh color was green with gelatinous-to-compact texture in F₁(cmsE × SI) and light green with a gelatinous texture in F₁(cmsE × SG).

Eggplant cv. ‘Emi’ produced the heaviest fruits (436.8 g), whereas F₁(cmsE × SI) and F₁(cmsE × SG) had the lowest fruit weight (15.9 and 19.9 g, respectively). Fruits of S. integrifolium and S. gilo were approximately twice as heavy as those of the interspecific hybrids (37.5 and 35.4 g, respectively). The longest fruit peduncle was recorded in ‘Emi’ (8.6 cm), followed by the interspecific hybrids (3.0 and 2.8 cm, respectively), while S. integrifolium and S. gilo had the shortest peduncles (1.7 and 1.4 cm, respectively).

Regarding fruit dimensions, cv. ‘Emi’ had the largest polar and equatorial diameters (15.4 and 8.7 cm, respectively), while S. integrifolium and S. gilo had significantly smaller polar (2.9 and 2.6, respectively) and equatorial (5.2 and 4.8 cm, respectively) diameters. F₁(cmsE × SI) had a significantly smaller value for polar diameter than S. integrifolium, while the opposite was observed for F₁(cmsE × SG) and S. gilo. The interspecific hybrids recorded the smallest values for equatorial diameter (3.6–3.8) among the plant material. Consistent with its fruit shape, eggplant ‘Emi’ showed the highest equatorial/polar diameter ratio (1.78), while S. integrifolium and S. gilo exhibited the lowest ratios (0.55 and 0.54). On the other hand, the hybrids had intermediate values, with F₁(cmsE × SI) producing slightly more flattened fruits (0.72) than F₁(cmsE × SG) (0.79).

Interestingly, the highest number of calyx prickles was observed in F₁(cmsE × SI), which formed 6.7 prickles, followed by ‘Emi’ (3.7) and S. integrifolium (1.8). Solanum gilo and its hybrid did not form calyx prickles. Finally, no significant differences were observed for the number of locules, which ranged from 4.9 in ‘Emi’ to 6.0 in S. integrifolium.

3.1.6. Principal Component Analysis

The 47 traits used in PCA correspond to all seedling, plant, leaf, and flower descriptors (Table A1), while fruit traits were excluded because most hybrids were sterile and did not produce fruit. Principal component analysis identified two principal components, PC1 and PC2, explaining 37.7 and 30.5% of the variation, respectively (Figure 6). These components were used as the x and y axes of the trait variables loading plot and genotype score plot. The loading plot indicated a considerable degree of trait grouping with respect to the plant organ and function. For example, seedling anthocyanin traits such as anthocyanin presence in the hypocotyl (S_HA), midrib color (S_LMC), and anthocyanin intensity (S_LMA_US, S_LMA_LS) were closely grouped on the upper right quadrant, while seedling leaf prickles anthocyanin intensity (S_LPA_US, S_LPA_LS) were located on the lower right quadrant close to the x-axis. Seedling cotyledon length (S_CL) and length/width ratio (S_CL/W) were positioned on the upper right and left quadrant, respectively, close to the y-axis. Seedling cotyledon width (S_CW) and leaf prickle presence (S_LPP) were in the lower right quadrant.

Plant traits were located on the right upper and lower quadrants (Figure 6). Specifically, growth-related traits, including height (P_H), estimated leaf area (P_ELA), number of shoots (P_NS), leaves (P_NL), and internode length (P_INL), as well as shoot anthocyanin intensity (P_SAI), were closely positioned on the upper right quadrant above the x-axis, suggesting a close relationship. On the other hand, plant traits regarding prickle presence (P_NPI), density (P_NPIcm), and anthocyanin presence (P_SC, P_SPC) were grouped on the lower right quadrant, together with inflorescence (P_IT, P_NFlInfl) and infructescence (P_NFrInfr) traits.

Most leaf traits were also located on the right quadrants (Figure 6). Traits related to leaf growth such as length (L_L), width (L_W), area (L_A), petiole length (L_PL) and diameter (L_PD) were mainly positioned on the upper right quadrant, while leaf traits related to the presence and density of prickles (L_PN_US, L_PNcm_US, L_PN_LS, L_PNcm_LS, L_PNP and L_PNPcm) were on the lower right. Leaf midrib color (L_MC) and sinuation of margin (L_SM) were on the upper and lower right quadrants, respectively, while the number of lobes (L_NL) was on the lower left quadrant.

Most of the flower traits were located on the upper and lower left quadrants (Figure 6). Some quantitative flower traits, including the number of anthers (F_NA), petals (F_NP), and sepals (F_NS), were closely located in the lower-left quadrants close to the x-axis, with flower stigma color (F_SC) also on the same quadrant but lower. Flower corolla color (F_CC) and diameter (F_CD) were on the upper-left quadrant, whereas the presence of flower stripes (F_CS) and the number of stigma lobes (F_NSL) were on the upper-right quadrant.

In addition, PCA clearly distinguished the genotypes examined into five distinct groups as demonstrated in the score plot in Figure 6. The first group, located on the upper-left quadrant, consisted of the three eggplant cultivars, while the second and third groups were represented by S. gilo and S. integrifolium, respectively, positioned on the lower-left and lower-right quadrants. The interspecific hybrids of S. integrifolium formed the fourth group with cmsT × SI and cmsL × SI on the upper-right and cmsE × SI in the lower-right, very close to the x-axis, while the interspecific hybrids of S. gilo were the fifth group on the upper right quadrant close to both axes, at the same height as the respective eggplant cultivars.

3.2. Heterosis and Heterobeltiosis

The interspecific hybrids displayed remarkable heterosis and heterobeltiosis for most of the traits studied (Figure 7), with distinct magnitudes and heterotic patterns among genotypes (Figure A2). In general, the hybrids of female parent cmsT were the most heterotic, with F₁(cmsT × SI) being the most heterotic for the majority of the traits, followed by F₁(cmsT × SG) in the remaining traits. Hybrids of cmsE exhibited statistically significant, yet less pronounced, heterotic responses. With respect to the male parent, hybrids of S. integrifolium were more heterotic compared to those of S. gilo.

Solanum integrifolium hybrids exhibited significant heterosis and heterobeltiosis for plant height, with the highest values (67.0% and 55.31%, respectively) recorded in F₁(cmsT × SI), followed by F₁(cmsE × SI) and F₁(cmsL × SI). Significant heterotic responses, although of lesser magnitude (30.8–36.6%), were also recorded in the hybrids of S. gilo F₁(cmsT × SG) and F₁(cmsE × SG), while heterobeltiosis was significant in the former (29.0%) and non-significant in the latter (18.6%). For internode length, heterosis was significant in all hybrids except for F₁(cmsE × SI), with F₁(cmsT × SI) having the highest value (50.0%), followed by F₁(cmsL × SI) and F₁(cmsE × SG) with 30.3% and 25.9%, respectively. However, nearly all heterobeltiosis estimates were non-significant for this trait, except for F₁(cmsT × SI), which surpassed its best parent by 36.4%.

The trait with the more pronounced heterosis and heterobeltiosis was the number of shoots per plant, with statistically significant values in all hybrids (Figure 7). F₁(cmsT × SG) was the most heterotic genotype (316.5% and 266.7%), followed by F₁(cmsL × SI) and F₁(cmsT × SI). Similar responses were observed for the number of leaves per plant, with heterosis ranging from 97.4% in F₁(cmsE × SG) to 184.2% in F₁(cmsL × SI), and heterobeltiosis ranging from 80.3% in F₁(cmsE × SG) to 151.2% in F₁(cmsT × SG). Significant heterosis and heterobeltiosis were also recorded for estimated leaf area, particularly in F₁(cmsT × SI) (186.8% and 170.2%) and F₁(cmsT × SG) (160.1% and 154.2%), while lower, yet still significant values (98.2–133.37%), were observed in F₁(cmsE × SI) and F₁(cmsE × SG).

The strongest heterotic responses for flower number per inflorescence were observed in the hybrids of S. gilo (Figure 7). That was an expected result, given the contrasting phenotype of the hybrids and their parental genotypes (compound vs. simple inflorescence). F₁(cmsE × SG) and F₁(cmsT × SG) displayed exceptionally high heterosis (237.6% and 233.3%, respectively) and heterobeltiosis (204.3% and 199.4%, respectively). In contrast, S. integrifolium hybrids showed weak heterosis (12.3–37.7%) and negative heterobeltiosis, which ranged from −20.8 to −31.7%.

Heterosis for leaf blade length ranged from 15.3% in F₁(cmsL × SI) to 36.0% in F₁(cmsE × SI), with the former being the only non-significant estimate (Figure 7). Heterobeltiosis for the same trait was non-significant in all hybrids except F₁(cmsE × SI), which surpassed its better parent by 18.9%. For leaf blade width, significant heterosis and heterobeltiosis were recorded in all hybrids, with S. integrifolium hybrids showing uniform and somewhat greater values compared to those of S. gilo. Heterosis ranged from 19.4% in F₁(cmsE × SG) to 32.4% in F₁(cmsT × SI), with comparable heterobeltiosis values, ranging from 18.8% in F₁(cmsE × SG) to 31.2% in F₁(cmsT × SI). Leaf blade area followed a similar heterotic pattern with heterosis ranging from 42.8% F₁(cmsE × SG) to 66.8% in F₁(cmsT × SI). Heterobeltiosis for this trait was not significant in F₁(cmsL × SI) and F₁(cmsE × SG), while significant values ranged from 30.8% in F₁(cmsT × SG) to 52.6% in F₁(cmsT × SI).

3.3. Male Fertility

Stereomicroscopic examination confirmed that the anthers of eggplant, S. integrifolium, and S. gilo had typical morphology and functionality, with terminal pores that dehisced normally during anthesis to release pollen (Figure 8a,d). In contrast, the interspecific hybrids displayed structural and functional abnormalities, including constriction below the terminal pore, partially or slightly dehisced pores, and pores covered by hair-like structures. In addition, anther homeotic changes, i.e., petaloidy and pistilody, were also observed in the hybrids (Figure 8c,f). Pollen viability was high in eggplant cultivars and their allied species, ranging from 90.4% in S. integrifolium to 93.4% in eggplant ‘Emi’ (Figure 8g–i; Figure 9a). Conversely, all interspecific hybrids were completely male sterile, with no viable pollen grains detected.

3.4. Female Fertility

Female fertility of the plant material was assessed based on the percentage of successful backcrosses to the corresponding eggplant cultivars and allied species and the percentage of backcrosses that yielded seeded fruits (Figure 9b). Solanum melongena, S. integrifolium, and S. gilo set seeded fruits after controlled self-pollination, with a success rate ranging between 70.0 and 90.0%. In contrast, the interspecific hybrids failed to produce seeded fruits after selfing or backcrossing, with the exception of F₁(cmsE × SG) and F₁(cmsT × SG), which set one and two fruits, respectively, when backcrossed to S. gilo. These fruits were dissected, and it was concluded that only one fruit from the cross F₁(cmsT × SG) × S. gilo contained seemingly viable backcross seeds, while the other fruits were seedless (parthenocarpic) (Figure 5f and Figure 9b).

4. Discussion

4.1. Phenotypic Characterization and Uniformity of CMS-Based Interspecific Hybrids

One key feature of a reliable CMS system is the ability to ensure genetic purity of hybrid seed by preventing self-pollination [43]. In the present study, the CMS-based interspecific hybrids between S. melongena and the two S. aethiopicum groups exhibited complete phenotypic uniformity across all developmental stages, with no segregation for the qualitative morphological traits evaluated (Table 1, Table 2, Table 3, Table 4 and Table 5, Figure 2, Figure 3, Figure 4 and Figure 5). This confirms the effectiveness of the maternal CMS eggplant lines in preventing self-pollination and ensuring genetic purity of the interspecific hybrid seed.

The phenotype of the hybrids was intermediate relative to their parental species (Figure 3, Figure 4 and Figure 5), which is typical for eggplant wide crosses [44,45,46]. Some traits, including prickliness, anthocyanins, and compound inflorescence, were dominantly inherited, consistent with previous reports [47,48,49,50,51], and confirmed hybridity. The fact that parthenocarpic hybrid fruits more closely resembled those of the allied species suggests dominant inheritance of fruit traits from the S. aethiopicum groups, whereas their reduced size can be attributed to the absence of seeds due to pollen sterility [46]. From the breeders’ point of view, the complete absence of prickles in S. gilo and its hybrids with eggplant, possibly reflecting the different domestication paths of these species, is a trait that could be exploited for the development of prickle-free eggplants through introgression breeding.

Vegetative development traits, including leaf area, shoot number, and internode length, were remarkably enhanced in all hybrids, which is typical for eggplant interspecific hybrids [44,45,52]. Principal component analysis clearly separated the eggplant cultivars, S. integrifolium, S. gilo, and their CMS-based hybrids into distinct clusters (Figure 6), reflecting both taxonomic differentiation and the strong influence of the male parent on hybrid morphology. Collectively, these results indicate that overall morphological development was not impaired by their maternal CMS background.

4.2. Expression of Heterosis and Parental Effects

In eggplant wide hybridization, heterotic responses are often maximized due to large genetic and phenotypic distances and genomic complementation [53]. In this study, all CMS-based interspecific hybrids had a very vigorous development, exhibiting significant heterosis and, in most cases, heterobeltiosis for major growth-related traits (Figure 3 and Figure 7), consistent with overdominant allelic interactions for these traits [54]. This demonstrates that heterosis was not compromised despite the unusual nuclear/cytoplasmic constitution of the hybrids (Figure 1). The combination of high vegetative heterosis and extensive root systems observed in the CMS-based hybrids further supports their potential as rootstocks.

Heterosis analysis showed that the magnitude of heterosis was strongly dependent on the parental combination, in agreement with previous reports on eggplant interspecific hybrids [53]. Hybrids involving the CMS line of ‘Tsakoniki’ were the most heterotic, particularly when crossed with S. integrifolium (Figure 7). This indicates that, in addition to the allied species genome, the nuclear background of the CMS eggplant parent plays a critical role in determining hybrid performance, highlighting the importance of parental selection for CMS-based rootstock breeding.

4.3. Fertility and CMS Effects in the Interspecific Hybrids

All CMS-based interspecific hybrids were completely male sterile, as evidenced by the absence of viable pollen and the presence of severe anther abnormalities (Figure 8 and Figure 9a). Although pollen sterility in eggplant wide hybrids may result from meiotic irregularities caused by parental genomic divergence [55,56,57,58,59,60,61], the petaloid and pistilloid homeotic anther phenotypes are characteristic of CMS expression in eggplant and other species [36,62,63,64], demonstrating a stable effect of the maternal sterile cytoplasm across all hybrid nuclear backgrounds. The partial anther dehiscence observed in the hybrids, in contrast to the complete indehiscence of their maternal CMS lines, indicates the absence of effective fertility-restorer alleles for this CMS system in the S. integrifolium and S. gilo accessions tested.

Female fertility was also strongly reduced, as revealed by the failure of all backcrosses to both parents, except for a single successful cross of F₁(cmsT × SG) with S. gilo (Figure 9b). Female sterility is commonly observed in interspecific Solanum hybrids due to genomic incompatibilities, but is not a limiting factor for the practical use of these hybrids as rootstocks.

In eggplant, CMS arises by mitochondrial–nuclear conflicts in alloplasmic plants: open reading frames (orfs) related to mitochondrial genes are expressed after nucleus substitution, thus negatively affecting male fertility [65,66]. Because cytoplasmic organelles are maternally inherited in eggplant wide crosses and female fertility typically improves with successive backcross generations [67,68,69], the recovered BC₁ population represents a genetic bridge for transferring the CMS trait into the cultivated Gilo.

4.4. Breeding Implications of CMS-Based Interspecific Hybridization

The results of this study define two breeding pathways: (i) the direct use of CMS systems to produce vigorous interspecific rootstocks and (ii) the introgression of CMS cytoplasm to Solanum gilo and related Solanum germplasm.

4.4.1. CMS-Based Rootstock Breeding

The combination of phenotypic uniformity and remarkable heterosis makes CMS-based interspecific hybridization a practical system that can be exploited for the development of vigorous commercial rootstocks. Given the broad cross-compatibility of eggplant with allied Solanum species, a wide array of CMS-based interspecific rootstocks can potentially be developed. In this context, future work should focus on the evaluation of grafting compatibility of the CMS-based interspecific rootstocks and their effect on scion yield, quality, and stress tolerance.

4.4.2. CMS Introgression into S. gilo and Related Germplasm

The BC₁ population obtained in the present study provides the first practical entry point for transferring the CMS cytoplasm into the cultivated Gilo group through recurrent backcrossing using S. gilo as the recurrent parent. Marker-assisted backcrossing could be used to accelerate the recovery of the S. gilo nuclear genome while retaining the CMS phenotype. The same strategy could be extended to other cultivated Solanum germplasm, including the Kumba and Shum Groups of S. aethiopicum and S. macrocarpon, which currently lack CMS systems.

5. Conclusions

This study evaluated the morphology, heterosis, and fertility in cytoplasmic male sterility (CMS)-based interspecific eggplant hybrids involving Solanum integrifolium and S. gilo from the S. aethiopicum complex. The hybrids exhibited stable, predominantly intermediate morphology combined with enhanced vegetative vigor, while multivariate analysis clearly discriminated parental and hybrid groups. Heterotic responses were not compromised by CMS, supporting its use for the large-scale production of vigorous eggplant rootstocks. The presence of homeotic anther transformations and impaired pollen release ability in all hybrids indicated CMS effects across all nuclear backgrounds. Importantly, the recovery of BC₁ progeny after backcrossing to S. gilo demonstrates the feasibility of CMS introgression within the S. aethiopicum complex. Overall, CMS-based wide hybridization provides an effective strategy for developing uniform, high-vigor interspecific rootstocks and for transferring CMS into cultivated allied species.

Author Contributions

Conceptualization, K.K., A.M., F.B. and D.R.; methodology, K.K., A.M., F.B. and D.R.; formal analysis, K.K. and A.M.; investigation, K.K. and A.M.; data curation, K.K.; writing—original draft preparation, K.K.; writing—review and editing, K.K., A.M., F.B. and D.R.; supervision, K.K. and D.R. All authors have read and agreed to the published version of the manuscript.

Funding

This research received no external funding.

Data Availability Statement

The original contributions presented in this study are included in the article. Further inquiries can be directed to the corresponding authors.

Conflicts of Interest

The authors declare no conflicts of interest.

Appendix A

Appendix A.1. Digital Estimation of Plant Leaf Area

Plant leaf area was digitally estimated using ImageJ software (version 1.50; National Institutes of Health and the Laboratory for Optical and Computational Instrumentation, University of Wisconsin–Madison, Madison, WI, USA) [70]. Each plant was photographed from a height of 150 cm using a Sony DSC-H9 digital camera (Sony Corporation, Tokyo, Japan), with a 12 cm paper strip placed nearby as a scale reference. The photographs, representing the non-shaded leaf area under overhead sunlight, were processed in ImageJ. A 12 cm line was digitally added along the paper strip and set as the image scale, and non-green pixels were excluded using the Threshold function. The remaining green pixels, representing leaf area, were recognized as objects by ImageJ’s ‘ROI Manager’. Area measurements were obtained, converted to cm², and exported in .csv format for subsequent statistical analysis.

Appendix A.2. Digital Measurement of Leaf Blade Dimensions

Leaf blade dimensions were digitally determined following a similar digital procedure described for leaf area (see Appendix A.1). In brief, leaves (excluding the petiole) collected from the middle part of the plant were scanned at a 150 dots per inch (dpi) resolution (equivalent to 59 dots per cm) using an EPSON SX535WD digital scanner (Seiko Epson Corporation, Suwa, Japan). Each leaf was positioned with the midrib aligned vertically, the background was digitally removed, and leaf blade measurements were obtained using the scan resolution as the image scale.

Table A1. Descriptor list of the morphological traits evaluated in seedlings, plants, leaves, flowers, and fruits of the plant material.

Trait	Abbreviation	Units/Scale	Methods	Methods	Authority
Seedling
Hypocotyl anthocyanins	S_HA	1 = Absent; 9 = Present	visual observation	VO ¹	CPVO
Cotyledon length	S_CL	cm	ruler	M
Cotyledon width	S_CW	cm	ruler	M
Cotyledon length/width ratio	S_CLW			C
Leaf midrib color	S_LMC	1 = Green; 9 = Purple; 10 = Vinaceous	visual observation	VO
Leaf midrib anthocyanin intensity (US)	S_LMA_US	1 = Very weak; 9 = Very strong	visual observation	VO
Leaf midrib anthocyanin intensity (LS)	S_LMA_LS	1 = Very weak; 9 = Very strong	visual observation	VO
Leaf prickles presence	S_LPP	1 = Absent; 9 = Present	visual observation	VO
Leaf prickles anthocyanin intensity (US)	S_LPA_US	1 = Very weak; 9 = Very strong	visual observation	VO
Leaf prickles anthocyanin intensity (LS)	S_LPA_LS	1 = Very weak; 9 = Very strong	visual observation	VO
Plant
Growth habit	P_GH	1 = erect; 3 = semi-erect; 7 = horizontal	visual observation	VO	CPVO
Shoot color	P_SC	1 = green; 9 = purple; 10 = vinaceous	visual observation	VO
Shoot intensity of anthocyanins	P_SAI	1 = very weak; 9 = very strong	visual observation	VO	CPVO
Height	P_H	cm	graded stake	M
Estimated leaf area	P_ELA	cm²	ImageJ software	DA
Number of shoots	P_NS		counting	M
Number of leaves	P_NL		counting	M
Internode length	P_INL	cm	ruler	M
Number of prickles per internode	P_NPI		counting	M
Internode prickles color	SPC		visual observation	VO
Number of prickles per internode/cm	P_NPIcm			C
Inflorescence type	P_IT	1 = simple; 2 = compound	visual observation	VO	modified from CPVO
Number of flowers per inflorescence	P_NFInfl		counting	M
Number of fruits per infructescence	P_NFInfr		counting	M
Leaf
Midrib color	L_MC	1 = green; 9 = purple; 10 = vinaceous	visual observation	VO
Blade length	L_L	cm	ImageJ software	DA
Blade width	L_W	cm	ImageJ software	DA
Blade length/width ratio	L_LW			C
Blade area	L_A	cm²	ImageJ software	DA
Margin sinuation	L_SM	1 = absent or very weak; 9 = very strong	visual observation	VO	CPVO
Number of lobes	L_NL		counting	M
Number of prickles (upper surface)	L_PN_US		counting	M
Number of prickles per cm² (upper surface)	L_PNcm_US			C
Number of prickles (lower surface)	L_PN_LS		counting	M
Number of prickles per cm² (lower surface)	L_PNcm_LS			C
Petiole length	L_PL	cm	ruler	M
Petiole diameter	L_PD	cm	caliper	M
Petiole number of prickles	L_PNP		counting
Petiole number of prickles/cm	L_PNPcm
Flower
Corolla color	Fl_CC		visual observation	VO
Corolla stripe	Fl_CS		visual observation	VO
Corolla diameter (cm)	Fl_CD	cm	caliper	M
Stigma color	Fl_SC		visual observation	VO
Number of anthers	Fl_NA
Number of petals	Fl_NP		counting	M
Number of sepals	Fl_NS		counting	M
Number of stigma lobes	Fl_NSL		counting	M
Fruit
Shape	Fr_Sh	0 = flattened; 1 = flattened to globular; 2 = globular; 3 = ovoid; 4 = obovate; 5 = pear shaped; 6 = club-shaped; 7 = ellipsoid; 8 = cylindrical	visual observation	VO	modified from CPVO
Ribs	Fr_R	1 = absent or very weak; 9 = very strong	visual observation	VO	CPVO
Apex	Fr_A	1 = indented; 2 = flattened; 3 = rounded; 4 = acute	visual observation	VO	CPVO
Size of pistil scar	Fr_SPS	1 = very small; 9 = very large	visual observation	VO	CPVO
Depth of indentation of pistil scar	Fr_DIPS	1 = absent or very shallow; 5 = very deep	visual observation	VO	CPVO
Main skin color (HM)	Fr_CHM	1 = white; 2 = green; 3 = purple	visual observation	VO	CPVO
Main skin color (HM)	Fr_CPM	1 = yellow; 2 = brown; 3 = red; 4 = orange	visual observation	VO
Glossiness (HM)	Fr_Gl	3 = weak; 7 = strong	visual observation	VO	CPVO
Density of stripes (HM)	Fr_DStr	3 = sparse; 7 = dense	visual observation	VO	CPVO
Stripes (HM)	Fr_Str	1 = absent; 9 = present	visual observation	VO	CPVO
Prominence of stripes (HM)	Fr_PStr	3 = weak; 7 = strong	visual observation	VO	CPVO
Flesh color (PM)	Fr_FC	1 = whitish; 2 = greenish; 3 = light green; 4 = green; 5 = orange	visual observation	VO	modified from CPVO
Flesh texture (PM)	Fr_FT	1 = gelatinous; 2 = gelatinous to compact; 3 = compact	visual observation	VO
Weight	Fr_W	g	digital scale	M
Peduncle length	Fr_PL	cm	ruler	M
Polar diameter	Fr_PD	cm	caliper	M
Equatorial diameter	Fr_ED	cm	caliper	M
Polar/equatorial diameter ratio	Fr_PDED			C
Number of calyx prickles	Fr_NCP		counting	M
Number of locules	Fr_NL		counting	M

¹ VO = visual observation, M = measurement, C = calculation, DA = digital analysis.

Table A2. ANOVA table obtained from the statistical analysis of the traits evaluated.

Plant Stage/Organ	Trait	Abbreviation	Source of Variation	Sum of Squares	df	Mean Square	F	Sig.
Seedling	Cotyledon length	S_CL	Genotype	29.060	9	3.229	71.319	0.000
			Error	7.697	170	0.045
			Total	36.756	179
	Cotyledon width	S_CW	Genotype	1.590	9	0.177	24.013	0.000
			Error	1.251	170	0.007
			Total	2.841	179
	Cotyledon length/ width ratio	S_CLW	Genotype	26.499	9	2.944	48.451	0.000
			Error	10.270	170	0.061
			Total	36.769	179
Plant	Height	P_H	Genotype	56,562.761	9	6284.751	18.632	0.000
			Error	30,357.858	90	337.310
			Total	86,920.619	99
	Number of shoots	P_NS	Genotype	13,767.048	9	1529.672	14.211	0.000
			Error	9687.600	90	107.640
			Total	23,454.648	99
	Number of leaves	P_NL	Genotype	84,885.294	9	9431.699	16.397	0.000
			Error	51,768.795	90	575.209
			Total	136,654.089	99
	Internode length	P_INL	Genotype	444.537	9	49.393	10.761	0.000
			Error	413.100	90	4.590
			Total	857.637	99
	Number of prickles per internode	P_NPI	Genotype	276.116	9	30.680	25.047	0.000
			Error	110.239	90	1.225
			Total	386.355	99
	Number of prickles per internode/cm	P_NPIcm	Genotype	2.936	9	0.326	37.150	0.000
			Error	0.790	90	0.009
			Total	3.726	99
	Number of flowers per inflorescence	P_NFInfl	Genotype	854.616	9	94.957	57.832	0.000
			Error	147.776	90	1.642
			Total	1002.392	99
Leaf	Blade length	L_L	Genotype	652.412	9	72.490	20.863	0.000
			Error	163.308	50	3.475
			Total	815.720	59
	Blade width	L_W	Genotype	280.148	9	31.128	13.181	0.000
			Error	110.994	50	2.362
			Total	391.142	59
	Blade length/width ratio	L_LW	Genotype	1.721	9	0.191	30.371	0.000
			Error	0.296	50	0.006
			Total	2.016	59
	Blade area	L_A	Genotype	188,765.861	9	20,973.985	10.284	0.000
			Error	95,859.124	50	2039.556
			Total	284,624.985	59
	Number of lobes	L_NL	Genotype	6.750	9	0.750	1.087	0.391
			Error	31.750	50	0.690
			Total	38.500	59
	Number of prickles (upper surface)	L_PN_US	Genotype	984.524	9	109.392	25.756	0.000
			Error	199.617	50	4.247
			Total	1184.140	59
	Number of prickles (lower surface)	L_PN_LS	Genotype	1840.402	9	204.489	54.722	0.000
			Error	175.633	50	3.738
			Total	2016.035	59
	Number of prickles per cm2 (upper surface)	L_PNcm_US	Genotype	0.022	9	0.002	18.997	0.000
			Error	0.006	50	0.0001
			Total	0.028	59
	Number of prickles per cm2 (lower surface)	L_PNcm_LS	Genotype	0.053	9	0.006	36.342	0.000
			Error	0.008	50	0.0002
			Total	0.061	59
	Petiole length	L_PL	Genotype	345.762	9	38.418	23.960	0.000
			Error	75.360	50	1.603
			Total	421.122	59
	Petiole diameter	L_PD	Genotype	22.377	9	2.486	22.196	0.000
			Error	5.619	50	0.112
			Total	27.995	59
	Petiole number of prickles	L_PNP	Genotype	135.752	9	15.084	14.683	0.000
			Error	48.283	50	1.027
			Total	184.035	59
	Petiole number of prickles/cm	L_PNPcm	Genotype	1.852	9	0.206	12.860	0.000
			Error	0.752	50	0.016
			Total	2.604	59
Flower	Corolla diameter (cm)	Fl_CD	Genotype	89.821	9	9.980	175.486	0.000
			Error	5.687	90	0.057
			Total	95.508	99
	Number of anthers	Fl_NA	Genotype	20.896	9	2.322	7.307	0.000
			Error	32.095	90	0.318
			Total	52.991	99
	Number of petals	Fl_NP	Genotype	20.510	9	2.279	7.843	0.000
			Error	29.346	90	0.291
			Total	49.856	99
	Number of sepals	Fl_NS	Genotype	21.635	9	2.404	5.810	0.000
			Error	41.788	90	0.414
			Total	63.423	99
	Number of stigma lobes	Fl_NSL	Genotype	21.461	9	2.385	7.013	0.000
			Error	34.003	90	0.340
			Total	55.464	99

Figure A1. Climatic data recorded during the experimental period. (a) Maximum, mean, and minimum temperatures (°C). (b) Precipitation (mm) and relative humidity (%).

Figure A2. Heatmap based on standardized values of heterosis and heterobeltiosis for nine plant growth traits in five CMS-based eggplant interspecific hybrids with Solanum integrifolium (SI) and S. gilo (SG). Abbreviations: P_H = Plant height, P_INL = Internode length, P_NS = Number of shoots, P_NL = Number of leaves, P_ELA = Estimated leaf area, P_NFI = Number of flowers per inflorescence, L_L = Lead length, L_W = Leaf width, and L_A = Leaf area.

References

FAOSTAT. Statistical Database of the Food and Agriculture Organization of the United Nations. Online Database. Available online: https://www.fao.org/faostat/en/#data/QCL (accessed on 4 August 2022).
Pugalendhi, L.; Veeraragavathatham, D.; Natarjan, S.; Praneetha, S. Utilizing Wild Relative (Solanum viarum) as Resistant Source to Shoot and Fruit Borer in Brinjal (Solanum melongena Linn.). Electron. J. Plant Breed. 2010, 1, 643–648. [Google Scholar]
Liu, J.; Zheng, Z.; Zhou, X.; Feng, C.; Zhuang, Y. Improving the Resistance of Eggplant (Solanum melongena) to Verticillium Wilt Using Wild Species Solanum linnaeanum. Euphytica 2015, 201, 463–469. [Google Scholar] [CrossRef]
Taher, D.; Rakha, M.; Ramasamy, S.; Solberg, S.; Schafleitner, R. Sources of Resistance for Two-Spotted Spider Mite (Tetranychus urticae) in Scarlet (Solanum aethiopicum L.) and Gboma (S. macrocarpon L.) Eggplant Germplasms. HortScience 2019, 54, 240–245. [Google Scholar] [CrossRef]
Bletsos, F.; Roupakias, D.; Tsaktsira, M.; Scaltsoyjannes, A. Production and Characterization of Interspecific Hybrids between Three Eggplant (Solanum melongena L.) Cultivars and Solanum macrocarpon L. Sci. Hortic. 2004, 101, 11–21. [Google Scholar] [CrossRef]
Toppino, L.; Tassone, M.R.; Ribolzi, S.; Sala, T.; Azzimonti, M.T.; Barchi, L.; Luciana, G.; Shaaf, S.; Rossini, L.; Rotino, G.L. Construction of a Living Library of Solanum tomentosum Introgression Lines into the S. melongena Genome: A Tool to Exploit Novel Genetic Diversity for Eggplant Breeding. Hortic. Plant J. 2025, in press. [Google Scholar] [CrossRef]
Toppino, L.; Valè, G.; Rotino, G. Inheritance of Fusarium Wilt Resistance Introgressed from Solanum aethiopicum Gilo and Aculeatum Groups into Cultivated Eggplant (S. melongena) and Development of Associated PCR-Based Markers. Mol. Breed. 2008, 22, 237–250. [Google Scholar] [CrossRef]
Rakha, M.; Namisy, A.; Chen, J.R.; El-Mahrouk, M.E.; Metwally, E.; Taha, N.; Prohens, J.; Plazas, M.; Taher, D. Development of Interspecific Hybrids between a Cultivated Eggplant Resistant to Bacterial Wilt (Ralstonia solanacearum) and Eggplant Wild Relatives for the Development of Rootstocks. Plants 2020, 9, 1405. [Google Scholar] [CrossRef] [PubMed]
Cebeci, E. Comparative Evaluation of Eggplant Genotypes with Their Wild Relatives under Gradually Increased Drought Stress. Bragantia 2024, 83, e20230246. [Google Scholar] [CrossRef]
Cebeci, E.; Boyacı, H.; Doğan, Y.; Toppıno, L.; Rotino, G. Determination of Heat and Drought Tolerant Lines in Segregating Populations Produced by Interspecific Crosses in Eggplant. Ekin J. Crop Breed. Genet. 2023, 9, 81–90. [Google Scholar]
Krommydas, K.; Papa, E.; Gaitani, P.; Papadopoulou, A.; Mellidou, I.; Bouloumpasi, E.; Kadoglidou, K.I. Comparative Drought Response of Solanum melongena, S. macrocarpon, S. dasyphyllum, and S. melongena × S. dasyphyllum Interspecific Hybrids. Agronomy 2025, 15, 2516. [Google Scholar] [CrossRef]
Ortega-Albero, N.; González-Orenga, S.; Vicente, O.; Rodríguez-Burruezo, A.; Fita, A. Responses to Salt Stress of the Interspecific Hybrid Solanum insanum × Solanum melongena and Its Parental Species. Plants 2023, 12, 295. [Google Scholar] [CrossRef] [PubMed]
Cebeci, E.; Boyaci, H.; Kıran, S.; Ellialtioglu, S. Assessment Results of Salinity Stressed F2 Population Originated from Interspecific Hybridization of Eggplant with Wild Relative S. incanum L. Hortic. Stud. 2024, 41, 50–59. [Google Scholar] [CrossRef]
Villanueva, G.; Plazas, M.; Gramazio, P.; Moya, R.D.; Prohens, J.; Vilanova, S. Evaluation of Three Sets of Advanced Backcrosses of Eggplant with Wild Relatives from Different Gene Pools under Low N Fertilization Conditions. Hortic. Res. 2023, 10, uhad141. [Google Scholar] [CrossRef]
Kaushik, P.; Gramazio, P.; Vilanova, S.; Raigón, M.D.; Prohens, J.; Plazas, M. Phenolics Content, Fruit Flesh Colour and Browning in Cultivated Eggplant, Wild Relatives and Interspecific Hybrids and Implications for Fruit Quality Breeding. Food Res. Int. 2017, 102, 392–401. [Google Scholar] [CrossRef]
Mennella, G.; Rotino, G.L.; Fibiani, M.; D’Alessandro, A.; Francese, G.; Toppino, L.; Cavallanti, F.; Acciarri, N.; Lo Scalzo, R. Characterization of Health-Related Compounds in Eggplant (Solanum melongena L.) Lines Derived from Introgression of Allied Species. J. Agric. Food Chem. 2010, 58, 7597–7603. [Google Scholar] [CrossRef] [PubMed]
Gramazio, P.; Prohens, J.; Plazas, M.; Herraiz, F.J.; Castillo, E.; Knapp, S.; Meyer, R.S.; Vilanova, S. Location of Chlorogenic Acid Biosynthesis Pathway and Polyphenol Oxidase Genes in a New Interspecific Anchored Linkage Map of Eggplant. BMC Plant Biol. 2014, 14, 350. [Google Scholar] [CrossRef]
Rosa-Martínez, E.; Adalid-Martínez, A.M.; García-Martínez, M.D.; Mangino, G.; Raigón, M.D.; Plazas, M.; Gramazio, P.; Prohens, J.; Vilanova, S. Fruit Composition of Eggplant Lines with Introgressions from the Wild Relative S. incanum: Interest for Breeding and Safety for Consumption. Agronomy 2022, 12, 266. [Google Scholar] [CrossRef]
Bletsos, F.A.; Olympios, C.M. Rootstocks and Grafting of Tomatoes, Peppers and Eggplants for Soil-Borne Disease Resistance, Improved Yield and Quality. Eur. J. Plant Sci. Biotechnol. 2008, 2, 62–73. [Google Scholar]
Khan, M.M.R.; Isshiki, S. Cytoplasmic Male Sterility in Eggplant. Hortic. J. 2016, 85, 1–7. [Google Scholar] [CrossRef]
Doganlar, S.; Frary, A.; Daunay, M. A Comparative Genetic Linkage Map of Eggplant (Solanum melongena) and Its Implications for Genome Evolution in the Solanaceae. Genetics 2002, 161, 1697–1711. [Google Scholar] [CrossRef]
Doganlar, S.; Frary, A.; Daunay, M.C.; Huvenaars, K.; Mank, R.; Frary, A. High Resolution Map of Eggplant (Solanum melongena) Reveals Extensive Chromosome Rearrangement in Domesticated Members of the Solanaceae. Euphytica 2014, 198, 231–241. [Google Scholar] [CrossRef]
Toppino, L.; Ribolzi, S.; Shaaf, S.; Bassolino, L.; Carletti, G.; Fadda, S.; Rossini, L.; Boyaci, H.F.; Caliskan, S.; Unlu, A.; et al. Development of an Introgression Lines Population and Genetic Mapping of Novel Traits Linked to Key Breeding Traits in Eggplant. In Proceedings of the LXII SIGA Annual Congress, Verona, Italy, 25–28 September 2018. [Google Scholar]
Mangino, G.; Plazas, M.; Vilanova, S.; Prohens, J.; Gramazio, P. Performance of a Set of Eggplant (Solanum melongena) Lines with Introgressions from Its Wild Relative S. incanum under Open Field and Screenhouse Conditions and Detection of QTLs. Agronomy 2020, 10, 467. [Google Scholar] [CrossRef]
Gisbert, C.; Prohens, J.; Nuez, F. Performance of Eggplant Grafted onto Cultivated, Wild and Hybrid Materials of Eggplant and Tomato. Int. J. Plant Prod. 2011, 5, 367–380. [Google Scholar]
Krommydas, K.; Mavromatis, A.; Bletsos, F.; Roupakias, D. Suitability of CMS-Based Interspecific Eggplant (Solanum melongena L.) Hybrids as Rootstocks for Eggplant Grafting. J. Agric. Ecol. Res. Int. 2018, 15, 1–15. [Google Scholar] [CrossRef]
Sabatino, L.; Iapichino, G.; Rotino, L.G.; Palazzolo, E.; Mennella, G.; D’Anna, F. Solanum aethiopicum Gr. Gilo and Its Interspecific Hybrid with S. melongena as Alternative Rootstocks for Eggplant: Effects on Vigor, Yield, and Fruit Physicochemical Properties of Cultivar ‘Scarlatti’. Agronomy 2019, 9, 223. [Google Scholar] [CrossRef]
Ibrahim, M.; Munira, M.K.; Kabir, M.S.; Islam, A.K.M.S.; Miah, M.M.U. Seed Germination and Graft Compatibility of Wild Solanum as Rootstock of Tomato. J. Biol. Sci. 2001, 1, 701–703. [Google Scholar] [CrossRef][Green Version]
Bharathi, S.; Pugalendhi, L.; Priya, R.S.; Uma, D.; Tamilselvi, N.A. Grafting Studies of Tomato with Wild Solanum Rootstocks. J. Pharmacogn. Phytochem. 2021, 10, 2210–2213. [Google Scholar] [CrossRef]
Ranil, R.H.G.; Niran, H.M.L.; Plazas, M.; Fonseka, R.M.; Fonseka, H.H.; Vilanova, S.; Andújar, I.; Gramazio, P.; Fita, A.; Prohens, J. Improving Seed Germination of the Eggplant Rootstock Solanum torvum by Testing Multiple Factors Using an Orthogonal Array Design. Sci. Hortic. 2015, 193, 174–181. [Google Scholar] [CrossRef]
Musa, I.; Rafii, M.Y.; Ahmad, K.; Ramlee, S.I.; Hatta, M.A.M.; Oladosu, Y.; Muhammad, I.; Chukwu, S.C.; Sulaiman, N.N.M.; Ayanda, A.F.; et al. Effects of Grafting on Morphophysiological and Yield Characteristic of Eggplant (Solanum melongena L.) Grafted onto Wild Relative Rootstocks. Plants 2020, 9, 1583. [Google Scholar] [CrossRef]
Petran, A.; Hoover, E. Solanum Torvum as a Compatible Rootstock in Interspecific Tomato Grafting. J. Hortic. 2014, 1, 103. [Google Scholar] [CrossRef]
Isshiki, S.; Kawajiri, N. Effect of Cytoplasm of Solanum Violaceum Ort. on Fertility of Eggplant (S. melongena L.). Sci. Hortic. 2002, 93, 9–18. [Google Scholar] [CrossRef]
Hasnunnahar, M.; Khan, M.R.; Isshiki, S. Inheritance Analysis of Fertility Restoration Genes (Rf) in a Male Sterile System of Eggplant Using Cytoplasm of Solanum Grandifolium. Aust. J. Crop Sci. 2012, 6, 475–479. [Google Scholar]
Fang, M.; Mao, R.; Xie, W. Breeding of Cytoplasmically Inherited Male Sterile Lines of Eggplant (Solanum melongena L.). Acta Hortic. Sin. 1985, 12, 261–266. [Google Scholar]
Krommydas, K.S.; Tzikalios, Z.; Madesis, P.; Bletsos, F.A.; Mavromatis, A.; Roupakias, D. Development and Fertility Restoration of CMS Eggplant Lines Carrying the Cytoplasm of Solanum violaceum. J. Agric. Sci. 2016, 8, 10–26. [Google Scholar] [CrossRef]
Han, M.; Opoku, K.N.; Bissah, N.A.B.; Su, T.; Mauro, P.; Nicoletto, C.; Sabatino, L. Solanum aethiopicum: The Nutrient-Rich Vegetable Crop with Great Economic, Genetic Biodiversity and Pharmaceutical Potential. Horticulturae 2021, 7, 126. [Google Scholar] [CrossRef]
Daunay, M.-C. Eggplant. In Vegetables II SE—5; Prohens, J., Nuez, F., Eds.; Handbook of Plant Breeding; Springer: New York, NY, USA, 2008; Volume 2, pp. 163–220. ISBN 978-0-387-74108-6. [Google Scholar]
Ali, M.; Fujieda, K. Cross Compatibility between Eggplant (Solanum melongena L.) and Wild Relatives. J. Jpn. Soc. Hortic. Sci. 1990, 58, 977–984. [Google Scholar] [CrossRef][Green Version]
Premabati Devi, C.; Munshi, A.D.; Behera, T.K.; Choudhary, H.; Vinod; Gurung, B.; Saha, P. Cross Compatibility in Interspecific Hybridization of Eggplant, Solanum melongena, with Its Wild Relatives. Sci. Hortic. 2015, 193, 353–358. [Google Scholar] [CrossRef]
CPVO Protocol for Distinctness, Uniformity and Stability Tests. Available online: https://cpvo.europa.eu/en/applications-and-examinations/technical-examinations/technical-protocols/cpvo-technical-protocols?page=3 (accessed on 5 December 2025).
Fehr, W.R. Principles of Cultivar Development: Theory and Technique; Fehr, W.R., Fehr, E.L., Jessen, H.J., Eds.; Macmillan: New York, NY, USA, 1987. [Google Scholar]
Bohra, A.; Tiwari, A.; Pareek, S.; Joshi, R.; Satheesh Naik, S.J.; Kumari, K.; Verma, R.L.; Parihar, A.K.; Patil, P.G.; Dixit, G.P. Past and Future of Cytoplasmic Male Sterility and Heterosis Breeding in Crop Plants. Plant Cell Rep. 2025, 44, 33. [Google Scholar] [CrossRef]
Kaushik, P.; Prohens, J.; Vilanova, S.; Gramazio, P.; Plazas, M. Phenotyping of Eggplant Wild Relatives and Interspecific Hybrids with Conventional and Phenomics Descriptors Provides Insight for Their Potential Utilization in Breeding. Front. Plant Sci. 2016, 7, 677. [Google Scholar] [CrossRef]
Ghani, M.A.; Abbas, M.M.; Ziaf, K.; Azam, M.; Ali, B.; Amjad, M.; Anjum, R.; Noor, A.; Zahid, M. Production and Characterization of Inter and Intraspecific Hybridization Eggplant. Hortic. Bras. 2020, 38, 407–414. [Google Scholar] [CrossRef]
Prohens, J.; Plazas, M.; Raigón, M.D.; Seguí-Simarro, J.M.; Stommel, J.R.; Vilanova, S. Characterization of Interspecific Hybrids and First Backcross Generations from Crosses between Two Cultivated Eggplants (Solanum melongena and S. aethiopicum Kumba Group) and Implications for Eggplant Breeding. Euphytica 2012, 186, 517–538. [Google Scholar] [CrossRef]
Tigchelaar, E.C.; Janick, J.; Erickson, H.T. The Genetics of Anthocyanin Coloration in Eggplant (Solanum melongena L.). Genetics 1968, 60, 475–491. [Google Scholar] [CrossRef]
Honda, T.; Zushi, K.; Matsuzqe, N. Inheritance of Anthocyanin Pigment and Photosensitivity in Eggplant (Solanum melongena) Fruit. Environ. Control Biol. 2012, 50, 75–80. [Google Scholar] [CrossRef][Green Version]
Zhang, L.; Zhang, R.; Yan, P.; Zeng, L.; Zhao, W.; Feng, H.; Mu, R.; Hou, W. PE (Prickly Eggplant) Encoding a Cytokinin-Activating Enzyme Responsible for the Formation of Prickles in Eggplant. Hortic. Res. 2024, 11, uhae134. [Google Scholar] [CrossRef] [PubMed]
Li, S.; He, Y.; Li, D.; Shi, S.; Wang, Y.; Tang, X.; Ge, H.; Liu, Y.; Chen, H. Fine Mapping an AUXIN RESPONSE FACTOR, SmARF18, as a Candidate Gene of the PRICKLE LOCUS That Controls Prickle Absence/Presence on Various Organs in Eggplant (Solanum melongena L.). Sci. Hortic. 2024, 327, 112874. [Google Scholar] [CrossRef]
Miyatake, K.; Saito, T.; Nunome, T.; Yamaguchi, H.; Negoro, S.; Ohyama, A.; Wu, J.; Katayose, Y.; Fukuoka, H. Fine Mapping of a Major Locus Representing the Lack of Prickles in Eggplant Revealed the Availability of a 0.5-Kb Insertion/Deletion for Marker-Assisted Selection. Breed. Sci. 2020, 70, 438–448. [Google Scholar] [CrossRef] [PubMed]
Borgato, L.; Conicella, C.; Pisani, F.; Furini, A. Production and Characterization of Arboreous and Fertile Solanum melongena + Solanum marginatum Somatic Hybrid Plants. Planta 2007, 226, 961–969. [Google Scholar] [CrossRef] [PubMed]
Kouassi, K.B.A.; N’dri, K.E.; Asseh, E.E.; Kouassi, A.B.; Koffi, E.-B.Z.; Yao, S.D.M. Heterosis in Interspecific Hybrids between Solanum melongena L. and Species from the Primary and Secondary Gene Pools. Afrique Sci. 2024, 25, 1–14. [Google Scholar]
Kumar, A.; Sharma, V.; Jain, B.T.; Kaushik, P.; Kumar, A.; Sharma, V.; Jain, B.T.; Kaushik, P. Heterosis Breeding in Eggplant (Solanum melongena L.): Gains and Provocations. Plants 2020, 9, 403. [Google Scholar] [CrossRef]
Kirti, P.B.; Rao, B.G.S. Cytological Studies on F₁ Hybrids of Solanum Integrifolium with S. melongena and S. melongena Var. insanum. Genetica 1982, 59, 127–131. [Google Scholar] [CrossRef]
Daunay, M.-C.; Salinier, J.; Aubriot, X. Crossability and Diversity of Eggplants and Their Wild Relatives. In The Eggplant Genome. Compendium of Plant Genomes; Springer: Cham, Switzerland, 2019; ISBN 9783319992082. [Google Scholar]
Kirti, P.B.; Rao, B.G.S. Chromosome Relationships of Spinous Solanums. Proc. Indian Acad. Sci.-Sect. B Part 3 Plant Sci. 1982, 91, 83–91. [Google Scholar] [CrossRef]
Kirti, P.B.; Rao, B.G.S. Chromosome Pairing in Reciprocal Hybrids of Solanum integrifolium and S. indicum Var. Multiflora. Caryologia 1980, 33, 289–294. [Google Scholar] [CrossRef]
Schaff, D.; Jelenkovic, G. Hybridization and Fertility of Hybrid Derivatives of Solanum melongena L. and Solanum macrocarpon L. Theor. Appl. Genet. 1982, 153, 149–153. [Google Scholar] [CrossRef]
Kumchai, J.; Wei, Y.-C.; Lee, C.-Y.; Chen, F.-C.; Chin, S.-W. Production of Interspecific Hybrids between Commercial Cultivars of the Eggplant (Solanum melongena L.) and Its Wild Relative S. torvum. Genet. Mol. Res. 2013, 12, 755–764. [Google Scholar] [CrossRef]
Isshiki, S.; Taura, T. Fertility Restoration of Hybrids between Solanum melongena L. and S. aethiopicum L. Gilo Group by Chromosome Doubling and Cytoplasmic Effect on Pollen Fertility. Euphytica 2003, 131, 195–201. [Google Scholar] [CrossRef]
Linke, B.; Börner, T. Mitochondrial Effects on Flower and Pollen Development. Mitochondrion 2005, 5, 389–402. [Google Scholar] [CrossRef]
Farbos, I.; Mouras, A.; Bereterbide, A.; Glimelius, K. Defective Cell Proliferation in the Floral Meristem of Alloplasmic Plants of Nicotiana tabacum Leads to Abnormal Floral Organ Development and Male Sterility. Plant J. 2001, 26, 131–142. [Google Scholar] [CrossRef] [PubMed]
Liu, B.; Ou, C.; Chen, S.; Cao, Q.; Zhao, Z.; Miao, Z.; Kong, X.; Zhuang, F. Differentially Expressed Genes between Carrot Petaloid Cytoplasmic Male Sterile and Maintainer during Floral Development. Sci. Rep. 2019, 9, 17384. [Google Scholar] [CrossRef]
Deng, W.; Gan, G.; Li, W.; Yu, C.; Jiang, Y.; Li, D.; Yang, Q.; Li, W.; Wang, P.; Wang, Y. Comparative Analysis of the Mitochondrial Genome of Eggplant (Solanum melongena L.) to Identify Cytoplasmic Male Sterility Candidate Genes. Int. J. Mol. Sci. 2024, 25, 9743. [Google Scholar] [CrossRef]
Yoshimi, M.; Kitamura, Y.; Isshiki, S.; Saito, T.; Yasumoto, K.; Terachi, T.; Yamagishi, H. Variations in the Structure and Transcription of the Mitochondrial Atp and Cox Genes in Wild Solanum Species That Induce Male Sterility in Eggplant (S. melongena). Theor. Appl. Genet. 2013, 126, 1851–1859. [Google Scholar] [CrossRef] [PubMed]
Khan, M.M.R.; Isshiki, S. Development of a Cytoplasmic Male-Sterile Line of Eggplant (Solanum melongena L.) with the Cytoplasm of Solanum anguivi. Plant Breed. 2011, 130, 256–260. [Google Scholar] [CrossRef]
Khan, M.M.R.; Isshiki, S. Development of the Male-Sterile Line of Eggplant Utilizing the Cytoplasm of Solanum aethiopicum L. Aculeatum Group. J. Jpn. Soc. Hortic. Sci. 2010, 79, 348–353. [Google Scholar] [CrossRef]
Khan, M.M.R.; Isshiki, S. Development of a Male Sterile Eggplant by Utilizing the Cytoplasm of Solanum virginianum and a Biparental Transmission of Chloroplast DNA in Backcrossing. Sci. Hortic. 2008, 117, 316–320. [Google Scholar] [CrossRef]
Schneider, C.A.; Rasband, W.S.; Eliceiri, K.W. NIH Image to ImageJ: 25 Years of Image Analysis. Nat. Methods 2012, 9, 671–675. [Google Scholar] [CrossRef] [PubMed]

Figure 1. General schematic representation of the crossing scheme used to develop the CMS-based eggplant interspecific hybrids and their expected nuclear and cytoplasmic compositions. Alloplasmic CMS Solanum melongena (SM) lines carrying the S. violaceum cytoplasm were used as female parents and crossed with the allied species S. integrifolium (a) and S. gilo (b), producing interspecific hybrids with hybrid nuclear genomes and a common CMS cytoplasm.

Figure 2. Seedling morphological characterization. Hypocotyl coloration in seedlings of eggplant cv. ‘Langada’ (L), F₁(cmsL × SI), and Solanum integrifolium (SI) (a–c). Hypocotyl coloration in seedlings of eggplant cv. ‘Emi’ (E), F₁(cmsE × SG), and S. gilo (SG) (d–f). Presence of prickles and anthocyanin prickles on the upper and lower leaf surfaces of S. integrifolium (g,h), and (i) on the lower leaf surface of F₁(cmsE × SI), respectively.

Figure 3. Plant morphology. Top row: Solanum gilo (SG), F₁(cmsT × SG), and eggplant cv. ’Tsakoniki’ (T). Middle row: F₁(cmsL × SI), eggplant cv. ‘Langada’ (L) and S. integrifolium (SI). Bottom row: Root systems of eggplant cv. ‘Langada’, F₁(cmsL × SI), S. integrifolium, cv. ‘Tsakoniki’, F₁(cmsT × SG) and S. gilo.

Figure 4. Floral morphology of the examined plant material. Top row: Flowers of Solanum integrifolium (SI) (a), S. gilo (SG) (b), F₁(cmsL × SI) (c), and eggplant cv. ‘Langada’ (L) (d). Middle row: Flowers of F₁(cmsE × SI) (e), F₁(cmsE × SG) (f), and eggplant cv. ‘Emi’ (E) (g). Bottom row: Flowers of F₁(cmsT × SI) (h), F₁(cmsT × SG) (i), and eggplant cv. ‘Tsakoniki’ (T) (j). (scale bar = 1 cm, same scale for all photographs).

Figure 5. Fruit morphology. Immature fruits of S. integrifolium (SI) and S. gilo (SG) (a,b), and parthenocarpic fruits F₁(cmsE × SI) and F₁(cmsE × SG) (c,d). Mature fruit of eggplant cv. ‘Emi’ (E) (left), and parthenocarpic fruit of F₁(cmsE × SI) (center) and fruit S. integrifolium (right) (e). Fruits at physiological maturity of eggplant cv. ‘Tsakoniki’ (T) (left), seeded fruit of F₁(cmsT × SG) (center top), parthenocarpic fruit of F₁(cmsT × SG) (center bottom), and fruit of S. gilo (right) (f). White scale bar = 5 cm.

Figure 6. Loading plot of traits (a) and score plot of genotypes (b) from the principal component analysis of 47 seedling, plant, leaf, and flower traits in S. melongena cultivars ‘Langada’ (L), ‘Emi’ (E), and ‘Tsakoniki’ (T), as well as S. integrifolium (SI), S. gilo (SG), and their CMS-based interspecific hybrids. The plots were based on the first and second principal components (PC1 and PC2), explaining 37.7% and 30.5% of variation, respectively.

Figure 7. Mid-parent heterosis (MPH) and better-parent heterosis (BPH) for plant height (a), internode length (b), number of shoots (c), number of leaves (d), estimated leaf area (e), number of flowers per inflorescence (f), leaf blade length (g), leaf blade width (h), and leaf blade area (i) in five F₁ hybrids (cmsL × SI, cmsE × SI, cmsT × SI, cmsE × SG, and cmsT × SG). Values are given as percentages. Significant and non-significant differences between the hybrid and the corresponding mid-parent value for MPH or better-parent value for BPH are indicated by * and ns, respectively, according to the t-test (p < 0.05).

Figure 8. Anther morphology and pollen viability assessment in Solanum integrifolium (SI) and S. gilo (SG), as well as the eggplant cultivars ‘Langada’ (L), ‘Emi’ (E), and ‘Tsakoniki’ (T), in addition to their respective CMS-based interspecific hybrids. Top row: Normal (dehisced) anther pore in S. integrifolium (a), closed pore covered by hair-like structures in F₁(cmsT × SI) (b), and petaloid anthers in F₁(cmsE × SI) (c). Middle row: Normal pore in S. gilo (d), partially opened pores with constriction in F₁(cmsE × SG) (e), and anthers showing petaloidy and pistilody in F₁(cmsT × SG) (f). Bottom row: Microphotographs of pollen grains stained with 2% acetocarmine solution in eggplant cv. ‘Emi’ (g), F₁(cmsE × SG) (h), and S. gilo (i). Blue and orange arrows indicate viable and non-viable pollen grains, respectively.

Figure 9. Pollen viability (%) in Solanum integrifolium (SI) and S. gilo (SG), the eggplant cultivars ‘Langada’ (L), ‘Emi’ (E), and ‘Tsakoniki’ (T), as well as their respective CMS-based interspecific hybrids. Means followed by different letters differ significantly at p ≤ 0.05 according to Duncan’s multiple range test (a). Percentage (%) of successful artificial self-pollination in the parental lines (n = 10) and backcrosses of the interspecific hybrids (n = 30) (b).

Table 1. Seedling morphological traits in Solanum integrifolium (SI), S. gilo (SG), the eggplant cultivars ‘Langada’ (L), ‘Emi’ (E), and ‘Tsakoniki’ (T), and their respective CMS-based interspecific hybrids.

Trait	SI	SG	L	F₁(cmsL × SI)	E	F₁(cmsE × SI)	F₁(cmsE × SG)	T	F₁(cmsT × SI)	F₁(cmsT × SG)
Hypocotyl anthocyanins	absent	absent	present	present	present	present	present	present	present	present
Cotyledon
Length (cm)	2.37 d ¹	2.24 d	3.41 a	2.84 c	3.38 a	2.89 c	3.22 b	2.91 c	3.41 a	2.86 c
Width (cm)	1.20 a	0.91 d	0.94 d	1.06 c	1.04 c	1.06 c	1.18 a,b	1.02 c	1.13 b	1.18 a,b
Length/width ratio	1.98 g	2.50 f	3.66 a	2.67 e	3.27 b	2.73 d,e	2.74 d,e	2.89 c,d	3.04 c	2.86 c,d
Leaf midrib
Color	vinaceous	green	purple	purple	purple	purple	purple	purple	purple	purple
Anthocyanin intensity (US) ²	medium	absent	medium	medium	medium	medium	medium	medium	medium	medium
Anthocyanin intensity (LS)	absent	absent	medium	medium	medium	medium	medium	medium	medium	medium
Leaf prickles
Presence	present	absent	absent	present	absent	present	absent	absent	present	absent
Anthocyanin intensity (US)	medium	-	-	very strong	-	very strong	-	-	very strong	-
Anthocyanin intensity (LS)	absent	-	-	medium	-	medium	-	-	medium	-

¹ Different letters within each row indicate significant differences between genotypes, according to Tukey’s test at p < 0.05. ² US = upper surface; LS = lower surface.

Table 2. Plant morphological traits in Solanum integrifolium (SI), S. gilo (SG), the eggplant cultivars ‘Langada’ (L), ‘Emi’ (E), and ‘Tsakoniki’ (T), and their respective CMS-based interspecific hybrids.

Trait	SI	SG	L	F₁(cmsL × SI)	E	F₁(cmsE × xSI)	F₁(cmsE × SG)	T	F₁(cmsT × SI)	F₁(cmsT × SG)
Plant
Growth habit	semi-erect	semi-erect	erect	semi-erect	semi-erect	semi-erect	horizontal	erect	semi-erect	horizontal
Shoot color	vinaceous	green	purple	purple	purple	purple	purple	purple	purple	purple
Shoot intensity of anthocyanins	medium	absence	medium	very strong	medium	very strong	medium	medium	very strong	medium
Height (cm)	74.0 d ¹	70.0 d	95.17 b,c	131.67 a	78.75 c,d	119.0 a	101.60 b	86.0 b–d	133.57 a	102.0 b
Estimated leaf area (cm²)	942.36 b	1116.57 b	1388.84 b	2984.01 a	1202.08 b	2381.91 a	2709.74 a	1066.02 b	2880.01 a	2838.60 a
Number of shoots	6.83 c	11.40 c	9.17 c	30.33 b	9.0 c	22.50 b	25.80 b	8.67 c	27.29 b	41.80 a
Number of leaves	26.0 c	42.60 c	39.67 c	93.33 a,b	35.0 c	80.0 b	76.80 b	36.67 c	87.0 a,b	107.0 a
Internode
Length (cm)	9.0 d	8.90 d	13.0 a–c	14.33 a,b	11.75 c	12.33 b,c	13.0 a–c	11.0 c	15.0 a	12.90 b,c
Number of prickles	4.33 a	0.0 d	0.0 d	2.83 b	0.0 d	1.60 c	0.0 d	0.0 d	2.50 bc	0.0 d
Prickle color	vinaceous	-	-	purple	-	purple	-	-	purple	-
Number of prickles per cm	0.48 a	0.0 d	0.0 d	0.20 b	0.0 d	0.13 c	0.0 d	0.0 d	0.17 c	0.0 d
Inflorescence
Type	Compound	Simple	Simple	Compound	Simple	Compound	Compound	Simple	Compound	Compound
Number of flowers	9.60 a	1.67 e	1.44 e	7.60 b	2.08 e	6.56 b,c	6.33 c	1.33 e	6.71 b,c	5.0 d
Number of fruits	1.55 c	1.42 c	1.00 c	-	1.25 c	3.73 a	2.50 b	1.00 c	-	-

¹ Different letters within each row indicate significant differences between genotypes according to Tukey’s test at p < 0.05.

Table 3. Leaf morphological traits in Solanum integrifolium (SI), S. gilo (SG), the eggplant cultivars ‘Langada’ (L), ‘Emi’ (E), and ‘Tsakoniki’ (T), and their respective CMS-based interspecific hybrids.

Trait	SI	SG	L	F₁(cmsL × SI)	E	F₁(cmsE × SI)	F₁(cmsE × SG)	T	F₁(cmsT × SI)	F₁(cmsT × SG)
Blade
Midrib color	vinaceous	green	purple	purple	purple	purple	purple	purple	purple	purple
Length (cm)	16.77 c ¹	16.17 c	26.16 a	24.74 a,b	22.39 b	26.63 a	23.25 b	22.36 b	24.47 a,b	23.12 b
Width (cm)	15.67 b	15.60 b	15.58 b	20.44 a	15.78 b	20.70 a	18.74 a	14.88 b	20.23 a	18.65 a
Length/width ratio	1.07 d	1.04 d	1.68 a	1.21 c	1.42 b	1.29 c	1.24 c	1.51 b	1.21 c	1.24 c
Area (cm²)	179.88 e,f	168.31 f	259.79 b–d	314.51 a,b	230.32 c–e	330.90 a	284.55 a–c	216.99 d-f	331.06 a	283.86 a–c
Sinuation of margin	very strong	very strong	absent or very weak	very strong	absent or very weak	very strong	very strong	absent or very weak	very strong	very strong
Number of lobes	8.60 ns	8.50 ns	8.25 ns	8.00 ns	8.00 ns	8.67 ns	7.67 ns	8.80 ns	8.00 ns	8.17 ns
Blade prickles
Number (US ²)	11.00 a	0.0 d	0.75 d	8.00 b	4.83 c	9.0 a,b	0.0 d	1.40 d	7.83 b	0.0 d
Number per cm² (US)	0.06 a	0.0 c	0.0 c	0.03 b	0.02 b	0.03 b	0.0 c	0.01 c	0.02 b	0.0 c
Number (LS)	17.50 a	0.0 d	0.50 d	7.17 c	1.33 d	9.67 b	0.0 d	0.20 d	7.17 c	0.0 d
Number per cm² (LS)	0.10 a	0.0 c	0.0 c	0.02 b	0.01 c	0.03 b	0.0 c	0.0 c	0.02 b	0.0 c
Petiole
Length (cm)	5.25 e	5.38 d,e	9.45 b,c	12.08 a	4.75 e	9.42 b,c	8.67 c	6.90 d	10.83 a,b	9.92 b,c
Diameter (cm)	0.55 d	0.62 c,d	0.88 a	0.83 a,b	0.83 a,b	0.83 a,b	0.82 a,b	0.70 b,c	0.78 a,b	0.82 a,b
Number of prickles	2.67 b	0.0 c	0.75 c	3.33 a,b	0.83 c	4.17 a	0.0 c	0.40 c	3.00 a,b	0.0 c
Number of prickles per cm	0.50 a	0.0 c	0.07 c	0.28 b	0.16 b,c	0.45 a	0.0 c	0.06 c	0.28 b	0.0 c

¹ Different letters within each row indicate significant differences between genotypes according to Tukey’s test at p < 0.05. ² US = upper surface and LS = lower surface.

Table 4. Flower morphological traits in Solanum integrifolium (SI), S. gilo (SG), the eggplant cultivars ‘Langada’ (L), ‘Emi’ (E), and ‘Tsakoniki’ (T), and their respective CMS-based interspecific hybrids.

Trait	SI	SG	L	F₁(cmsL × SI)	E	F₁(cmsE × SI)	F₁(cmsE × SG)	T	F₁(cmsT × SI)	F₁(cmsT × SG)
Corolla
Color	white	white	medium purple	white	medium purple	white	white	light purple	white	white
Stripe	absent	absent	present	present	present	present	present	present	present	present
Diameter (cm)	2.26 e ¹	2.29 e	5.26 a,b	3.00 c	5.48 a	2.72 d	3.05 c	5.20 b	2.73 d	2.93 c,d
Stigma color	light orange	light orange	green	light green	green	light green	light green	light green	light green	light green
Number of
Anthers	6.47 b	6.00 b,c	6.20 b,c	5.86 b–d	7.50 a	6.0 bc	5.73 c,d	6.40 b	5.77 c,d	5.36 d
Petals	6.42 b	6.14 b,c	6.00 b,c	5.76 c,d	7.50 a	6.0 b,c	5.73 c,d	6.20 b,c	5.69 c,d	5.36 d
Sepals	6.47 b,c	6.14 b,c	6.40 b,c	5.86 c,d	7.50 a	6.18 b,c	5.87 c,d	6.60 b	5.77 c,d	5.36 d
Stigma lobes	3.47 c,d	3.29 d	4.75 a	4.62 a–c	4.50 a,b	4.09 a,b	4.33 a,b	4.40 a,b	4.15 a,b	4.00 b,c

¹ Different letters within each row indicate significant differences between genotypes according to Tukey’s test at p < 0.05.

Table 5. Fruit morphological traits of S. integrifolium (SI), S. gilo (SG), the eggplant cultivar ‘Emi’ (E), and their respective CMS-based interspecific hybrids.

Trait	SI	SG	Ε	F₁(cmsE × SI)	F₁(cmsE × SG)
Shape	flattened	flattened	obovate	flattened to globular	flattened to globular
Ribs	very strong	strong	absent	medium	medium
Apex	indented	indented to flattened	flattened	indented to flattened	indented to flattened
Size of pistil scar	very large	very large	medium	very small	very small
Depth of indentation of pistil scar	very deep	deep	absent or very shallow	shallow	shallow
Main skin color (HM ¹)	green	green	purple	green	green
Main skin color (PM)	red	red	brown	orange	orange
Glossiness (HM)	strong	medium	medium to strong	strong	strong
Presence of stripes (HM)	present	present	absent	present	present
Prominence of stripes	weak	weak	-	strong	strong
Density of stripes	sparse	sparse	-	medium	medium
Flesh color (PM)	white	orange	white	green	light green
Flesh texture (PM)	gelatinous to compact	gelatinous	compact	gelatinous to compact	gelatinous
Weight (g)	37.50 b ²	35.40 b	436.77 a	15.85 c	19.87 c
Peduncle length (cm)	1.68 c	1.44 c	8.56 a	2.98 b	2.77 b
Polar diameter (cm)	2.87 b	2.60 c	15.42 a	2.58 c	2.94 b
Equatorial diameter (cm)	5.20 b	4.80 c	8.68 a	3.57 d	3.81 d
Polar/equatorial diameter	0.55 d	0.54 d	1.78 a	0.72 c	0.79 b
Number of calyx prickles	1.83 c	0.0 c	3.70 b	6.69 a	0.0 c
Number of locules	6.0 ns	5.83 ns	4.90 ns	5.60 ns	5.30 ns

¹ HM = Harvest maturity; PM = Physiological maturity. ² Different letters within each row indicate significant differences between genotypes according to Tukey’s test at p < 0.05.

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Krommydas, K.; Mavromatis, A.; Bletsos, F.; Roupakias, D. Morphology, Heterosis, and Fertility of Novel CMS-Based Solanum melongena × S. aethiopicum Hybrids. Agronomy 2026, 16, 306. https://doi.org/10.3390/agronomy16030306

AMA Style

Krommydas K, Mavromatis A, Bletsos F, Roupakias D. Morphology, Heterosis, and Fertility of Novel CMS-Based Solanum melongena × S. aethiopicum Hybrids. Agronomy. 2026; 16(3):306. https://doi.org/10.3390/agronomy16030306

Chicago/Turabian Style

Krommydas, Konstantinos, Athanasios Mavromatis, Fotios Bletsos, and Demetrios Roupakias. 2026. "Morphology, Heterosis, and Fertility of Novel CMS-Based Solanum melongena × S. aethiopicum Hybrids" Agronomy 16, no. 3: 306. https://doi.org/10.3390/agronomy16030306

APA Style

Krommydas, K., Mavromatis, A., Bletsos, F., & Roupakias, D. (2026). Morphology, Heterosis, and Fertility of Novel CMS-Based Solanum melongena × S. aethiopicum Hybrids. Agronomy, 16(3), 306. https://doi.org/10.3390/agronomy16030306

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Menu

Morphology, Heterosis, and Fertility of Novel CMS-Based Solanum melongena × S. aethiopicum Hybrids

Abstract

1. Introduction

2. Materials and Methods

2.1. Plant Material

2.2. Experimental Design, Site, and Climatic Conditions

2.3. Phenotypic Characterization

2.3.1. Seedling Traits

2.3.2. Plant Traits

2.3.3. Leaf Traits

2.3.4. Flower Traits

2.3.5. Fruit Traits

2.4. Evaluation of Heterosis

2.5. Male Fertility Assessment

2.6. Female Fertility Assessment

2.7. Statistical Analysis

3. Results

3.1. Morphological Characterization

3.1.1. Seedling

3.1.2. Plant

3.1.3. Leaf

3.1.4. Flower

3.1.5. Fruit

3.1.6. Principal Component Analysis

3.2. Heterosis and Heterobeltiosis

3.3. Male Fertility

3.4. Female Fertility

4. Discussion

4.1. Phenotypic Characterization and Uniformity of CMS-Based Interspecific Hybrids

4.2. Expression of Heterosis and Parental Effects

4.3. Fertility and CMS Effects in the Interspecific Hybrids

4.4. Breeding Implications of CMS-Based Interspecific Hybridization

4.4.1. CMS-Based Rootstock Breeding

4.4.2. CMS Introgression into S. gilo and Related Germplasm

5. Conclusions

Author Contributions

Funding

Data Availability Statement

Conflicts of Interest

Appendix A

Appendix A.1. Digital Estimation of Plant Leaf Area

Appendix A.2. Digital Measurement of Leaf Blade Dimensions

References

Share and Cite

Article Metrics

Article Access Statistics

Further Information

Guidelines

MDPI Initiatives

Follow MDPI