Cucurbit Genetics Cooperative
Other Crop Genetics Cooperatives
Home About Membership Reports Gene Lists Conferences Links Search NCSU
Cucurbit Genetics Cooperative Report 17:102-105 (article 31) 1994

Determination of Watermelon Ploidy Level using Flow Cytometry

X.P. Zhang and B.N. Rhodes; J.F. Whitesides

Horticulture Department, Clemson University, Clemson, SC 29634-0375, Department of Poultry Science, Clemson University, Clemson, SC 29634-0379

Several approaches have been used to identify ploidy level in watermelon, citrullus lanatus. Generally, ploidy levels are suggested by plant morphology, leaf shape/size, pollen size, guard cell size, chloroplast number in guard cells, fruit shape/size and the size of blossom end scar. The ploidy level is determined directly by chromosome number (Angel, 1969; Kihara, 1951; Lower et al., 1969). Chromosome numbers are considered proof of ploidy level. However, because of the difficulties associated with chromosome staining and metaphase arrest, chromosome counts are time consuming and experience dependent. Tetraploid breeding for triploid parents os one of the major breeding objectives in watermelon, and therefore, ploidy level determinations are frequently made. Unfortunately, the watermelon chromosome observations are difficult because of the small size and special difficulties associated with chromosome staining nd metaphase arrest. An efficient and reliably ploidy determination technique other than chromosome counting is specially useful to expedite watermelon tetraploid breeding.

Flow cytometry is a powerful technique for estimating plant nuclear DNA content because this technique permits sensitive measurement of fluorescence intensity of large numbers of stained nuclei within seconds (Agumuganathan and Earle, 1991a). Flow cytometry also permits greater statistical accuracy than either conventional Feulgen microdensitometry or reassociation kinetics (Leutwiler et al., 1984). Nuclear DNA content of more than 100 important plant species have been measured by flow cytometry (Arumuganathan and Earle, 1991a, 1991b). Chromosome number increases or decreases will results in the increase or decrease of nuclear DNA content. therefore, flow cytometry can be used for determining ploidy level (De Laat et al., 1987; De Rocher et al., 1990). This experiment was conducted to confirm putative tetraploids regenerated from several diploid watermelon genotypes using tissue culture.

Materials. Putative tetraploid regenerants 4xG17AB (G4x), 4xFY91, (YF4x) and 4AuP (AuP4x) and their original diploids of 617AB (G2x), and their original diploids 617AV (G2x). YF91-1 (YF2x and AuP7 (Aup2x), respectively, were used in this experiment. Leaf materials for nuclear isolation were taken from greenhouse-grown plants in October 1993.

Methods. Nuclear isolation procedures were from Arumuganathan and Earle (1991a), and Michaelson et al. (1991) with minor modification. About 25 milligrams of newly expanded true leaf tissue were sliced as thin as possible in Mgso4 buffer. The nuclei isolated were stained with propidium iodide (PI) in a solution containing DNase-free RNase. Samples were analyzed on an Epics 751 flow cytometer (Coulter Corporation) equipped with a data acquisition system. Excitation of PI was provided by the 488 nm line (400 mW) of an argon laser (model I-90, Coherent) and the red fluorescence emitted by PI was collected through a 635 nm band pass filter. Chic ken red blood cells (CRBC), with nuclear DNA content of 2.397 pg as calibrated by using male human white blood cells (7.0 pg/nucleus) (Tiersch and Chandler, 1989), were used as an internal standard. Five thousand cells per sample were analyzed and the data were represented as histograms. The formula, plant nuclear DNA amount = (position of plant nuclear peak/position of CRBC nuclear peak) x 2.397 pg, was used to calculate DNA content of different genotypes with different ploidy levels.

Results. High quality nuclear preparation was obtained from watermelon leaf tissue. Newly expanded true leaves provided a better source for nuclear isolation than cotyledons. All nuclear preparations from the true leaves gave sharp peaks (Figure 1). Nuclear preparations from cotyledons gave "fat" peaks as a results of a large amount of starch particles in the preparation (histograms are not shown). The yields of nuclei from all of the leaf materials were good enough for flow cytometry to distinguish different categories of the nuclei. However, more nuclei were isolated from the same amount of diploid leaf tissue than from the putative tetraploid leaf tissue.

Nuclei of all six preparations produced two peaks of fluorescence intensity besides the peak produced by CRBC (Figure 1a,b,d,e,f), one corresponding to G0 - G1 phase cells (2C and 4C complement of DNA for diploid and tetraploid, respectively) and the other to G2 + M phase cells (4C and 8C complement of DNA for diploid and tetraploid, respectively). The G2 + M nuclei produced twice the amount of fluorescence intensity of the G0 + G1 nuclei because they contained twice as much DNA. Mixed nuclei of G2x and G4x produced four peaks of fluorescence intensity, for 2C, 4C, 2C + 4C and 8C complement of DNA, besides the peak produced by CRBC (Figure 1c). The positions of fluorescence peaks of the putative tetraploids were almost in the expected position. Therefore, the tetraploidy pf these putative tetraploid regenerants were confirmed by the histogram of flow cytometric analysis.

The peak position of G0 - G1 + 0.111 for diploids, and 1.593 + 0.103 for tetraploids. Nuclear DNA content of tetraploids appeared 6% smaller than the expected amount from their original diploids. However, the chi-square statistic indicated that the measured nuclear DNA content of the tetraploids has a probability of 0.96-0.975 to be the expected content. The nuclear DNA content determined in our experiment was very close to the reported numbers 0.88 and 0.90 pg (Arumuganthan and Earle, 1991b). Watermelon genome size is small compared to wheat (33.09 pg), leek (50.27 pg), onion (31.69 pg) and tomato (1.88-2.07 pg), and is only about three times of that of Arabidopsis (Arumuganathan and Earle, 1991b), Therefore, a high density watermelon genetic map would not require very many markers.

In summary, the published plant nuclear isolation protocols (Arumuganthaan and Earle, 1991ba; Michaelson et al. 1991) worked well for watermelon nuclei isolation from a young true leaf. Flow cytometry is a very rapid and efficient technique for determining watermelon ploidy and, therefore, is helpful for watermelon polyploid breeding. The histogram obtained from flow cytometric analysis can be directly used for determining tetraploidy if the original diploid nuclei are included in the sample preparation.

Acknowledgments. The flow cytometer was made available through the Biotechnology Initiative of the College of Agriculture Science and the South Carolina Experiment Station. For information on polyploid screening of plants contact Dr. John F. Whitesides at (803)-656-0460.

Table 1. Nuclear DNA content (pg) of diploid and tetraploid watermelons as determined by flow cytometry.








Mean S.E.

0.878 + 0.111
1.593 + 0.103

Literature Cited

  1. Angel, I. 1969. A comparative study of en, 3n and 4n forms of Citrullus vulgaris.Genitika 5:103-111.
  2. Arumuganathan, K. and E. Earle. 1991a. Estimation of nuclear DNA content of plants by flow cutometry. Plant Mol. Biol. Reporter 9:221-231.
  3. Arumugenathan, K., and E. Earle. 1991b. Nuclear DNA content of some important plant species. Plant Mol. Biol. Reporter 9:208-218.
  4. De Laat, R.M.M., W. Gohde, and M.J.D.C. Vogelzang. 1987. Determination of ploidy of single plants and plant populations by flow cytometry. Plant Breeding 919: 303-307.
  5. De Rocher, E.J., K.R. Harkins, O.W. Galbraith, and H.J. Bohnert. 1990. Developmentally regulated systemic endopolyploidy in succulents with small genomes. Science 250:99-101.
  6. Kihara, H. 1951. Triploid watermelon. J. Amer. Soc. Hort. Sci. 58:217-230.
  7. Leutwiler, L.S., B.R. Hough-Evans, and E.M. M<eyerowitz. 1984. The DNA of Arabidopsis thaliana. Mol. Gen. Genet. 194:15-23.
  8. Lower, R.L., and K.IU. Johnson. 1969. Observation of sterility of induced autotetraploid watermelon. J. Amer. Soc. Hort. Sci. 94-367-369.
  9. Michaelson, M.J., H.J. Price, J.R. Ellison, and J.S. Johnston. 1991. Comparison of plant DNA contents determined by Feulgen microspectrophotometry and laser flow cytometry. Amer. J. MBot. 78:83-188.
  10. Tiersch, T.R. and R.W. Chandler. 1989. Chicken erythrocytes as an internal reference for analysis of DNA content by flow cytometry in grass carp. Transaction of the American Fisheries Society 118:713-717.
Home About Membership Reports Gene Lists Conferences Links Search NCSU
Department of Horticultural Science Box 7609North Carolina State UniversityRaleigh, NC 27695-7609919-515-5363
Page citation: Wehner, T.C., Cucurbit Genetics Cooperative;
Created by T.C. Wehner and T. Ng, 1 June 2005; design by C.T. Glenn;
send questions to T.C. Wehner; last revised on 11 December, 2009