Abstract
Currently, consumer demands are rapidly and continuously evolving, which accelerates the development and release of new cultivars. This dynamic process necessitates the integration of biotechnological approaches that can shorten the duration of plant breeding programs. Among these, the doubled haploid (DH) technique, which enables the rapid production of completely homozygous lines in F1 hybrid breeding programs, has been widely and successfully applied in many plant species. Unfertilized ovary culture is one such method and represents a prominent approach for haploid plant production, particularly in the Cucurbitaceae family. However, reports on its successful application in watermelon remain limited. Therefore, the present study aimed to evaluate the effects of culture media composition, pretreatments, and genotype on haploid plant production via ovary culture in watermelon. Although gynogenic embryos and in vitro plantlets were successfully obtained from all tested cultivars, the frequency of haploid induction was influenced by the composition of the culture media, pretreatment conditions, and genotype. The highest haploid induction frequency was achieved using culture media supplemented with a GPT mixture (500 mg L⁻¹ L-glutamine, 250 mg L⁻¹ L-proline, and 0.5 mg L⁻¹ thiamine). Among the tested media, M8 (MS + GPT + 1 mg/l 2,4-D + 1 mg/L BAP + 7 g/L Agar + 30 g/L sucrose + pH:5.8) was identified as the most effective, as it supported haploid plant production across all cultivars. Ploidy analysis of 94 regenerated plants revealed that 82 were haploid (n = 11), 9 were diploid (2n = 22), and 3 were mixoploid (n + 2n). Among the cultivars, Pastel F1 produced the highest number of haploid plants (52), followed by Paskal F1 (14), Crimson Sweet (10), and Crimson Tide F1 (6).
Keywords: Watermelon, Ovary culture, Haploidization
Background
Citrullus lanatus var. lanatus, a member of the Cucurbitaceae family, is the second most produced vegetable crop worldwide after tomatoes, with China ranking first at 61.5 million tons [1]. Watermelon is widely preferred due to its high water content and rich composition of sugars, vitamins, antioxidants, and other bioactive compounds.
F1 hybrid watermelon cultivars are favored for their high yield, superior quality, and resistance to diseases and pests, while also meeting consumer demands for nutritional value, taste, and flavor. In recent years, doubled-haploid (DH) technology has been increasingly utilized in the development of F1 hybrid vegetable cultivars, including melon, cucumber, pepper, and eggplant [2–5]. This technique significantly shortens the breeding cycle to as little as two years, enhances breeding efficiency by reducing population size, and lowers the overall cost of cultivar development. However, the results obtained from the technique vary depending on the species and even the genotypes; some resistant species and genotypes may not respond. On the other hand, some of the obtained DH lines may not possess the desired agronomic characteristics. In today’s context, where both producer and consumer demands are rapidly evolving, the ability to quickly release new F1 hybrid varieties has become essential. Consequently, the importance of DH technology in modern plant breeding has increased substantially [6].
Classical breeding procedures in the Cucurbitaceae family are time-consuming due to their predominantly open-pollinated nature. In contrast, haploidization techniques have been successfully applied in crops such as melon, cucumber, and summer squash as an alternative to conventional breeding methods. The main objective of doubled haploid (DH) technology is to produce haploid (n) plants through androgenesis, gynogenesis, or parthenogenesis, and subsequently develop fully homozygous doubled haploid lines. In open-pollinated species, obtaining pure lines through traditional breeding requires 8–10 generations; however, DH technology significantly reduces this period to approximately 1–2 years. As a result, F1 hybrid cultivars with desirable agronomic traits can be developed and released in a much shorter time.
Considering species within the Cucurbitaceae family, the irradiated pollen technique has been successfully applied in Cucumis melo, Citrullus lanatus, and Cucurbita pepo, while ovary culture is commonly used in Cucumis sativus for double-haploid production. In addition, several factors, including genotype, cultivation conditions for donor plants, seasonal variations, hormone combinations, pre-treatments, incubation conditions, and culture media composition, significantly influence gynogenic response and haploidy frequency in Cucurbitaceae [7–11]. However, only a limited number of studies have addressed ovary culture in watermelon, and an efficient, large-scale protocol for haploid plant production has not yet been established [12–18].
The study aims to determine the effectiveness of the dihaploidization process, specifically focusing on media compositions, pretreatments, and genotypes in producing pure lines of watermelon through unfertilized ovary culture. The ultimate goal is to reduce dependency on external seed suppliers by developing high-yielding, high-quality domestic cultivars that are resistant to diseases and pests. Additionally, the study seeks to contribute to the development of new, superior F1 hybrid watermelon cultivars.
Materials and methods
Plant material
Four commercial watermelon cultivars (Pastel F1, Paskal F1, Crimson Sweet, and Crimson Tide F1) were planted as donors. Pastel F1 was obtained from Petektar Seed Company, Paskal F1 from Fito Seed Company, Crimson Sweet from Asgen Seed Company, and Crimson Tide F1 from Syngenta Seed Company.
Experimental area and crop management practices
The experiment was conducted under controlled conditions in a plastic-covered heated greenhouse, a growth chamber, and a tissue culture laboratory in the 2024 spring season (March to June). Seeds were sown in 140 cm³ cell trays filled with peat moss. Planting rows were covered with black plastic mulch, and 40 seedlings from each cultivar were transplanted at a spacing of 60 × 90 cm. Based on soil analysis, fertilizers were applied via drip irrigation throughout the growing season at rates of 8.5 kg da− 1 nitrogen (N), 9.5 kg da− 1 potassium (K), and 5.5 kg da− 1 phosphorus (P) [19]. Plants were regularly pruned to promote the development of fresh female flower buds. Disease and pest control were managed through the application of systemic fungicides and pesticides.
Preparation of ovaries
Female flower buds were collected from cultivars one day before anthesis between 07:00 and 09:00 AM and immediately transferred to the laboratory. The ovaries were first rinsed under running tap water for 15 min, then surface-sterilized in 15% commercial bleach solution (containing ~ 5% NaOCl) for 20 min. Following sterilization, the ovaries were rinsed three times with sterile distilled water for 5 min each. Excess surface moisture was removed by placing the ovaries on sterile paper towels.
Media compositions and treatments
The composition of culture media and treatment regimes was optimized based on previously established protocols for watermelon [12–18, 20, 21], cucumber [10, 22], winter squash [23], pumpkin, and summer squash [12, 24–26]. To enhance embryogenic response, in addition to conventionally utilized basal media, Murashige and Skoog medium (MS) [27] was supplemented with a GPT additive complex consisting of 500 mg L⁻¹ L-glutamine, 250 mg L⁻¹ L-proline, and 0.5 mg L⁻¹ thiamine. This modification was informed by our preliminary experimental data demonstrating improved embryogenic competence. Treatment conditions were further refined while retaining the core compositional framework of previously reported media formulations (Table 1).
Table 1.
Media compositions and treatments
| Media | Media compositions | Treatments |
|---|---|---|
| M1 | MS + 3 mg L⁻¹ 2,4-D + 2 mg L⁻¹ BAP + 0.5 mg L⁻¹ NAA + 30 g l− 1 sucrose + 7 g/l agar + pH: 5.7 [10] |
1) Keep 35 °C for 4 days in the dark 2) Transferred light conditions on the 5th day 3) Transferred modified MS (GPT + 0.8 mg L⁻¹ BAP + 0.2 mg L⁻¹ NAA + 7 g L⁻¹ Agar + 30 g L⁻¹ sucrose + pH:5.7) at 14th day in growth chamber 4) Media refreshment every 3 weeks 5) Plant regeneration (MS + 0.4 mg L⁻¹ IBA + 30 g L⁻¹ sucrose + 7 g L⁻¹ agar + pH: 5.7) |
| M2 | MS + GPT + 3 mg L⁻¹ 2,4-D + 2 mg L⁻¹ BAP + 0.5 mg L⁻¹ NAA + 30 g L⁻¹ sucrose + 7 g L⁻¹ agar + pH: 5.7 (Modified from Zou et al. [10]) | |
| M3 | MS + 0.5 mg L⁻¹ NAA + 1 mg L⁻¹ 6-BA + 0.5 mg L⁻¹ KIN + 30 g L⁻¹ sucrose + 6 g L⁻¹ agar + pH: 5.7 [11] |
1) Keep 35 °C for 2 days in the dark 2) Keep in the dark for 10 days in the growth cabinet 3) Transferred modified MS (GPT + 0.02 mg L⁻¹ TDZ + 0.25 mg L⁻¹ NAA + 0.5 mg L⁻¹ BA + 30 g L⁻¹ sucrose + 6 g L⁻¹ agar + pH: 5.7) at 14th day in growth chamber 4) Media refreshment every 3 weeks 5) Plant regeneration (MS + 0.4 mg L⁻¹ IBA + 30 g L⁻¹ sucrose + 7 g L⁻¹ agar + pH: 5.7) |
| M4 | MS + GPT + 0.5 mg L⁻¹ NAA + 1 mg L⁻¹ 6-BA + 0.5 mg L⁻¹ KIN + 30 g L⁻¹ sucrose + 6 g L⁻¹ agar + pH: 5.7 (Modified from Zhu et al. [11]) | |
| M5 | MS + GPT + 0.04 mg L⁻¹ TDZ + 7 g L⁻¹ Agar + 30 g L⁻¹ sucrose + pH:5.8 (Modified from Baktemur et al. [6]) |
1) Keep 35 °C for 2 days in the dark 2) Keep in the dark for 5 days in the growth cabinet 3) Transferred modified MS (GPT + 0.2 mg L⁻¹ BAP + 0.05 mg L⁻¹ NAA + 7 g L⁻¹ Agar + 30 g L⁻¹ sucrose + pH:5.8) at 14th day in growth chamber 4) Media refreshment every 3 weeks 5) Plant regeneration (MS + 0.4 mg L⁻¹ IBA + 30 g L⁻¹ sucrose + 7 g L⁻¹ agar + pH: 5.7) |
| M6 | MS + GPT + 0.15 mg L⁻¹ 2,4-D + 1.5 mg L⁻¹ KIN + 7 g L⁻¹ Agar + 30 g L⁻¹ sucrose + pH:5.8 (Modified from Baktemur et al. [6]) | |
| M7 | MS + GPT + 0.1 mg L⁻¹ 2,4-D + 1 mg L⁻¹ BAP + 7 g L⁻¹ Agar + 30 g L⁻¹ sucrose + pH:5.8 | |
| M8 | MS + GPT + 1 mg L⁻¹ 2,4-D + 1 mg L⁻¹ BAP + 7 g L⁻¹ Agar + 30 g L⁻¹ sucrose + pH:5.8 | |
| M9 | MS + GPT + 0.2 mg L⁻¹ 2.4-D + 2 mg L⁻¹ BAP + 0.1 mg L⁻¹ NAA + 30 g L⁻¹ sucrose + 7 g/l agar; pH 5.8 | |
| M10 | MS + GPT + 5 mg L⁻¹ KIN + 0.2 mg L⁻¹ 2,4-D + 7 g L⁻¹ Agar + 30 g L⁻¹ sucrose + pH:5.8 (Modified from Demirel and Onus [17]) |
Ovary culture
The induction media were sterilized by autoclaving at 121 °C and 1.1 atm for 15 min. Following cooling, 8 mL aliquots of the medium were aseptically dispensed into sterile disposable petri dishes (6 × 1.5 cm) under a laminar flow cabinet. Ovaries were sectioned into slices approximately 1–2 mm in thickness. Subsequently, five to eight ovary segments were placed in each petri dish. The dishes were then sealed with sterile stretch film to prevent contamination.
Induction of callogenesis, embryogenesis, and plantlet formation
Heat shock was applied to the cultures as a pretreatment at 35 °C for 2 or 4 days in the dark. Subsequently, the cultures were transferred to dark or light conditions for varying days. At the end of these periods, ovary slices were incubated in the growth chamber at 26 ± 1 ◦ C with a light intensity of 22 µmol m -2 s -1 under a 16 h/8 h photoperiod (Table 1; Fig. 1). Once the ovary slices had differentiated (Fig. 2A, B) and developed into embryogenic calli (Fig. 2C, D), they were transferred to 300 ml culture vessels containing 40 ml of medium every 3 weeks. Subcultures were maintained in the same growth chamber conditions. For plantlet induction, the cultures were refreshed three times at three-week intervals (Fig. 2E, F).
Fig. 1.
Cultured ovary slices; before pretreatments (A), after pretreatments (B)
Fig. 2.
Callus induction (A and B); embryogenesis (C and D); and plantlet formation from ovaries (E and F)
Transformation into plants and ploidy analysis
The plantlets, approximately 1 cm in length, were first transferred to culture tubes containing 10 ml of regeneration media (Fig. 3A). As they developed roots and grew longer, they were transferred to larger culture vessels so they could continue growing into complete plants (Fig. 3B and C). Once the plants were well developed, they underwent acclimatization following the method described by Kurtar and Balkaya [23] (Fig. 3D and E).
Fig. 3.
Root and shoot development (A and B); complete plants (C); acclimatization (D and E)
The ploidy level was determined using chloroplast counts in stomatal guard cells and flow cytometry (FCM). To assess chloroplast number, the third and fourth leaves from the shoot apex were examined [9].
For FCM analysis, around 50 mg of fresh leaves of watermelon seedlings and 50 mg of the fresh leaves of rapeseed seedlings (the internal reference standard) were placed in a Petri dish containing 1.5 mL of a seed isolation liquid placed above an ice-filled box. The leaf tissue was chopped into small strips using a razor blade. Each seed sample solution was passed through a nylon filter and stained with 5 µL of the 1 mg/L prepared propidium iodide stock solution. Nuclear DNA content in samples was analyzed with a Cell Lab Quanta SC flow cytometer (Beckman-Coulter). The ploidy levels of watermelons were determined by comparing the nuclear DNA content in samples with that of the internal reference. Nuclear DNA contents of all materials were calculated according to the formula described by Podwyszynska et al. [28] as follows.
 |
For chloroplast numbers, counts were performed on 10 stomata in three different areas of each of the two leaves per plant.
Statistical evaluations
Because the number of explants and replicates used in the study was unequal, statistical analysis could not be performed; therefore, evaluations were based solely on the averages of the obtained data. The following parameters were assessed: cultured ovary slices (CS), differentiated ovary slices (DS), differentiation rate (DR, %), callus-forming ovary slices (CFS), callus formation rate (CFR, %), calli with embryo-like structures (CES), embryogenic callus rate (ECR, %), regenerated plantlets (RP), regeneration rate (RR, %), and haploidization efficiency (HE; haploid plant per ovary slices).
Results
Induction of callus, differentiation, maturation, and embryogenesis
During the third and fourth weeks of culture, differentiation and color changes were observed in the ovary slices across all cultivars and media. In contrast, the control treatment (M0) showed a low differentiated ovary slices (DR) of 25.0%. Variations were detected among cultivars and induction media in terms of callus-forming ovary slices (CFS) and callus formation rate (CFR, %). The CFR ranged from 27.8% (Pastel F1 in M4) to 100%. However, no callogenic response was observed in M5, M6, and M7 for Crimson Tide F1, and in M7 for Crimson Sweet (Table 2). Additionally, although some ovary slices in the control (M0) showed differentiation, they did not develop into a callus.
Table 2.
Cultured ovary slices (CS), differentiated ovary slices (DS), differentiation rate (DR; %), callus-forming ovary slices (CFS), callus formation rate (CFR; %), callus with the embryo-like structure (CES), embryogenic callus rate (ECR; %), regenerated plantlets (RP), regeneration rate (RR;%) in genotypes (G) and media (M)
| G | M | CS | DS | DR | CFS | CFR | CES | ECR | RP | RR |
|---|---|---|---|---|---|---|---|---|---|---|
| Control | M0 | 12 | 3 | 25,0 | 0 | 0,0 | 0 | 0,0 | 0 | 0,0 |
| Pastel | M1 | 20 | 20 | 100,0 | 20 | 100,0 | 1 | 5,0 | 0 | 0,0 |
| M2 | 24 | 24 | 100,0 | 24 | 100,0 | 2 | 8,3 | 0 | 0,0 | |
| M3 | 19 | 19 | 100,0 | 19 | 100,0 | 0 | 0,0 | 0 | 0,0 | |
| M4 | 18 | 18 | 100,0 | 5 | 27,8 | 0 | 0,0 | 0 | 0,0 | |
| M5 | 11 | 11 | 100,0 | 8 | 72,7 | 0 | 0,0 | 0 | 0,0 | |
| M6 | 14 | 14 | 100,0 | 11 | 78,6 | 0 | 0,0 | 0 | 0,0 | |
| M7 | 17 | 17 | 100,0 | 17 | 100,0 | 3 | 17.6 | 24 | 141.2 | |
| M8 | 19 | 19 | 100,0 | 19 | 100,0 | 5 | 26.3 | 31 | 163.2 | |
| M9 | 14 | 14 | 100,0 | 14 | 100,0 | 0 | 0,0 | 0 | 0,0 | |
| M10 | 19 | 19 | 100,0 | 15 | 78,9 | 1 | 5,3 | 1 | 5,3 | |
| ∑/ Mean | 175 | 175 | 100,0 | 152 | 86,9 | 12 | 6.9 | 56 | 32.0 | |
| Paskal | M1 | 15 | 15 | 100,0 | 15 | 100,0 | 0 | 0,0 | 0 | 0,0 |
| M2 | 13 | 13 | 100,0 | 13 | 100,0 | 3 | 23,1 | 2 | 15,4 | |
| M3 | 14 | 14 | 100,0 | 14 | 100,0 | 0 | 0,0 | 0 | 0,0 | |
| M4 | 16 | 16 | 100,0 | 7 | 43,8 | 0 | 0,0 | 0 | 0,0 | |
| M5 | 11 | 11 | 100,0 | 5 | 45,5 | 0 | 0,0 | 0 | 0,0 | |
| M6 | 14 | 14 | 100,0 | 7 | 50,0 | 0 | 0,0 | 0 | 0,0 | |
| M7 | 13 | 13 | 100,0 | 13 | 100,0 | 0 | 0,0 | 0 | 0,0 | |
| M8 | 11 | 11 | 100,0 | 11 | 100,0 | 4 | 36,4 | 14 | 127.3 | |
| M9 | 14 | 14 | 100,0 | 14 | 100,0 | 0 | 0,0 | 0 | 0,0 | |
| M10 | 16 | 16 | 100,0 | 16 | 100,0 | 0 | 0,0 | 0 | 0,0 | |
| ∑/ Mean | 137 | 137 | 100,0 | 115 | 83,9 | 7 | 5,1 | 16 | 11.7 | |
| C. Sweet | M1 | 21 | 21 | 100,0 | 21 | 100,0 | 0 | 0,0 | 0 | 0,0 |
| M2 | 21 | 21 | 100,0 | 21 | 100,0 | 2 | 9,5 | 0 | 0,0 | |
| M3 | 12 | 12 | 100,0 | 12 | 100,0 | 0 | 0,0 | 0 | 0,0 | |
| M4 | 13 | 13 | 100,0 | 6 | 46,2 | 0 | 0,0 | 0 | 0,0 | |
| M5 | 15 | 15 | 100,0 | 11 | 73,3 | 0 | 0,0 | 0 | 0,0 | |
| M6 | 12 | 12 | 100,0 | 5 | 41,7 | 0 | 0,0 | 0 | 0,0 | |
| M7 | 17 | 17 | 100,0 | 0 | 0,0 | 0 | 0,0 | 0 | 0,0 | |
| M8 | 18 | 18 | 100,0 | 18 | 100,0 | 2 | 11,1 | 2 | 11.1 | |
| M9 | 41 | 41 | 100,0 | 21 | 100,0 | 15 | 36,6 | 11 | 26.8 | |
| M10 | 19 | 19 | 100,0 | 19 | 100,0 | 0 | 0,0 | 0 | 0,0 | |
| ∑/ Mean | 189 | 189 | 100,0 | 134 | 81,5 | 19 | 10,1 | 13 | 6.9 | |
| C. Tide | M1 | 19 | 19 | 100,0 | 19 | 100,0 | 0 | 0,0 | 0 | 0,0 |
| M2 | 32 | 32 | 100,0 | 20 | 90,9 | 4 | 12,5 | 1 | 3,1 | |
| M3 | 18 | 18 | 100,0 | 18 | 100,0 | 0 | 0,0 | 0 | 0,0 | |
| M4 | 18 | 18 | 100,0 | 15 | 83,3 | 0 | 0,0 | 0 | 0,0 | |
| M5 | 11 | 11 | 100,0 | 0 | 0,0 | 0 | 0,0 | 0 | 0,0 | |
| M6 | 13 | 13 | 100,0 | 0 | 0,0 | 0 | 0,0 | 0 | 0,0 | |
| M7 | 17 | 17 | 100,0 | 0 | 0,0 | 0 | 0,0 | 0 | 0,0 | |
| M8 | 18 | 18 | 100,0 | 18 | 100,0 | 2 | 11,1 | 5 | 27.8 | |
| M9 | 15 | 15 | 100,0 | 15 | 100,0 | 1 | 6,7 | 3 | 20.0 | |
| M10 | 11 | 11 | 100,0 | 11 | 100,0 | 1 | 9,1 | 0 | 0,0 | |
| ∑/ Mean | 172 | 172 | 100,0 | 116 | 73,3 | 8 | 4,7 | 9 | 5.2 | |
| Overall | 673 | 673 | 100,0 | 517 | 93,3 | 46 | 6,8 | 94 | 14,1 |
Embryogenic calli were visualized at the end of the subculture periods. The embryos were detected in some calli after 3–4 months of culture. Both cultivar and medium composition had an effect on callus with embryo-like structures (CES) and the embryogenic callus rate (ECR, %). Embryogenesis (CES formation) was observed in all genotypes; however, the response varied depending on the culture medium, and some media did not induce embryogenic responses. Considering the average values, among the tested media, M2 and M8 were the most effective, promoting embryogenic callus formation in all cultivars. In contrast, no embryogenic callus was obtained from media M3, M4, M5, and M6 across all cultivars. In total, CES values varied by cultivar, ranging from 7 in ‘Pascal F1’ to 19 in ‘Crimson Sweet’. Similarly, embryogenic callus rate (ECR) differed among cultivars and media, with frequencies ranging from 5.0% (‘Pastel F1’ in M1) to 36.6% (‘Crimson Sweet’ in M9) (Table 3).
Table 3.
Cultured ovary slices (CS), regenerated plantlets (RP), haploid (H), diploid (D), and mixoploid (M), and haploidization efficiency (HE; ovary slices/haploid plant)
| G | M | CS | RP | H | D | M | HE |
|---|---|---|---|---|---|---|---|
| PASTEL | M7 | 17 | 24 | 21 | 2 | 1 | 1,24 |
| M8 | 19 | 31 | 31 | 0 | 0 | 1,63 | |
| M10 | 19 | 1 | 0 | 1 | 0 | 0 | |
| ∑/Mean | 55 | 56 | 52 | 3 | 1 | 0,96 | |
| PASKAL | M2 | 13 | 2 | 0 | 2 | 0 | 0 |
| M8 | 11 | 14 | 14 | 0 | 0 | 1,27 | |
| ∑/ Mean | 24 | 16 | 14 | 2 | 0 | 0,58 | |
| C. SWEET | M8 | 18 | 2 | 1 | 1 | 0 | 0,06 |
| M9 | 41 | 11 | 9 | 1 | 1 | 0,22 | |
| ∑/ Mean | 59 | 13 | 10 | 2 | 1 | 0,17 | |
| C. TİDE | M2 | 32 | 1 | 0 | 1 | 0 | 0 |
| M8 | 18 | 5 | 4 | 0 | 1 | 0,22 | |
| M9 | 15 | 3 | 2 | 1 | 0 | 0,13 | |
| ∑/ Mean | 65 | 9 | 6 | 2 | 1 | 0,09 | |
| Overall | 203 | 94 | 82 | 9 | 3 | 0,40 |
Relatively higher CFS values were detected in the cultivars Pastel F1 in M3 (19), Paskal F1 in M1 and M5 (15 and 16), Crimson Sweet in M10 and M1 (19 and 21), and Crimson Tide F1 in M4 and M2 (15 and 19); however, these calli did not produce CES. Regarding the culture media, M2 and M8 yielded satisfactory results in callus formation, with some calli developing embryo-like structures. When the mean values were examined, a relationship was observed between ECR and CES, and relatively high ECR values were obtained across the cultivars. Among the cultivars tested, Crimson Sweet exhibited the best performance, with mean CES and ECR values of 19 and 10.1%, respectively.
Embryo development and plantlet formation
From 673 cultured ovary slices, a total of 517 calli were obtained; however, only 46 of these developed into embryogenic calli. The calli exhibiting embryogenic structures were transferred to embryo development and plantlet formation media between the 11th and 13th weeks of culture. Plantlets subsequently emerged from the embryogenic calli during the 14th to 16th weeks (Table 1). Although responses varied among cultivars, regenerated plantlets (RP) were observed on M2, M7, M8, M9, and M10 media.
All regenerated plantlets were obtained from media supplemented with the GPT mixture (500 mg L⁻¹ L-glutamine, 250 mg L⁻¹ L-proline, and 0.5 mg L⁻¹ thiamine). Among the tested media, M8 supported regeneration in all varieties and proved to be the most productive medium. In total, the Pastel F1 cultivar showed the highest regeneration efficiency, producing 56 plants from 12 embryogenic calli. In contrast, Crimson Tide F1 exhibited the lowest response, with only 9 plants regenerated from 8 embryogenic calli. Overall, a total of 94 plants were obtained from 46 embryogenic calli.
Transformation into plants and ploidy analysis
In ovary culture, most of the plantlets transformed into mature plants. The plants that reached approximately 1 cm in length were cultured in regeneration media. The plantlets turned into complete plants within 2 or 4 weeks of culture. A total of 97 plantlets underwent the acclimatization process; 94 of these were successfully adapted to external conditions and used for ploidy analysis. The survival rate following acclimatization was 97%. The ploidy status of plants was determined by both counting the chloroplast number of guard cells and flow cytometry (FCM). While haploid (n = 11) plants have 6–8 chloroplasts in the stomata of guard cells, it was 10–12 chloroplasts in diploid (2n = 22) plants. Both haploid and diploid stomatal structures have been observed in mixoploid (n + 2n) plants (Fig. 4). The FCM histograms of diploid and haploid plants are presented in Fig. 5.
Fig. 4.
The chloroplast number of haploid (A), diploid (B), and mixoploid (C) plants
Fig. 5.
The FCM histogram of diploid (A) and haploid (B) plants (1st peak)
Ploidy analysis of 94 plants revealed that 82 were haploid, 9 were diploid, and 3 were mixoploid. Among the genotypes, Pastel F1 produced the highest number of haploid plants (52), followed by Paskal F1 (14), Crimson Sweet (10), and Crimson Tide F1 (6) (Table 2). Haploidy efficiency (HE), calculated based on the number of cultured ovaries, varied notably among genotypes. The highest HE was observed in Pastel F1 (0.96), whereas the lowest was recorded in Crimson Tide F1 (0.09). Haploid plants were successfully regenerated from M7, M8, and M9 media, with M8 proving to be the most productive. Examination of the media compositions revealed that all effective media contained a combination of GPT along with varying concentrations of 2,4-D and BAP.
Discussion
Ovary culture represents a valuable approach within the double-haploidization (DH) process, offering a time-efficient strategy for the development of F1 hybrid vegetable cultivars. Considerable progress has been achieved in optimizing ovary culture protocols for several cucurbit species, including Cucumis melo L., Cucumis sativus L., Cucurbita pepo L., Cucurbita maxima Duch., and Cucurbita moschata Duch. However, in sweet watermelon (Citrullus lanatus var. lanatus), the outcomes remain variable and insufficiently characterized. Although Kara et al. [18] reported the successful induction of haploid plants via ovary culture in watermelon, the absence of a standardized and genotype-independent protocol constitutes a major limitation. This gap has provided the primary rationale for the present study. Therefore, the effectiveness of different medium compositions and pretreatments on DH production in watermelon via ovary culture was investigated. Although limited studies have been conducted on ovary culture in sweet watermelon, our findings contribute to the understanding of this process in other species of the Cucurbitaceae family.
The induction of embryogenesis and plant regeneration in ovary culture was critically dependent on both genotype and culture medium composition. Considering the average values, different responses were observed among the tested media, with some formulations promoting high embryogenic competence, while others exhibited minimal or no morphogenic response. Callus induction and subsequent differentiation rates differed markedly across cultivars, indicating strong genotype-dependent effects. Furthermore, the media formulations demonstrated considerable heterogeneity in their capacity to support embryogenesis and plantlet regeneration, with certain treatments resulting in comparatively low frequencies of embryo formation and shoot initiation. These outcomes suggest that embryogenic induction and plantlet development are highly regulated by the qualitative and quantitative balance of nutrients and plant growth regulators within the culture medium.
MS [27] media supplemented with different amounts of 2,4-D, NAA, BAP, KN, and TDZ [11, 14–18, 29, 30], putrescine, and spermidine [31] have been used for the induction of gynogenic response in the watermelon. Furthermore, transcriptome-based studies have revealed that heat shock promotes the development of the unfertilized ovules by inducing the biosynthesis of amino acids, suggesting the need to add certain amino acids to the growing medium in watermelon ovary cultures [32]. Indeed, promising results have been reported in cucurbit ovary culture using amino acids such as proline, serine, and glutamine [33]. It has been shown that amino acids are the precursors of protein biosynthesis, that heat shock can activate the ABA signaling pathway, and that it can promote the accumulation of storage substances such as amino acids and unsaturated fatty acids in ovary culture [34].
Although all media influenced the differentiation of cultured ovaries, callogenesis did not occur at a comparable rate across treatments. Our findings revealed that the responses to callogenesis and embryogenesis varied among the culture media and genotypes tested. While some culture media and genotypes exhibited favorable responses, others showed little or no response under the experimental conditions. Callus formation was observed in all media for the Pastel F1 and Pascal F1 cultivars; however, it was absent in Crimson Sweet on M7 and in Crimson Tide F1 on M5, M6, and M7. Embryogenic capacity rate (ECR, %) and regeneration potential (RP) were identified as key indicators of ovary culture fecundity. Regarding embryogenesis, only media M2 and M8 induced embryonic structures in all cultivars. The highest embryogenic capacity (36.6%) was recorded in M9 media for the Crimson Sweet cultivar. Despite these findings, no relationship was detected between the embryogenic callus ratio and the number of regenerated plants.
Although embryogenic calli were successfully obtained, plantlet regeneration was not achieved in the M1 medium. In this study, embryo development and plantlet regeneration were achieved in all culture media supplemented with the GPT mixture. In the Crimson Sweet cultivar, only 11 plants were regenerated from the M9 medium, despite it producing the highest number of embryogenic structures. In contrast, the Pastel F1 cultivar showed the highest regeneration efficiency, with 1, 24 and 31 plantlets regenerated from the M10, M7 and M8 media, which produced 1, 3 and 5 embryogenic calli, respectively. Notably, M8 was the only medium that enabled the production of haploid plants across all varieties.
When comparing cultivars, Pastel F1 showed high fecundity, whereas Crimson Tide F1 exhibited a lower regeneration rate and was more recalcitrant. These findings are consistent with previous reports in Citrullus lanatus [12, 17–20, 21, 30–32, 35–38]. The influence of genotype and induction medium on ovary culture in Cucurbitaceae has also been documented in cucumber [10, 21, 29, 39], melon [7, 40, 41], summer squash [42], and winter squash [9, 25, 28].
The number of guard cells and FCM analyses were used to determine the ploidy status, and these techniques were found to be effective in watermelon. Thus, the counting of chloroplast numbers was a quick, reliable, and efficient method. The results of the ploidy analyses, diploid, haploid, and mixoploid plants, were determined. Diploid plants can form diploid ovary tissues or undergo spontaneous double-haploidization. Spontaneous double-haploids are valuable for the dihaploidisation process because they do not need chromosome doubling [9, 29]. The production of mixoploid plants that subsequently spontaneously doubled into double-haploid (2n) lines in dihaploidization breeding was found to be noteworthy.
The present study reports a significant advancement in the production of haploid plants via ovary culture in watermelon (Citrullus lanatus) cultivars. The regenerated haploid plants constitute valuable genetic resources for future breeding programs, particularly for the development of superior F1 hybrid cultivars. Molecular analyses are required to determine whether plants identified as diploid are spontaneous doubled haploids (DHs) or originate from diploid tissues. Diploid plants confirmed as DHs through these analyses can be directly incorporated into breeding programs. Furthermore, the findings indicate that the supplementation of the induction medium with specific amino acids, namely proline and glutamine, in combination with thiamine, enhances the frequency of haploid induction in watermelon ovary cultures. These results highlight the potential of optimizing culture media composition to improve the efficiency of haploidization techniques in watermelon breeding.
Conclusions
Doubled haploid (DH) technology is used to rapidly produce homozygous, fertile lines, particularly in open-pollinated watermelon. This study reports the successful induction of plantlets through ovary culture in watermelon, a species considered recalcitrant to such techniques. The results showed that the success of ovary culture is highly dependent on both genotype and the composition of the culture medium. Notably, media supplemented with GPT enhanced callus formation (callogenesis), embryo development (embryogenesis), and plantlet regeneration. In conclusion, future studies should focus on optimizing induction media by supplementing them with polyamines (spermidine, spermine, and putrescine), selected amino acids (glutamine, proline, arginine, alanine, and glycine), and novel plant growth regulators such as brassinosteroids, jasmonates, karrikinolide, triacontanol, and meta-topolin at varying concentrations. Additionally, factors including the growing season and culture conditions of donor plants, ovary pretreatments, and subculture frequency should be systematically evaluated to enhance gynogenic responses in recalcitrant watermelon lines.
Acknowledgements
The authors thank TUBITAK for their support. The ovary culture studies from the second year of the project formed part of Deniz Metin’s doctoral thesis. We would also like to thank Prof. Dr. Ali Ramazan ALAN, Prof. Dr. Fevziye Çelebi TOPRAK, and Assoc. Prof. Dr. Musa SEYMEN for their valuable contributions.
Authors’ contributions
D.M. and E.S.K. conceived and designed the study, conducted the experiments, analyzed the data, and wrote the manuscript.
Funding
This study was funded by the Scientific and Technological Research Council of Turkey (TUBITAK) under the Grant Number 221O338. The authors thank TUBITAK for their support. The ovary culture studies from the second year of the project formed part of Deniz Metin’s doctoral thesis.
Data availability
The data obtained throughout the study were organized into tables in docx format using Word software and stored electronically.
Declarations
Ethics approval and consent to participate
Not applicable.
Consent for publication
Not applicable.
Competing interests
The authors declare no competing interests.
Footnotes
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.FAO. Food and Agriculture Organization of the United Nations. http://faostat.fao.org. 2024.
- 2.Kambelash F, Moieni V, Ashtiyani AK, Mohammad RS. Parthenogenesis-based production of doubled haploids in melon (Cucumis melo L.). Plant Cell, Tissue and Organ Culture (PCTOC). 2025;162. 10.1007/s11240-025-03196-y. [DOI]
- 3.Nyirahabimana F, Solmaz I. Haploid induction through ovary culture in cucumber. Vitro Cell Dev Biology - Plant. 2024;60. 10.1007/s11627-023-10406-9. [DOI]
- 4.Mir R, Calabuig-Serna A, Seguí-Simarro J. Doubled Haploids Eggplant Biology. 2021;10. 10.3390/biology10070685. [DOI] [PMC free article] [PubMed]
- 5.Misal S, Das A. Development of a unique protocol for the production of doubled haploids in Hot Pepper. 2023; 10.21203/rs.3.rs-3294597/v1. [DOI]
- 6.Kurtar ES, Seymen M, Çetin AN, Türkmen Ö. Dihaploidization in promising summer squash genotypes (Cucurbita pepo L.) via irradiated pollen technique. Yuzuncu Yıl Univ J Agricultural Sci. 2021;31(1):42–51. [Google Scholar]
- 7.Dong YQ, Zhao WX, Li XH, Liu XC, Gao NN, Huang JH, Wang WY, Xu XL. Androgenesis, gynogenesis, and parthenogenesis haploids in cucurbit species. Plant Cell Rep. 2016;35(10):1991–2019. [DOI] [PubMed] [Google Scholar]
- 8.Kurtar ES, Balkaya A, Kandemir D. Evaluation of haploidization efficiency in winter squash (Cucurbita maxima Duch.) and pumpkin (Cucurbita moschata Duch.) through anther culture. Plant Cell Tissue Org Cult. 2016;127:497–511. 10.1007/s11240-016-1074-6. [DOI] [Google Scholar]
- 9.Kurtar ES, Seymen M. Gynogenesis in Cucurbita species. Doubled Haploid Technology: Volume 3: Emerging Tools, Cucurbits, Trees, Other Species; Methods in Molecular Biology. Springer US, New York, NY, USA, p. 2021; 123–133. Segui-Simarro, J. M., Ed.ISBN 978-1-07-161331-3. 10.1007/978-1-0716-1331-3_8 [DOI] [PubMed]
- 10.Baktemur G, Keles D, Kara E, Yıldız S, Taskın H. Effects of genotype and nutrient medium on obtaining haploid plants through ovary culture in cucumber. Mol Biol Rep. 2022;49:5451–8. 10.1007/s11033-022-07238-y. [DOI] [PubMed] [Google Scholar]
- 11.Metin D, Atakul Z, Kurtar ES, Seymen M, Alan AR, Toprak FC. Callogenesis, embryogenesis, and plantlet initiation in citron watermelon (Citrullus lanatus var. citroides) via anther and unfertilized ovary culture. Sci Hort. 2024;337:113493. [Google Scholar]
- 12.Gürsoy I, Solmaz I, Deliboran S, Sari N. In vitro ovule and ovarium culture in watermelon. In: Proceedings of the 10th EUCARPIA Meeting on Genetics and Breeding of Cucurbitaceae, Cucurbitaceae. Antalya, Turkey, University of Cukurova. 2012;799–804.
- 13.Zhang LJ, Yang Y, Shen WJ, Wang JS. Haploid plant and its identification of watermelon using gynogenesis in vitro induction. J Agric Sci. 2015;36(4):8–11. [Google Scholar]
- 14.Zou T, Su HN, Wu Q, Sun XW. Haploid induction via unfertilized ovary culture in watermelon. Plant Cell Tissue Organ Cult. 2018;135. 10.1007/s11240-0181454-1. 179 – 18. [DOI]
- 15.Zhu Y, Sun D, Deng Y, Li W, An G, Si W, Liu J. Regeneration of doublehaploid plants from unpollinated ovary cultures of watermelon. 2019; 10.21203/rs.2.14098/v1. [DOI]
- 16.Sari N, Solmaz I. Doubled haploid production in watermelon. Doubled Haploid Technology. New York, NY: Humana; 2021. p. 97–110. [DOI] [PubMed] [Google Scholar]
- 17.Isak MA, Majeed H, Yetisir H, Simsek O. Evaluation of factors affecting embryo-like structure and callus formation in unpollinated ovary culture of watermelon (Citrullus lanatus). Int J Agric Nat Sci. 2022;15(3):319–35. [Google Scholar]
- 18.Kara E, Taşkın H, Açar E, Solmaz I, Aka Kaçar Y, Karaköy T, Baktemur G. Development of homozygous lines in watermelon (Citrullus lanatus L.) by ovary culture. Euphytica. 2026;222(1):1. [Google Scholar]
- 19.Bao L, Zhang SF, Qin JM, Zhang HM, Wang PZ, Zhou JB, Chen Z. Recommended fertilization rates based on watermelon yield and fertilizer reduction potential. J Plant Nutr Fertilizers. 2022;28:1558–68. [Google Scholar]
- 20.Yue W, Yiqin G, Bo D, Tonghui R. Induction of watermelon anther callus. Hubei Agric Sci. 2005;5:93–5. [Google Scholar]
- 21.Da Silva CMJ, Dias RCS, Melo NF. The effect of temperature and growth regulators on callus induction in watermelon anthers. Braz Arch Biol Technol. 2021;64:e21180505. 10.1590/1678-4324-2021180505. [DOI] [Google Scholar]
- 22.Demirel E, Onus AN. Obtaining haploid embryo and plant by gynogenesis in some cucumber (Cucumis sativus L.) cultivars and types. Int J Agric Wildl Sci. 2021;7(3):360–7. 10.24180/ijaws.969741. [DOI] [Google Scholar]
- 23.Kurtar ES, Balkaya A. Production of in vitro haploid plants from in situ induced haploid embryos in winter squash (Cucurbita maxima Duchesne ex Lam.) via irradiated pollen. Plant Cell Tissue Org Cult. 2010;102:267–77. 10.1007/s11240-010-9729-1. [DOI] [Google Scholar]
- 24.Kurtar ES, Balkaya A, Ozbakir Ozer M. Production of callus mediated gynogenic haploids in winter squash (Cucurbita maxima Duch.) and pumpkin (Cucurbita moschata Duch). Czech J Genet Plant Breed. 2018;54(1):9–16. 10.17221/30/2017-CJGPB. [DOI] [Google Scholar]
- 25.Kurtar ES, Seymen M. Anther culture in cucurbita species. Doubled Haploid Technology: Volume 3: Emerging Tools, Cucurbits, Trees, Other Species; Methods in Molecular Biology. Springer US, New York, NY, USA. 2021; 111–121. Segui-Simarro, J.M., Ed.ISBN 978-1-07-161331-3. 10.1007/978-1-0716-1331-3_7 [DOI]
- 26.Kurtar ES, Seymen M. Induction of parthenogenesis by irradiated pollen in Cucurbita species. Doubled Haploid Technology Jose M. Seguí-Simarro Methods in Molecular Biology 2289 Volume 3 Emerging Tools, Cucurbits, Trees, Species O. 135–145. Methods in Molecular Biology. Springer US, New York, NY, USA. 2021; 123–133. Segui-Simarro, J. M., Ed.ISBN 978-1-07-161331-3. 10.1007/978-1-0716-1331-3_9 [DOI] [PubMed]
- 27.Murashige T, Skoog F. A revised medium for rapid growth and bioassays with tobacco cultures. Physiol Plant. 1962;15:473–97. [Google Scholar]
- 28.Podwyszynska M, Kruczynska D, Machlanska A, Dyki B, Sowik I. Nuclear DNA content and ploidy level of apple cultivars including Polish ones in relation to some morphological traits. Acta Biol Crac Ser Bot. 2016;58:81–93. 10.1515/abcsb-2016-0008. [DOI] [Google Scholar]
- 29.Kurtar ES, Seymen M, Kal Ü. An overview of doubled haploid plant production in Cucurbita species. Yuz Yıl Univ J Agric Sci. 2020;29(2):510–20. 10.29133/yyutbd.741087. [DOI] [Google Scholar]
- 30.Nyirahabimana F, Solmaz I, Sarı N. An overview of haploid plant production in cucurbits. Agricultural Practices and Sustainable Management ın Türkiye. 2021; 31–56. In: BELLİTÜRK, K.; SOLMA, Y (Eds.)., Iksad Publications, ISBN: 978-625-8213-37-9.
- 31.Yıldız C, Solmaz I. Obtention of haploid plant in watermelon genetic resources by ovule-ovary culture method. Çukurova J Agric Food Sci. 2020;35(1):35–42. 10.36846/CJAFS.2020.16. [DOI] [Google Scholar]
- 32.Zhu Y, Sun D, Deng Y, An G, Li W, Si W, Liu J, Sun X. Comparative transcriptome analysis of the effect of different heat shock periods on the unfertilized ovule in watermelon (Citrullus lanatus). J Integr Agric. 2020;19:528–40. [Google Scholar]
- 33.Domblides E, Shmykova N, Khimich G, Korotseva I, Kan L, Domblides A, Pivovarov V, Soldatenko A. Production of doubled haploid plants of Cucurbitaceae family crops through unpollinated ovule culture in vitro. Acta Hortic. 2020;1294:19–28. [Google Scholar]
- 34.Gong S, Tang B, Dai Y, Sun X, Song H, Xiong C, Sun X. Transcriptome and proteome reveal heat shock promotes haploid induction rate via activating ABA signal transduction in watermelon. Agronomy. 2025;15(5):1063. [Google Scholar]
- 35.Tulukoglu SK. Effects of cold application, Thidiazuron (TDZ) and 2,4-D applications on haploid embryo stimulation in anther and ovule culture in watermelons. Çukurova University Institute of Science, Department of Horticulture. Master’s Thesis. 2014;78. https://tez.yok.gov.tr/ID367745.
- 36.Abdollahi MR, Darbandi M, Hamidvand Y, Majdi M. The influence of phytohormones, wheat ovary co-culture, and temperature stress on anther culture response of watermelon (Citrullus lanatus L). Braz J Bot. 2015;38(3):447–56. 10.1007/s40415-015-0152-z. [DOI] [Google Scholar]
- 37.Zhu Y, Sun D, Deng Y, An G, Li W, Liu J. Effects of preliminary treatment on watermelon anther culture callus induction. China Cucurbits Veg. 2012;25(5):17–9. [Google Scholar]
- 38.Akbas FC, Solmaz I. Obtention of haploid plant in Citrullus lanatus var. lanatus and Citrullus lanatus var. citroides species by anther culture method. Int J Environ Res Technol. 2019;2(3):25–36. [Google Scholar]
- 39.Asadi A, Seguí-Simarro JM. Production of doubled haploid plants in cucumber (Cucumis sativus L.) through anther culture. Doubled Haploid Technology. New York, NY: Humana; 2021. pp. 71–85. 10.1007/978-1-0716-1331-3_4. [DOI] [PubMed] [Google Scholar]
- 40.Nguyen ML, Ta THT, Huyen TNBT, Voronina AV. Anther-derived callus formation in bitter melon (Momordica charantia L.) as influenced by microspore development stage and medium composition. Sel’skokhozyaistvennaya Biol. 2019;54(1):140–8. [Google Scholar]
- 41.De Oliveira FIC, Carvalho AVF, de Arag˜ao FAS, de Carvalho ACPP. Growth regulators in the induction of calli in anthers of the Goldex hybrid yellow melon. Rev Agro@ mbiente On-line. 2022;16:14. 10.18227/1982-8470ragro.v16i0.7224. [DOI] [Google Scholar]
- 42.Refaei EFS, Mohamed MF, Shalaby GI. Planting season and dates of donor plants influence in vitro androgenesis in summer squash. J Agric Res Tanta Univ. 2002;28(3III):973–80. [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The data obtained throughout the study were organized into tables in docx format using Word software and stored electronically.