Molecular Breeding

, Volume 24, Issue 3, pp 201–211

Improved watermelon quality using bottle gourd rootstock expressing a Ca2+/H+ antiporter


  • Jeung-Sul Han
    • Department of Ecological Environment ConservationKyungpook National University
  • Sunghun Park
    • Department of Horticulture, Forestry & Recreation Resources (HFRR) 2021 Throckmorton Plant Science CenterKansas State University
  • Toshiro Shigaki
    • Baylor College of Medicine, Departments of Pediatrics, and Human and Molecular GeneticsUSDA/ARS Children’s Nutrition Research Center
  • Kendal D. Hirschi
    • Baylor College of Medicine, Departments of Pediatrics, and Human and Molecular GeneticsUSDA/ARS Children’s Nutrition Research Center
    • Department of Environmental HorticultureKyungpook National University

DOI: 10.1007/s11032-009-9284-9

Cite this article as:
Han, J., Park, S., Shigaki, T. et al. Mol Breeding (2009) 24: 201. doi:10.1007/s11032-009-9284-9


Bottle gourd (Lagenaria siceraria Standl.) has been commonly used as a source of rootstock for watermelon. To improve its performance as a rootstock without adverse effects on the scion, the bottle gourd was genetically engineered using a modified Arabidopsis Ca2+/H+ exchanger sCAX2B. This transporter provides enhanced Ca2+ substrate specificity and decreased Mn2+ transport capability. Our previous work demonstrated that sCAX2B bottle gourds were more robust and nutrient dense than controls. Here, several cucurbit crops were test-grafted onto the transgenic bottle gourd to determine its effect on the scions. The grafted watermelons and melons onto the transgenic rootstocks appeared to show more robust growth than the controls 35 days after greenhouse transplanting. Watermelon fruits with the watermelon/transgenic bottle gourd (scion/rootstock) combination demonstrated higher osmotic pressure and more soluble solids than controls. These results suggest that sCAX2B expression in the bottle gourd rootstock facilitates improved watermelon quality through the translocation of nutrients and/or water toward enhancing the biomass of scion.


Bottle gourd (Lagenaria siceraria Standl.)CalciumGraftingTransformationTransportRootstockWatermelon



Cation exchanger


Polymerase chain reaction


Uptake and translocation of cationic nutrients in plants plays an essential role in physiological processes such as growth, signal transduction, and development (Flowers and Colmer 2008; Wang et al. 2006). The differential partitioning of cations is crucial for biochemical processes in the cell, and transporters play a critical role in maintaining the proper concentrations of these ions in various cellular compartments (Cohen and Nelson 2000; Shigaki et al. 2003). Two Arabidopsis cation/H+ exchanger genes, CAX1 and CAX2, can function in yeast (Saccharomyces cerevisiae) mutants as vacuolar Ca2+ transporters (Hirschi et al. 1996). The transport properties of CAX2 suggested the potential for broad substrate specificity, whereas CAX1 appeared to be a specialized Ca2+ transporter. Subsequent studies have demonstrated that besides Ca2+, CAX2 can also transport Mn2+, Cd2+ and Zn2+ (Hirschi et al. 2000; Pittman et al. 2004; Schaaf et al. 2002; Shigaki et al. 2003). To identify the domains that determine Mn2+ specificity in CAX2, six CAX2 variants without the N-terminal autoinhibitory domain (i.e., without the first 42 amino acids) (sCAX2s) were constructed by replacing the regions of divergent sequence in CAX2 with the corresponding regions from CAX1 (Shigaki et al. 2003). When the sCAX2 variant genes were expressed in yeast mutants defective in vacuolar Ca2+ transport, the yeast strains expressing sCAX2A and sCAX2B exhibited diminished growth on Ca2+-containing media, and completely failed to suppress Mn2+ sensitivity (Shigaki et al. 2003). These results imply that among the six sCAX2 variants, sCAX2A and sCAX2B are preferred candidates to genetically engineer a robust crop for its moderate but more specific ability of Ca2+ transport and diminished ability to accumulate metals such as Mn2+. Indeed, potato tubers expressing sCAX2B contained 50–65% more Ca2+ than wild-type tubers although these plants are as vigorous as controls, and did not show any significant increase of Mn2+ (Kim et al. 2006). In contrast, the transgenic crops ectopically expressing sCAX1 displayed alterations in plant development and morphology, including increased incidence of blossom-end rot and tip-burning despite the transgenic plants having elevated Ca2+ levels (Hirschi 1999; Park et al. 2004, 2005a).

Recently bottle gourd (Lagenaria siceraria Standl.), with its long history of cultivation (Erickson et al. 2005), has attracted public attention because the crop has been used as a rootstock for watermelon in the Far Eastern countries (Lee and Oda 2003). The grafting onto a bottle gourd generally provides the resistance to soil born diseases such as watermelon Fusarum wilt (Fusarium oxysprum f. sp. niveum), and confers tolerance to low-soil temperatures. However, a recent epidemic of sudden wilt in cucurbit crops has been reported (Edelstein et al. 1999; Gal-On et al. 2005; Park et al. 2005c). Therefore, it is desirable to improve the performance of bottle gourd rootstocks for enhanced stress tolerance.

In this study, we characterize the bottle gourd ectopically expressing the sCAX2B and the phenotypes of cucurbit crops grafted onto the transgenic bottle gourd rootstock. Our study revealed that both watermelon and melon grafted onto the sCAX2B-expressing bottle gourd, as well as the self-rooted transgenic bottle gourd, have enhanced biomass without any adverse traits compared to controls.

Materials and methods

Generation of sCAX2B-expressing bottle gourd via Agrobacterium-mediated transformation

The coding region of sCAX2B (Shigaki et al. 2003) was cloned into pBIN19 (CLONTECH Laboratories, Palo Alto, CA), which contained the T-DNA cassette consisting of Nos-pro/NptII/Nos-ter/35S-pro/sCAX2B/Nos-ter (Kim et al. 2006). The plasmid was introduced into Agrobacterium tumefaciens strain GV3101 using the freeze-thaw method (Holsters et al. 1978), which was used for the genetic transformation.

Bottle gourd (Lagenaria siceraria Standl.) transformation was performed via Agrobacterium-mediated transformation method using the cotyledon explants of G5 inbred line as previously reported (Han et al. 2004, 2005) except that 100 mg l−1 kanamycin monosulfate was used as a selective agent for selecting transgenic shoots.

Nucleic acid analyses of transgenic bottle gourd

Bottle gourd genomic DNA was extracted from the leaf tissue, according to the protocol developed for cotton by Paterson et al. (1993). Four grams of fresh young leaves were ground with a mortar and pestle in liquid nitrogen, and the fine powder was mixed with 20 ml of the ice-cold cotton DNA extraction buffer. The sample was then centrifuged at 2,700×g, and the pellet was suspended with 8 ml of cotton nuclei lysis buffer, and incubated at 65°C for 20 min. A 10 ml of chloroform-isoamyl alcohol (24:1) was added, mixed, and then the mixture was centrifuged at 2,700×g. The upper phase was transferred to a 15 ml tube, mixed with 5.4 ml of ice-cold isopropanol, and centrifuged at 10,000×g. After pouring off the supernatant, the pellet was washed using 70% ethanol, and resuspended in TE buffer (pH 8.0) by incubating in a 65°C water bath for 20 min. Finally, the suspension was centrifuged at 10,000×g to pellet residual impurities. The supernatant was transferred to a clean tube, and used for DNA analyses. For PCR amplification of sCAX2B gene, one set of primers (5′-GAATGTGACAGAGCTGATCA-3′ and 5′-GATCGAGGACCCAATAGCCA-3′) was used.

For Southern blot analysis on the randomly selected PCR-positive plants, DNA (5–10 μg) was digested with XbaI (which made only one cut within the T-DNA), and was then separated by electrophoresis in a 0.9% agarose gel, and blotted onto a nylon membrane (Zeta-probe GT membrane, Bio-Rad Laboratories) according to the manufacturer’s instructions. The probe for the sCAX2B gene was isolated from an XbaI-SacI (1.2-kb) restriction fragment from the plasmid (Kim et al. 2006). The membrane was pre-hybridized overnight at 65°C in 7% SDS and 0.25 M Na2HPO4, and then hybridized overnight at 65°C in the same solution containing the probe labeled with 32P-dCTP using NEBlot kit (NEB BioLabs). The membrane was washed twice for 30 min each with 20 mM Na2HPO4 and 5% SDS at 65°C, and then washed twice again for 30 min each with 20 mM Na2HPO4 and 1% SDS 65°C. The membrane was exposed to X-ray film at −80°C.

Total RNA was also extracted from the newly developing young leaves of the randomly selected PCR-positive plants using RNeasy Plant Kits (Qiagen, Hilden, Germany) according to the manufacturer’s instructions.

For Northern-blot analysis, total RNA (7 μg) was separated on a 1.2% agarose gel containing 1.5% formaldehyde, blotted onto a Zeta-Probe GT membrane according to the manufacturer’s instructions. Hybridization and washing were as previously described in Southern-blot analysis.

A reverse transcriptional PCR (RT-PCR) analysis was performed to detect sCAX2B transcripts of transgenic plants in T3 generation (TG137 line). RNA samples were treated with DNase I (Invitrogen, Carlsbad, CA) two times to remove any contaminated genomic DNA. The first strands of cDNA were synthesized using 0.2 μg of total RNA as template in 20 μl of reaction mixture, which included 50 ng of random hexamers and 200 units of Superscript III RT (Invitrogen, Carlsbad, CA). A 4 μl of the first strand cDNA in 50 μl was used to amplify a sCAX2B gene-specific fragment. Forward and reverse primers as previously described in PCR analysis were used at concentrations of 0.8 μM each. The conditions for amplification were 95°C for 5 min followed by 35 cycles at 95°C for 30 s, 60°C for 45 s, 72°C for 1 min, and 72°C for 5 min. PCR products were run on 1.4% agarose gels and stained with ethidium bromide.

Hydroponic culture and cation analysis

Eight sCAX2B-expressing and wild-type bottle gourd seedlings were established in household hydroponic culture bathes (Hi-Green; Gwang Myeong Control Electric, Korea), which were equipped with the waterproof electric motors continuously circulating the medium at 5 l/min−1. Three bathes were used for each transgenic and wild-type treatment. The 12 l bathes were resupplied with the progressively fortified MS medium (Murashige and Skoog 1962) every 6 days in the order of 0× [electric conductivity (EC) 0.0 mS/cm], 1/8× (EC 0.8 mS/cm), 1/2× (EC 3.1 mS/cm), and final 2× (EC 11.2 mS/cm). De-ionized water was used for preparing all media, and the pH of the medium was adjusted to 5.8 using 1 N HCl and 1 N NaOH before applying. All roots and shoots from bottle gourd plants grown on the final 2× MS medium for 10 days were dried at 70°C for 4 days. A total of 48 dried samples were ground, and then 0.25 g (dry weight) was digested using a ternary solution (HNO3:H2SO4:HClO4 = 10:1:4 ratio by volume; Lee et al. 2002). Cation contents per gram of dry weight were determined by inductively coupled plasma emission spectrophotometry (Integra XM2, GBC Scientific Equipment Inc., Australia). One replicate was defined as the average value from three plants similarly positioned in three hydroponic baths. The comparison of ion concentrations between transgenic and wild-type bottle gourd plants was conducted using a t-test at the 5% level.

Grafting of cucurbit crops

Germinated seeds of several cucurbit crops [two commercial watermelons (Citrullus vulgaris Schrad.) (watermelon A; Gwigongja, Monsanto Korea Inc., South Korea: watermelon B; Obok-kkul, Daenong Seeds Co., South Korea), one commercial melon (Cucumis melo L. var. reticulatus Naud. melon A; Beauty, Monsanto Korea Inc., South Korea), and one commercial oriental melon (Cucumis melo L. var. makuwa Mak. melon B; Hwanggeum-kkul-euncheon, Daenong Seeds Co., South Korea) cultivars in addition to bottle gourds were sown in plastic plug trays filled with commercial compost. When each seedling had developed cotyledons, more than 50 seedlings of each cucurbit were test-grafted onto the transgenic and wild-type bottle gourds (8 combinations) by using the splice-grafting method (Lee 2003; Lee and Oda 2003). We monitored the fidelity of the graft-union for each combination for 2 weeks post grafting.

Field cultivation of grafted cucurbit plants, plant measurements, watermelon fruit set, and watermelon fruit measurements

In the summer of 2006, grafted cucurbit and self-rooted bottle gourd seedlings grown for 3 weeks after sowing were transplanted into three greenhouses located in Suwon city, South Korea. Efforts were made to maintain similar environmental conditions within all three greenhouses. Eight grafting combinations and two self-rooted bottle gourd genotypes were analyzed in each greenhouse, and each grafting combination or bottle gourd genotype consisted of five or six plants separated by one meter intervals and randomized in each greenhouse. After 35 days of cultivation in soil, several indices for plant growth were estimated for each grafting combination and self-rooted bottle gourd genotype. The data on growth were analyzed by t-test at the 5% level. To set fruits, artificial pollinations were conducted with paint brushes using the pollen collected from the male flowers of the same plants after 45–50 days in the greenhouse. Two watermelon fruits per plant were set at the similar position (approximately eleventh node) of watermelon branches. In addition to fruit weight, fruit height, and fruit diameter, osmotic pressure and soluble solid contents (°Brix) in fruit juice were measured by a micro-osmometer (Model 210, The Fiske, Norwood, Massachusetts) and a refractometer (RA-520 N, Kyoto Electronics, Japan) according to the manufacturer’s instructions, respectively.


Generation of sCAX2B-expressing bottle gourd

A total of 1,204 cotyledon explants co-cultivated with Agrobacterium were transferred to the selection medium with 100 mg l−1 kanamycin monosulfate. Regenerated shoots from 553 cotyledon explants were sub-cultured on the rooting medium with 50 mg l−1 kanamycin monosulfate. Out of the 243 shoots that elongated and rooted, 125 rooted plantlets were successfully acclimatized in a controlled growth room. Based on the PCR analysis for sCAX2B, the transformation frequency of bottle gourd was 0.8% (Fig. 1a). This transformation frequency was lower than that of our previous bottle gourd transformation experiments (Han et al. 2005), where a different Agrobacterium strain (AGL1) and selectable marker gene (Bar) were used. Nadolska-Orczyk and Orczyk (2000) reported that the Agrobacterium-mediated transformation efficiency in pea (Pisum sativum L.) was variable depending on the bacterial strain, selection gene, and acetosyringone. Three randomly selected PCR-positive transformants were subjected to Southern blot analysis for sCAX2B to confirm independent integration and determine copy number (Fig. 1b). As demonstrated in Fig. 1c, RNA gel blot demonstrated that the sCAX2B transcripts accumulated in all of the tested transgenic T0 plants. We selected the line termed TG137 to generate the T3 progenies along with T2, which contained one copy of the construct with strong expression of the transcript (Fig. 1b, c). We also confirmed that the majority of the TG137 progeny were derived from a non-segregating T2 population as the transgene expression was equivalent except in (137T3-15; Fig. 1d). The same TG137 T3 population was used for further experiments such as hydroponic tests, growth measurements, and test-grafting with other cucurbit crops in a greenhouse. The inability to detect Arabidopsis sCAX2B homologs in the wild-type bottle gourd plants by Southern blot, Northern blot, and RT-PCR analyses may be due to the stringency of hybridization or amplification methods used in this study.
Fig. 1

Molecular analyses of sCAX2B-expressing bottle gourds. a PCR detection of sCAX2B genes in the putative transgenic bottle gourd plants. Expected 670-bp fragments of sCAX2B were amplified in the asterisked plantlets (lanes 35, 126, 137, 180, 183, 196, 213, 222, and 225 lines). Lanes SM, molecular size marker; P, positive control (plasmid DNA); C, negative control (wild-type bottle gourd). b Southern blot analysis of the transgenic bottle gourd plants. 5–10 µg of genomic DNA of randomly selected PCR-positive bottle gourd plants were digested with XbaI, and hybridized with the sCAX2B probe. LanesC, negative control (wild-type bottle gourd); P, positive control (plasmid DNA); 35, 137, and 213, transgenic bottle gourds lines with sCAX2B gene. c Northern blot analysis of transgenic bottle gourd plants. 7 µg of total RNA from bottle gourd leaves were hybridized with the sCAX2B probe. Ethidium bromide-stained rRNA (bottom) is shown as a loading control. LanesC, negative control (wild-type bottle gourd); 35, 137, 213, and 225, transgenic bottle gourd lines with sCAX2B transcripts. d Detection of sCAX2B transcripts by RT-PCR in total RNA from transgenic bottle gourd plants in T3 generation (No. 137 line). 670-bp sCAX2B-specific fragments were amplified by RT-PCR in all tested plants. LanesSM, molecular size marker; C, negative control (wild-type bottle gourd), 137T3-1~137T3-15, independent plants in T3 generation of No.137 line; HeLa(+RTase) and HeLa(−RTase), human HeLa RNA as controls to confirm the reverse transcriptase activity during the first strand cDNA synthesis

Ionic alteration and enhanced biomass of sCAX2B-expressing bottle gourd plant grown under hydroponic culture

We previously demonstrated that the sCAX2B-expressing potatoes do not alter their morphology or growth traits, and contain increased Ca2+ in their tubers (Kim et al. 2006) while the deregulated sCAX1-expressing plants show Ca2+ deficiency-like symptoms (Hirschi 1999; Park et al. 2004, 2005a). In the field, potato plants exhibit altered growth parameters compared to bottle gourd. Therefore, we introduced a hydroponic culture to precisely assess the morphological and ionic alterations of the transgenic bottle gourd. Neither wild-type nor sCAX2B-expressing bottle gourds showed any adverse phenotypes in the progressively fortified MS nutrient media (Murashige and Skoog 1962; Fig. 2a). Interestingly, the total biomass (dry weight) of the transgenic bottle gourd plant was significantly (27.4%) greater than controls (wild-type) (data not shown). The total dry weight per plant was 2,848 ± 108.9 and 3,628 ± 153.7 (mean ± SE) mg in the wild-type and transgenic bottle gourd plants, respectively, which indicates that the sCAX2B-expressing bottle gourd plants grew larger than wild-type. To ascertain whether sCAX2B expression altered ion levels, cations including Ca2+ were measured in the shoots and roots of transgenic bottle gourds. The roots of sCAX2B-expressing plants showed an 8.7% increase in Ca2+ level, while there were no differences in the Ca2+ level of shoots as well as other metal ion (Mg2+, Zn2+, Mn2+, Fe2+, and Cu2+) levels in both plant parts (Fig. 2b). In addition, sCAX2B-expressing bottle gourd plants showed dramatic increases in K+ (7.7% in root) and Na+ (44.6 and 27.2% in shoot and root, respectively) levels (Fig. 2b).
Fig. 2

Phenotype of sCAX2B-expressing bottle gourd plants under a hydroponic culture system, and their cation contents in shoots and roots. a Phenotypes of sCAX2B-expressing bottle gourd plants growing under a hydroponic culture system. The media were progressively fortified every 6 days in order of 0× MS, 1/8× MS, 1/2× MS, and final 2× MS. After 10 days of supplying the final 2× MS, the contents of cations were measured by inductively coupled plasma emission spectrophotometer. b Data represent the mean values (±SE) obtained from three independent hydroponic culture systems in which eight bottle gourd seedlings were established. Values in each sub-frame of B frame followed by different letter are significantly different at the 5% level (t-test). C, control (wild-type bottle gourd); T, sCAX2B-expressing bottle gourd

Improved graft-unions using sCAX2B-expressing bottle gourd

Bottle gourd is mainly used as a rootstock for watermelon (Han et al. 2004); however, it is interesting to examine whether sCAX2B-expressing bottle gourd can be used as a rootstock with other cucurbit scions. Melon or oriental melon seedlings grafted onto a bottle gourd are not normally able to grow and set fruits despite the graft-unions success owing to their physiological incompatibility (Lee and Oda 2003). All trials of eight graft combinations using two watermelon, one oriental melon, and one melon cultivars as scions were successful to generate graft-unions with bottle gourd rootstocks followed by wound repair (Fig. 3a, b). These results indicate that the transgenic bottle gourd expressing sCAX2B enhances the graft-union with other cucurbit crops without any apparent adverse effects during physical cohesion.
Fig. 3

Generation of graft-unions in cucurbit seedlings grafted onto the sCAX2B-expressing bottle gourds. a Grafted cucurbit seedlings being acclimatized (taken in 10 days after grafting). More than 50 seedlings of each cucurbit were test-grafted onto the transgenic and wild-type bottle gourd seedlings, respectively. b Acclimatized seedlings in all combinations of cucurbit crops and bottle gourds (taken in 14 days after grafting)

Enhanced growth of sCAX2B-expressing bottle gourd, and its effect on the grafted cucurbit scions under soil cultivation

The self-rooted transgenic bottle gourds and four genotypes of cucurbits grafted onto the transgenic rootstock were cultivated in three similar environments. To assess the performance of sCAX2B-expressing bottle gourd as a rootstock, we measured the growth indices 35 days after greenhouse transplanting. The shoot length of self-rooted sCAX2B-expressing bottle gourds was significantly longer (17% increase) than the wild-type because of the increased number of nodes (Table 1). In hydroponic culture mentioned above, sCAX2B-expressing bottle gourd demonstrated enhanced biomass at the young developmental phase. Our findings here indicate that the enhanced growth in sCAX2B-expressing bottle gourd is maintained through the adult developmental phase. These observations suggest that sCAX2B-expressing bottle gourd rootstocks can positively impact both the watermelon scions, and the grafting-incompatible melon scions (Table 1; Fig. 4c). A majority of the growth indices in two watermelon genotypes grafted onto the sCAX2B-expressing bottle gourds were improved (Table 1). As reported previously, the seedlings of melon/bottle gourd (scion/rootstock) combinations failed to grow continuously, and did not set any fruit (Fig. 4a arrows; Lee and Oda 2003).
Table 1

Enhanced growth of the watermelons grafted onto the sCAX2B-expressing bottle gourds as well as the self-rooted transgenic bottle gourd itself when cultivated in the greenhouse


Shoot length (cm) (mean ± SE)

No. of node (mean ± SE)

No. of lateral branch (mean ± SE)

Diameter of main stem (mm) (mean ± SE)

Length of the largest (mean ± SE)

Width of the largest (mean ± SE)

Self-rooted bottle gourd


96.1 ± 4.7 b

12.8 ± 0.3 b

4.3 ± 0.3

8.0 ± 0.1

27.4 ± 0.6

23.2 ± 0.5


112.4 ± 5.1 a

14.3 ± 0.3 a

4.5 ± 0.3

8.1 ± 0.1

28.5 ± 0.5

23.0 ± 0.4

Mini watermelon grafted onto the bottle gourd

    Wild type

44.4 ± 2.2 b

10.2 ± 0.4 b

2.8 ± 0.2 b

5.3 ± 0.2 b

23.2 ± 0.3 b

15.7 ± 0.7 b


53.8 ± 2.7 a

12.3 ± 0.4 a

4.3 ± 0.4 a

6.4 ± 0.2 a

26.3 ± 0.8 a

18.7 ± 0.7 a

Normal watermelon grafted onto the bottle gourd

    Wild type

57.9 ± 3.0 b

12.3 ± 0.4 b

3.2 ± 0.2

5.9 ± 0.1 b

25.0 ± 0.7 b

16.7 ± 0.3 b


72.3 ± 5.2 a

14.0 ± 0.4 a

3.8 ± 0.3

6.7 ± 0.2 a

25.7 ± 0.6 a

19.1 ± 0.6 a

The indices were estimated to be 35 days after greenhouse transplanting

The data represent the mean values of 32 plants of self-rooted bottle gourds, and 10 plants of grafted-watermelon. Values in a column within the same row followed by a common letter are not significantly different at the 5% level (t-test)
Fig. 4

Growth aspects of cucurbit crops grafted onto the bottle gourd expressing sCAX2B in the greenhouse. a Three greenhouses were used for cultivation of grafted cucurbit crops and self-rooted bottle gourds. Arrows and arrow heads indicate melon/bottle gourd (scion/rootstock) and watermelon/bottle gourd combinations, respectively. b The fruits of mini watermelon cultivar (upper) and normal watermelon cultivar (lower) just before harvesting. c The growth situation of melon/bottle gourd combinations taken in 72 days after greenhouse transplanting. The plant volumes of melon/sCAX2B-expressing bottle gourd combinations were slightly bigger than wild-type bottle gourd combinations

The two watermelon cultivars/sCAX2B-expressing bottle gourd combinations grew vigorously during the cropping period (Fig. 4a arrow heads). We harvested the watermelon fruits after 45 days of artificial pollination (Fig. 4b, c), and then characterized the watermelon fruit. The fruit weight, fruit height, and fruit diameter were indistinguishable between the two genotypes of watermelon/wild-type bottle gourd and two genotypes of watermelon/sCAX2B-expressing bottle gourd combinations. However, when the soluble solids (°Brix) and osmotic pressure of watermelon fruit juice were measured, the fruits from transgenic bottle gourd combinations recorded higher osmotic pressure and more soluble solids than those with the wild-type bottle gourd combination (Table 2).
Table 2

Alterations in watermelon fruits from the grafting combination of watermelon scion/sCAX2B-expressing bottle gourd rootstock when cultivated in the greenhouse

Grafting combination (scion/rootstock)

Osmotic pressure of watermelon fruit juice (mOsm/kg)

Soluble solids (°Brix)

Mini watermelon/wild-type bottle gourd

545 ± 11.3 b

10.9 ± 0.2 b

Mini watermelon/transgenic bottle gourd

577 ± 6.5 a

11.6 ± 0.2 a

Normal watermelon/wild type bottle gourd

522 ± 7.0 b

9.9 ± 0.1 b

Normal watermelon/transgenic bottle gourd

546 ± 5.8 a

10.5 ± 0.1 a

The values were estimated for the fruits harvested in 45 days after artificial pollination. The data represent the mean values of more than ten fruits. Values in a column within the same row followed by a common letter are not significantly different at the 5% level (t-test)


The production of grafted plants first began in Japan and Korea in the late 1920s with watermelon grafted onto bottle gourd rootstock (Lee 1994). In this study, we have evaluated sCAX2B-expressing bottle gourd rootstocks as a mechanism to improve watermelon quality. Genetic modifications to CAX transporters, such as sCAX2B, allow scientists an opportunity to engineer plants that specifically accumulate particular nutrients (Shigaki and Hirschi 2006). A modified CAX may also enhance plant growth. Ectopic expression of deregulated ArabidopsisCAX1 (sCAX1) showed Ca2+ deficiency-like symptoms in Solanaceae species (Hirschi 1999; Park et al. 2005a, b). Meanwhile, soil grown tomato and potato expressing other cation/H+ antiporter (CAX4 and sCAX2B) had no morphological alterations, and contained increased Ca2+ in their fruits and tubers (Kim et al. 2006; Park et al. 2005a). These results imply that the ectopic expression of CAX antiporters produces variable effects depending on the particular plant species and the introduced CAXs.

In this study, we demonstrated that the sCAX2B-expressing bottle gourd has no adverse phenotypic alterations, reaffirming our previous report in potato expressing the same CAX variant. Interestingly, the biomass of the transgenic bottle gourd plants was dramatically increased. In addition, the Ca2+ levels were increased with no significant changes in Mn2+ levels (Fig. 2b). These findings are consistent with previous studies in sCAX2B-expressing yeast cells and potatoes (Kim et al. 2006; Shigaki et al. 2003).

The sCAX2B-expressing bottle gourd grown under a hydroponic culture system contained more K+ in roots and more Na+ in both roots and shoots parts than controls (Fig. 2b). CAX transporters influence the proton gradients in plant cells and thus these Na+ and K+ adjustments may be in an indirect consequence of CAX expression (Zhao et al. 2008). The increased biomass phenotypes mediated by sCAX2B are similar to that of plant overexpressing the vacuolar H+-pyrophosphatase AVP1 (Gaxiola et al. 2001). The enhanced growth by manipulation of AVP1 expression reveals a close connection between endomembrane proton pumps and auxin-mediated signals (Schumacher 2006). Another illuminating example of the breadth of functions of antiporters is seen with yeast NHX1 (Brett et al. 2005). While it transports both K+ and Na+, it also regulates luminal and cytoplasmic pH to control vesicle trafficking out of the endosome.

We speculate that the sCAX2B transporter in the transgenic bottle gourd plant indirectly influences K+ and Na+ ions followed by effective water use, which contributes to increased biomass of transgenic plant. Further studies, focusing on the other domains of CAX2 and the relation of the sCAX2B with other cellular transporters, pumps, and channels especially water channel, may validate this model.

The abilities of rootstock to uptake and transport water and nutrients can affect the growth of scions (Cohen and Naor 2002; Estan et al. 2005; Ruiz et al. 1997). The biomass of watermelons grafted onto the sCAX2B-expressing rootstocks was enhanced compared to controls (Table 1), which indicates that the enhanced vigor of rootstock by sCAX2B expression positively influenced the scion. In addition to the biomass increase of aerial parts (photosynthetic source) in watermelon/sCAX2B-expressing combinations, the fruit (strong photosynthetic sink) harvested from the transgenic combination recorded higher osmotic pressure and more soluble solids than control combinations (Table 2). Hence we suggest that sCAX2B-expressing bottle gourd rootstock has a higher hydraulic conductance than controls, which is connected with the biomass increase of the scion.

A wild-type bottle gourd is an important rootstock for cucurbit crops and has distinct growth and developmental habits distinct from those of potato. For example, the vining stage starts 14 days after emergence and is characterized by very rapid vine elongation. In fact, frequently Ca deficiencies are caused by rapidly growing or expanding distal tissues (Ho and White 2005). Thus, future studies need to be directed toward investigating the trait alterations caused by sCAX2B expression in a plant species with a rapid growth pattern.

In this report, we have demonstrated the ionic alterations in the bottle gourd expressing an Arabidopsis variant transporter (sCAX2B), and the ability of sCAX2B to enhance the biomass and quality of the wild-type watermelon grafted onto the transgenic rootstock as well as the transgenic rootstock itself.


This research was supported in part by the Cooperative Research Project (Agenda 2-8-14) between the NIHHS of RDA of Republic of Korea and the KSU of USA.

Copyright information

© Springer Science+Business Media B.V. 2009