Russian Journal of Genetics, Vol. 38, No. 5, 2002, pp. 526530. Translated from Genetika, Vol. 38, No.

5, 2002, pp. 641646.

Original Russian Text Copyright 2002 by Tsvetova, Elkonin.

PLANT
GENETICS

Instability of the Ploidy Level in Autotetraploid Sorghum Plants

from a Line with Variable Male Fertility
M. I. Tsvetova and L. A. Elkonin
Agricultural Research Institute for Southeast Region, Saratov, 410020 Russia; e-mail: elkonin@mail.saratov.ru
Received March 13, 2001; in final form, October 23, 2001

AbstractIt was found that, in some of the plants in generations C1C5 of induced tetraploids in the semisterile sorghum line AS-1-30, more than 30% of pollen grains (PGs) have sizes typical of haploid PGs. Pollen of
these plants was used to pollinate different lines of sorghum with cytoplasmic male sterility (CMS). Diploid
hybrids were obtained, which confirms the presence of haploid PGs in the tetraploids studied. When tetraploid
plants with an increased frequency of haploid PGs were pollinated with pollen of plants from fertile diploid
lines, diploid hybrids were also obtained. This demonstrates that the tetraploids studied had haploid egg cells.
In generation C4, a chimeric plant was found; one of its shoots was diploid (judging from morphological characters) and produced as many as 99% of haploid PGs. It is assumed that haploid gametes in autotetraploids are
formed through somatic reduction of chromosomes at different ontogenetic stages.

INTRODUCTION
The constancy of the number of chromosomes is a
fundamental biological phenomenon ensuring the stability of the transmission of genetic information in generations. However, polyploidization of differentiating
somatic cells occurs during ontogeny of the majority of
animal and plant species studied. This phenomenon is
termed polysomaty [1, 2]. After cell proliferation in
apical layers of the meristem, polyploidizing mitoses
start occurring even in its distal layers; however, the
basic number remains unchanged in germline cells,
which ensures genetic consistency of the species. Here,
we discuss the consequences of the changes in the number of chromosomes in apical meristematic cells from
which generative organs and gametes are formed in the
course of ontogeny. This phenomenon is termed mixoploidy [3]. Different authors use this term when referring
to two separate phenomena: (1) the presence of cells
with chromosome numbers forming aneuploid series in
plant tissues and (2) the presence of cells with different
levels of ploidy. Hereinafter, only the presence of cells
with different chromosome numbers that are multiples
of the basic number in plant tissues is termed mixoploidy.
There are numerous data on the presence of cells
with various ploidies in apical growing points, among
mother cells of microspores that appear as a result of
experimental exposure to external factors or spontaneously [46]. In some cases, the meristem of the same
plant contains cells with several ploidy levels, with the
chromosome numbers in some cells not being multiples
of 2 [79]. Plants have been described in which spontaneous doubling of the chromosome number has
occurred [1013]. In a few plants, cells with reduced
chromosome numbers may be found [11, 14].

In this study, we discovered instability of the ploidy

level in autotetraploids induced in the sorghum line AS-1.
This was obtained by self-pollination of a male-fertile
plant regenerated from the tissue culture of a plant with
cytoplasmic male sterility (CMS) [15]. The AS-1 line is
also characterized by elements of apomixis in female
generative system. Polyploidization of plants of this
line by means of colchicine led to an increase in the frequency of apospore structures in ovules [16].
MATERIALS AND METHODS
The AS-1 line was obtained from a plant regenerated from callus cultures that were induced from
fragments of young panicles of a completely sterile
[A1]Saratovskoe-3/S-752 F2 hybrid plant. This line
is characterized by partial male fertility: different
panicles of the same plant and different layers of the
same panicle contain 0 to 66% of viable pollen
grains (PGs) [15]. To obtain polyploids, we used
seedlings of subline AS-1-30, which originates from
a single plant.
Tetraploids were obtained via treatment of the apical meristems of the seedlings with colchicine for 24 h
[17]. Chromosomes were counted in root meristems
fixed in acetic alcohol (1 : 3) after pretreatment with a
solution of -monobromonaphthalene and stained with
acetohematoxylin. The panicles of the plants were isolated with parchment bags before blooming. For cytological analysis, pollen was shaken from the blooming
part of the panicle and analyzed in temporary preparations stained with I2KI solution. The diameters of PGs
were measured by means of an eyepiece micrometer.

1022-7954/02/3805-0526$27.00 2002 MAIK Nauka /Interperiodica

INSTABILITY OF THE PLOIDY LEVEL IN AUTOTETRAPLOID SORGHUM PLANTS

527

RESULTS
The size of PGs is a generally accepted criterion for
estimating the effectiveness of polyploidization factors,
because diploid PGs formed in the tetraploid sector of
the inflorescence have a larger diameter than haploid
PGs, and their amount in pollen is positively correlated
with the size of the polyploid sector. In the sorghum,
polyploid plants are easily detected in the offspring of
the inflorescences that contain more than 20% of large
PGs; both fertile and sterile pollen may be analyzed to
detect polyploidization. Earlier, we demonstrated that
sorghum fertile PGs with diameters greater than
54.2 m can confidently be considered diploid. Sterile
PGs of diploid sorghum plants are usually smaller than
41.7 m in diameter [17, 18]. Hereinafter, we will refer
to PGs with diameters larger than these values as
large PGs.
We analyzed pollen of 40 AS-1-30 plants treated
with a 0.2% colchicine solution for 24 h. Eight plants
with increased proportions of both fertile and sterile
large PGs were found (Table 1).
The AS-1-30 line is characterized by a high sterility
of pollen, and the effect of colchicine further increases
sterilization. Therefore, there was only a limited possibility to obtain tetraploid offspring as a result of selfpollination of these panicles. However, we obtained a
certain amount of kernels.
Cytological analysis of generation C1 revealed both
tetraploid and diploid plants. The tetraploids phenotypically differed from diploid sibs in a drastically
decreased growth, darker leaves, larger glumes, and an
increased panicle density. Analysis of the pollen of C1
plants identified as tetraploid according to the results of
cytological analysis of root meristems demonstrated
that some PGs had the sizes typical of haploid PGs
(Table 2). The pollen resembled in appearance the pollen of C0 mixoploid plants treated with colchicine. This
was also true for all plants from generations C1C3
examined in our study (figure).
To confirm that PGs smaller than 50.1 m in diameter were haploid, we crossed tetraploid plants with an
increased frequency of such PGs with plants from stable CMS lines, which did not set seeds in the absence
of pollen from other lines. Crossing between diploids
and tetraploids usually yield caryopses with a triploid

50 m
Size polymorphism of pollen grains in an autotetraploid
AS-1-30 plant (C4). The arrow indicates a haploid pollen
grain.

embryo and poorly developed tetraploid endosperm.

However, we obtained both underdeveloped and plump
seeds. Underdeveloped seeds could not germinate even
when cultivated in vitro. Plump seeds also had a
decreased germinating capacity; however, they could
produce diploid plants. Some of these plants were
hybrids, whereas the others were phenotypically identical to the plants of the mother line (matroclinal plants)
(Table 3). The formation of diploid hybrid plants unambiguously demonstrates that the tetraploid studied had
haploid PGs. Apparently, matroclinal plants grew from
embryos developing through mentoral apomixis, when
the embryo develops from the ovicell as a result of parthenogenesis, and endosperm forms after fertilization
of the nuclei of the central cell of the embryo sac. Note
that seeds were set at a low rate in these crosses. It is
possible that some of the pollen grains smaller than
50.1 m in diameter were aneuploid PGs, which are
characterized by decreased viability.
To detect haploid embryo sacs in the tetraploids
studied, we selected the panicles with a high sterility of
pollen and a high proportion of sterile PGs with sizes
typical of diploids (smaller than 41.7 m). Six of these
panicles were pollinated with pollen of fertile analogues of CMS lines used in the preceding crosses. As
a result, we obtained one hybrid plant.

Table 1. The amount of large pollen grains in pollen of some AS-1-30 plants treated with colchicine and the ploidy of their
offspring
Nos. of the plant
and panicle

Number
of pollen grains

208-5

Percentage of large pollen grains

Number of plants in C1

fertile

sterile

2n

4n

200

33.9

13.2

236-6

200

18.0

10.5

239-5

200

19.5

50.0

25

21

RUSSIAN JOURNAL OF GENETICS

Vol. 38

No. 5

2002

528

TSVETOVA, ELKONIN

Table 2. Distribution of fertile pollen grains with respect to diameter in some tetraploid AS-1-30 plants
Percentage of pollen grains with diameters:*
No. of family No. of plant

4.5

5.0

5.5

6.0

6.5

7.0

7.5

37.5

41.7

45.9

50.1

54.3

58.5

62.7

0.9

1996, C1
208-5

16.3

3
239-5

21.6

40.5

10.8

10.8

5.4

12.7

55.5

16.4

9.1

10.1

23.0

23.7

10.1

14.9

18.2

2.0

8.0

32.0

26.0

30.0

2.0

48.0

12.0

4.0

20.0

36.0

20.0

20.0

1999, C4
239-5

29-1

8.0

29-3

4.0

28.0

2000, C5
239-5

41.7

13.3

16.3

21.6

40.5

10.8

14

2.0

22.0

24.0

10.0

26.0

16.0

* In each specimen, 50 pollen grains were measured. Diameters of pollen grains are indicated in scale divisions of the eyepiece micrometer
(multiplying factor, 8.35 m) and in micrometers (the upper and lower lines, respectively).

Table 3. Results of pollination of CMS sorghum plants with pollen of tetraploid AS-1-30 plants
CMS line
[A1] Saratovskoe 3
Control*
[9E] Zheltozernoe 10
Control
[A3] Feterita 14
Control
[M35] Pishchevoe 1
Control
[A2] KVV-181
Control

Number of seeds

Number of plants in the offspring

Number of pollinated panicles

normal

undersized

hybrid

matroclinal

11
10
20
16
7
30
2
10
8
13

23
0
36
0
1
1
1
0
1
0

0
0
5
0
1
0
4
0
1
0

2
0
5
0
0
0
2
0
0
0

2
0
30
0
0
0
0
0
0
0

* As control, isolated nonpollinated panicles of sterile lines were used.

Thus, the data obtained indicate that tetraploids can

form both male and female haploid gametes. All the
remaining offspring was tetraploid; it is obvious that
these plants resulted from self-pollination with the few
fertile PGs that were contained in the pollen of the
maternal panicles.
The instability of the ploidy level was not observed
in the generative system of the induced AS-1-30 tetraploids alone. In addition, we found a plant (no. 29-5) in
generation C4 one of the shoots of which had morphological characters typical of the original diploid line.

Large PGs only accounted for 1% of pollen produced

by the panicle that had formed on this shoot. Pollination
of this panicle with pollen from the sorghum cultivar
Zheltozernoe 10 yielded more than 100 grains, from
which we obtained hybrid plants.
DISCUSSION
The results obtained indicate that tetraploids from
the sorghum line studied can form male and female
haploid gametes. This phenomenon is inherited and is

RUSSIAN JOURNAL OF GENETICS

Vol. 38

No. 5

2002

INSTABILITY OF THE PLOIDY LEVEL IN AUTOTETRAPLOID SORGHUM PLANTS

consistently expressed in sequential seed generations

up to C5. Note that we earlier pollinated 73 panicles of
stable CMS lines with pollen of autotetraploid analogs
of other sorghum cultivars and lines (Zheltozernoe 10
and Negrityanskoe Kremovo-Buroe k-3366/2) and Sudan
grass (Krasnoplenchataya k-16/1E) in order to obtain triploids, and not a single normal seed was obtained. This
agrees with the assumption that the AS-1-30 autotetraploid line is genetically predisposed to form haploid
gametes.
There are few published data on the formation of
haploid gametes in autotetraploid plants [1921] and
the cytological mechanisms that may underlie this phenomenon. For example, additional division of the
nuclei formed after meiotic anaphase II was observed
in the tetraploid potato species Solanum chacoense and
S. tarijense [20]. However, we cannot explain the formation of functionally normal PGs in these cases. Even
if the chromosome number is reduced, it is unclear how
the appropriate segregation of homologous chromosomes into the daughter cells is ensured (in a normal
reduction division, this is a result of chromosome
pairing).
Another mechanism of formation of haploid
gametes was described for tetraploid maize hybrids, in
which parental genomes were separated from each
other, and then reduction division occurred in each
genome [21].
However, the tetraploid sorghum line studied produced not only haploid gametes, but also diploid shoots
(one shoot in a C4 plant). It is most likely that the same
mechanismsomatic reduction of the chromosome
numberunderlies both phenomena.
Mutations at early ontogenetic stages are known to
cause the formation of genetically different shoots. At
later stages of ontogeny, similar events result in appearance of smaller sectors of mutant tissues within a shoot,
inflorescence, or anther [22]. The reduction of the chromosome number in plant no. 29-5 had most probably
occurred in early ontogeny. In other tetraploid plants
studied, somatic reduction occurred during the formation of generative organs and yielded small sectors of
diploid tissue, which may have given rise to diploid
gametes. Earlier, the same phenomenon was observed
in a 40-chromosome Sorghum halapence plant, which
produced several diploid shoots [14], and in a Pennisetum americanum plant, whose anthers contained both
diploid and haploid microspores [11].
Based on indirect evidence, the somatic reduction of
chromosomes was detected in tetraploid buckwheat
plants [23]. This explained the presence of diploids in
their offspring, as well as in the offspring of tetraploid
hemp plants [24]. However, it is theoretically possible
that the appearance of diploid plants in the offspring of
tetraploids is a consequence of parthenogenetic development of diploid egg cells; therefore, this fact does not
prove the presence of haploid gametes. The diploid
hybrids obtained in our study are a stronger evidence in
RUSSIAN JOURNAL OF GENETICS

Vol. 38

No. 5

529

favor of the presence of haploid gametes in the tetraploids studied.

We should like to emphasize that the formation of
haploid gametes in the AS-1-30 line is a hereditary phenomenon that was observed in five sequential generations. This unique cytogenetic instability may be determined by the genetic background of the sorghum line
studied. Therefore, this line may prove useful for studying the mechanisms of this phenomenon.
ACKNOWLEDGMENTS
This study was supported by the Russian Foundation for Basic Research, project no. 00-04-48686.
REFERENCES
1. Brodskii, V.Ya. and Uryvaeva, I.V., Kletochnaya poliploidiya. Proliferatsiya i differentsirovka (Cell Polyploidy: Proliferation and Differentiation), Moscow:
Nauka, 1981.
2. DAmato, F., Cytogenetics of Plant Cell Tissue Culture
and Their Regenerates, CRC Crit. Rev. Plant. Sci., 1985,
vol. 3, pp. 73112.
3. Nemec, B., Das Problem der Berfruchtungsvorgange
und andere zytologische Fragen, Berlin: Borntraeger,
1910.
4. Karpechenko, G.D., Eksperimentalnaya poliploidiya i
gaploidiya. Teoreticheskie osnovy selektsii rastenii (Experimental Polyploidy and Haploploidy: A Theoretical
Basis of Plant Breeding), Moscow, 1935, vol. 1, pp. 397
434.
5. Breslavets, L.P., Poliploidiya v prirode i opyte (Polyploidy in Nature and Experiments), Moscow: Akad. Nauk
SSSR, 1963.
6. Kunakh, V.A., Genome Variation of Plant Somatic Cells
and Factors Regulating This Process, Tsitol. Genet.,
1980, vol. 14, no. 1, pp. 7381.
7. Polumordvinova, I.V., Application of the Methods of
Cytological and Embryological Analysis to Breeding,
Selektsiya ovoshchnykh kultur (Vegetable Breeding),
Moscow, 1984, pp. 8795.
8. Lebedeva, S.K., Polyploidy, Aneuploidy, and Mixoploidy in Kentucky Bluegrass, Nauchno-Tekhn. Byul.
Vses. Inst. Rastenievod., 1985, no. 155, pp. 3940.
9. Kashin, A.S., Zalesnaya, S.A., and Titovets, V.V., Morphogenetic Potential of the Pilosella Agamous Complex:
Genome Variation in Populations and Progeny of Individual Plants, Bot. Zh., 2000, vol. 85, no. 12, pp. 1328.
10. Muntzing, A. and Prakken, R., Chromosomal Aberrations in Rye Populations, Hereditas (Lund, Swed.),
1941, vol. 27, nos. 34, pp. 273308.
11. Rao Panuganty, N. and Nirmala, A., Chromosome
Numerical Mosaicism in Pearl Millet (Pennistrum americanum (L.) Leeke), Can. J. Genet. Cytol., 1986, vol. 28,
pp. 203206.
12. Prakash, N.S., Lakshmi, N., and Harini, I., A Note on
Spontaneous Mixoploid in Capsicum, Curr. Sci., 1988,
vol. 57, no. 8, pp. 435436.
2002

530

TSVETOVA, ELKONIN

13. Maletskii, S.I. and Maletskaya, E.I., Self-Fertility and

Agamospermy in Sugar Beet Beta vulgaris L., Genetika
(Moscow), 1996, vol. 32, no. 12, pp. 16431650.
14. Raman, V.S. and Krishnaswami, N., A Chromosomal
Chimera in S. halepense (Linn.), Ind. J. Agric. Sci.,
1955, vol. 25, no. 1, pp. 4650.
15. Elkonin, L.A., Enaleeva, N.Kh., Tsvetova, M.I., et al.,
Partially Fertile Line with Apospory Obtained from Tissue Culture of Male-Sterile Plant of Sorghum (Sorghum
bicolor (L.) Moench), Ann. Bot., 1995, vol. 76, pp. 359
364.
16. Tsvetova, M.I., Belyaeva, E.V., and Enaleeva, N.H.,
Intensification of Tendency to Apomixis in Sorghum
Autotetraploids, Int. Sorghum Millet Newslett., 1998,
no. 39, pp. 6667.
17. Tsvetova, M.I., Study of the Regularities at Experimental Polyploidy in Sorghum, Cand. Sci. (Biol.) Dissertation, St. Petersburg: Vses. Inst. Rastenievod., 1997.
18. Tsvetova, M.I., Elkonin, L.A., and Ishin, A.G., The Use
of the Size of Sterile Pollen Grains to Evaluate Polyp-

19.
20.
21.

22.
23.
24.

loidization in a Male-Sterile Line of Sorghum, Dokl.

Ross. Akad. Skh. Nauk, 1998, no. 1, pp. 1011.
Marks, G.E., McKee, R.K., and Harborn, J.B., Double
Chromosome Reduction in a Tetraploid Solanum,
Nature, 1965, vol. 208, no. 5008, pp. 359361.
Shevtsov, I.A. and Davidenko, N.S., Specific Features of
Meiosis in Diploid and Tetraploid Forms of Potato, Tsitol. Genet., 1974, vol. 8, no. 3, pp. 245247.
Strelchuk, S.I., Segregation of Chromosomal Genomes
in Tetraploid Maize, 3-ya Natsionalnaya konferentsiya
tsitogenetikov s mezhdunarodnym uchastiem (3rd National
Conf. of Cytogenetics with Int. Attendance), Plovdiv,
1984, vol. 2, pp. 399401.
Shevchenko, V.V. and Grinikh, L.I., Khimernost u rastenii (Chimericism in Plants), Moscow: Nauka, 1981.
Sakharov, V.V., Somatic Reduction as a Cause of Specific Mosaicism in Tetraploid Buckwheat, Dokl. Akad.
Nauk SSSR, 1946, vol. 52, no. 4, pp. 349352.
Zhatov, A.I., Migal, N.D., and Kovalenko, V.M.,
A Cytological Study of Polyploid Hemp, Tsitol. Genet.,
1969, vol. 3, no. 1, pp. 2835.

RUSSIAN JOURNAL OF GENETICS

Vol. 38

No. 5

2002