Planta (1988)175:254~258
P l a n t a 9 Springer-Verlag 1988
Chronology of the differentiation of cotton (Gossypium hirsutum L.) fiber cells D . A . Graves 1 * and J . M . Stewart z ** 1 Graduate Program in Life Science, Plant Physiology and Genetics Program, Department of Plant and Soil Science, University of Tennessee, Knoxville, TN 37901, and 2 Department of Agronomy, University of Arkansas, Fayetteville, AR 72701, USA
Abstract. Cotton fibers are single elongated cells
that develop from epidermal cells of the ovule. The chronology of fiber differentiation was investigated using cultured ovules. Epidermal cells differentiate into fiber cells approx. 3 d before anthesis. When ovules were cultured on a defined medium, fiber growth could be initiated on ovules any time between 2 d preanthesis and the time of anthesis by adding indole-3-acetic acid and gibberellic acid to the medium. In the absence of phytohormones, fibers did not grow, and when ovules between 2 d preanthesis and anthesis were cultured without hormones past the day of anthesis and hormones then added, most ovules failed to produce fibers. The results define the timing of fiber differentiation from epidermal cells, and also define a window of time when differentiated ceils are capable of further development. During this window, fiber cells are latent awaiting appropriate stimulation which in the intact plant is apparently associated with anthesis. Key words: Cell elongation - Fiber g r o w t h - Gos-
sypium (fiber growth) - Ovule culture.
Introduction
The development of cotton fibers can be divided into four distinct periods - initiation, elongation, secondary-cell-wall deposition, and maturation (Basra and Malik 1984; DeLanghe 1986). Initiation occurs at anthesis; at this time certain epider-
mal cells swell and bulge above the surface of the ovule's epidermis (Stewart 1986) although ultrastructural examination permits detection of prefiber cells as early as 16 h before anthesis (Ramsey and Berlin 1976). Fiber initiation on the in-situ ovule begins at the chalazal end and progresses toward the micropylar end, presumably because of diffusion of phytohormones along the epidermis (Stewart 1975). Initiation is completed within 2-3 h. Examination of the protein constituency of ovules from 6 d before anthesis to several days after anthesis using two-dimensional gel electrophoresis with silver staining (Graves and Stewart 1988) or protein blotting followed by immunodetection using an antiserum developed against fiber proteins (Graves 1986) indicated that epidermal cells undergo determination as fiber cells a few days before initiation occurs. This concept was tested using the in-vitro ovule-culture system developed by Beasley and Ting (1973) and modified by Stewart and Hsu (1977). Indole-3-acetic acid (IAA) and gibberellic acid (GA3) were used in the culture medium, since these two hormones are essential for fiber growth on unfertilized ovules in culture (Beasley and Ting 1974). In this report we investigate the differentiation of fiber cells from epidermal cells and the relationship between fiber differentiation and anthesis. Ovules varying in age from 6 d before anthesis to 1 d after anthesis were cultured under different temporal treatments with phytohormones. Material and methods
*Present address: Oak Ridge National Laboratory, P.O. Box X, Oak Ridge, TN 37831, USA ** To whom correspondence should be addressed Abbreviations. acid
GA3 = gibberellic acid; IAA = indole-3-acetic
Ovules were collected from greenhouse-grown cotton plants, Gossypium hirsutum L. cv. Deltapine 61 (Delta and Pineland Seed Co., Scott, Miss., USA). Plants were grown in a 1 : 1 (v/v) soil-sand mixture under natural light supplemented with light from 400-W halide lamps providing a 14-h day. Plants were
D.A. Graves and J.M. Stewart: Cotton fiber development
255 1 2
A
B
3 5 6 1 2 3
4 5 6
!
- 6'
'-5 ;
-3
-
- 2.
-i.
.0
.1
.2 .
3
.
4
.
5
.
6
.
7
.
8
9
10
1 2 3
C 4 5 6 7 1 2
D 3 4 5
1
2
E2
5 6 1 2
F 3 4 5
watered daily and fertilized once a week with a complete nutrient solution. Preanthesis ovules were collected according to their position on the plant (Mauney 1986). Six-day preanthesis ovules were taken from the first bud distal to a flower. Threeday preanthesis ovules were taken from the first bud on the first branch above a branch containing a flower at the first node. Two-, four- and five-day preanthesis ovules were collected from buds tagged at 3 and 6 d preanthesis. One-day preanthesis ovules were collected from buds with expanded but unopened corollas. Zero-day ovules were taken from flowers at anthesis and 1-d post-anthesis ovules were taken from I-d-old flowers. In order to reduce between-ovary variability, two ovules from each of five different ovaries for a total of 10 ovules were used in each culture flask containing 20 ml of medium at pH 5.3 (Stewart and Hsu 1977). Indoleacetic acid and GA3 to make ] ~tM and 10 g M (Beasley and Ting 1974), respectively, were added to, or omitted from, the medium, as indicated in Fig. 1. Ovules were aseptically removed from the ovaries and floated on the surface of the medium, then cultured in the dark at 32~ in a 5% CO2 atmosphere (Marden and Stewart ]984). Each experiment was replicated four times with four individual cultures in each replication. At the end of the culture period, the ovules were collected and fiber growth was determined by visual inspection and by a modification of the fiber-staining procedure of Beasley and Ting (1973). Ovules were stained for 5 rain in a solution of 0.2 M citrate-K phosphate buffer, pH 4.2, to which 0.2 g/1 toluidine blue 0, and ] ml/1 Triton X-100 (octylphenoxypolyethoxyethanol) was added. The ovules were then destained for 1 h in 50% ethanol-water (v/v) and the A~4o of the destaining solution was measured against fresh solution. The data represent the mean staining response and standard deviation of the four replicates. Most cultures contained a few ( < 2 0 % ) aberrant ovules that did not respond like the rest. These abnormal ovules
Fig. 1. Experimental design to test timing of cotton fiber differentiation. Capital letters indicate individual experiments while the accompanying numbers represent different treatments within an experiment. The time line between Expts. B and C indicates days relative to anthesis (0). The lines indicate the time in culture with the left end showing the age of the ovules when the experiment started and the right end showing when the experiment ended. Solid lines indicate culture in 1.0 M I A A and ]0 M GA3 ; broken lines indicate hormone-free culture. The change from broken lines to solid lines in Expts. D, E and F show when I A A and GA3 were added. All experiments were repeated four times. Ten representative ovules from each time point were individually stained, and the mean and SD were plotted in the subsequent figures
were considered non-representative of the fiber-growth responses t6 the treatments and were not used for fiber staining. The percentage of ovules which produced fibers in each culture was determined by examining all ovules for the presence of fibers using a dissecting microscope at 10 x magnification. The presence of any visible fiber was considered a positive response.
Results
Experiment A was designed to establish the standard fiber-growth response to the culture conditions. Ovules were collected on the day of anthesis, the day fibers begin expansion in-vivo, and cultured for 2, 4, 6, 8, and 10 d as shown in Fig. 1. The amount of fiber growth at each time point was related to stain absorption by the fibers. The A64o of the destain solution was plotted against time in culture to give the standard growth curve shown in Fig. 2. Nearly 100% of the ovules responded to culture by producing fibers. Experiment B tested the ability of ovules at anthesis, I d after anthesis, and 1, 2, 3 and 6 d preanthesis to produce fibers in the absence ofphytohormones. Figure 3 shows that the staining characteristics of these ovules deviated from the standard growth curve. None of the ovules cultured in hormone-free medium produced fibers. However, anthesis and preanthesis ovules enlarged and often produced some callus tissue.
256
D.A. Graves and J.M. Stewart: Cotton fiber development =,
1
I
I
1
i
I
5.0
I
i
i
1
i
i
C_
5.0-
A
1.0
1.0 0.5
~ 2
////ibm,/
A640
o.,:
0.5
l
i
,
i!!111
1oo
N w n
1;;
"
ilJ
A640 0
,o I i i 1 1 1
0.05
R~SPONDING 25
mm.lo
o
1
0
2
4
6
8
10
12
DAYS IN CULTURE WITH IAA AND GA 3 Fig. 2. Experiment A, standard growth curve for in-vitro cultured cotton fibers/ovules. Numbered points correspond to the numbers in Fig. 1, Expt. A. The standard curve was constructed from fiber-staining data using the least-squares method with point 1 excluded. The inset shows the percentage of ovules which produced fibers at each time point. The standard curve was superimposed on the subsequent figures
I
I
t
I
I
I
B
5.0 j
1.0 0.5
t
A640 0.1
0.05
0
,,
I
2
I
4
I
6
I
8
1
10
12
DAYS IN CULTURE (NO IAA AND GA3) Pig. 3. Experiment B, staining response of cotton ovules after hormone-free culture
0
I
2
I
4
I
6
2
3
4
I,
8
5
I
.
6
10
.
.
.
7
12
DAYS IN CULTURE Fig. 4. Experiment C, fiber production on preanthesis cotton ovules
Experiment C was conducted to determine at what age preanthesis could be induced to produce fibers. Ovules 6 d preanthesis to anthesis were cultured in basal medium plus IAA and GA3. The staining response and the percentage of ovules with fibers indicated that the epidermis had developed full capacity for fiber initiation by 2 d preanthesis (Fig. 4). All ovules were grown to the equivalent of 6 d postanthesis; therefore, the 2-d preanthesis ovules were in culture for 8 d, the 1-d preanthesis ovules for 7 d, and ovules collected at anthesis for 6 d. The fiber-staining response indicated that these ovules began to form fibers soon after they were placed in culture; i.e., the 2-d preanthesis ovules stained like 0-d ovules grown for 8 d, the 1-d preanthesis ovules stained like 7-d cultures, and the ovules taken at anthesis stained like 6-d cultures. For each of these ages (2 d preanthesis to anthesis) more than 75% of the ovules produced fibers. Ovules younger than 2 d preanthesis when placed in culture did not stain like ovules cultured for 9-12 d, and the percentage of ovules which produced fibers declined to 20% or less for ovules cultured at 3 d preanthesis and younger. These results demonstrated that the epidermis of the cotton ovule was capable of producing fibers by 2 d preanthesis and that the epidermal cells were developmentally poised to become fibers by this time. In vivo, fiber cells do not expand until anthesis (Stewart 1975). Delayed cell expansion until an-
D.A. Graves and J.M. Stewart: Cotton fiber development Table 1. Percentage of preanthesis cotton ovules producing fibers following hormone-free culture and subsequent hormone addition Expt.
Age of ovules when placed in culture
Days of hormone-free growth
Ovules producing fibers (%)
D
3 d preanthesis
0 2 3" 4 5
22 40 20 8 5
E
2 d preanthesis
0 1 2a 3 4 6
78 86 90 32 24 2
F
1 dpreanthesis
0 Ia 2 3 5
96 87 20 5 1
" Period equivalent to the in-situ day of anthesis for the ovules being cultured
thesis, in vivo, is probably related to lack of endogenous hormones since the delay was overcome in vitro by the addition of IAA and GA3 to the culture medium. This observation indicates that, in vivo, the fiber cells are held in a latent state until they receive a hormonal stimulus. This assumption was tested in experiment E. Two-day preanthesis ovules were cultured without phytohormones for 0, 1, 2, 3, 4, and 6 d, IAA and GAa were then added to the cultures, and the incubation continued until the ovules were the equivalent of 6 d postanthesis. When hormones were added after 0, 1, or 2 d of hormone-fi'ee culture the ovules grew fibers. When hormones were added at 3 d and later the ovules did not grow fibers, but instead, they produced callus. Callus tissues as well as fibers stained with toluidine blue, therefore, this measurement of fiber growth could not be used for this experiment. However, the percentage of ovules without fibers showed that after 2 d in hormonefree culture the ability of the ovules to grow fibers was substantially reduced (Table 1, Expt. E). The capacity of 2-d preanthesis ovules to produce fibers up to and through the in-vivo day of anthesis indicated that fiber initiation was coupled to anthesis in the intact plant. Alternatively, this response could be the consequence of degeneration of the ovules after 2 d in hormone-free culture. To resolve this issue, experiments F and D were conducted.
257
In experiment F, 1-d preanthesis ovules were cultured for 0, 1, 2, 3 and 5 d without hormones, then I A A and GA3 were added, and incubation was continued until the ovules reached the equivalent of 6 d postanthesis. If the 2-d preanthesis ovules in Expt. E were competent for 2 d but degenerated thereafter, then the 1-d preanthesis ovules should also remain competent for 2 d in hormone-free culture. Conversely, if the ability to grow fibers was developmentally linked to anthesis, then the 1-d preanthesis ovules should be able to produce fibers in response to hormones after I d in hormone-free culture but not after 2 d. The results in Table 1, Exp. F showed the latter was true. The percentage of ovules forming fibers in response to hormones declined after 1 d in hormone-free culture. Experiment D was conducted with 3-d preanthesis ovules. Although only 20% of the ovules produced fibers, this age was included to address the possible error in trying to define accurately responses that are only 1 d different. The same argument used for l-d preanthesis ovules was applied to 3-d preanthesis ovules; if fiber production is linked to anthesis, the ovules should be capable of producing fibers for 3 d, 2 d longer than the 1-d preanthesis ovules. The data in Table 1, Expt. D show that immediate addition of hormones or a 3-d delay before the addition of hormones resulted in the same percentage of fiberproducing ovules. Hormone-free culture for 4 d reduced the percentage of fiber-producing ovules to less than half of the 3-d value. Discussion The results of experiment C indicate that fibers can be initiated and grown on a high percentage of ovules as young as 2 d preanthesis. The lower percentage of 3-d preanthesis ovules that produced fibers indicates that, under the growth conditions of the parent plants, the differentiation event is completed between 2 and 3 d preanthesis. The observation that preanthesis ovules grown in the presence of IAA and GAa stained like ovules cultured on the day of anthesis and grown for an equivalent amount of time demonstrated that preanthesis fiber cells responded quickly to the phytohormone stimulus. This response is reminescent of the in-plant response of fiber cells on the day of anthesis (Stewart 1975). Based on Expts. D, E, and F, fiber initiation seems linked to the time of anthesis in the intact plant. This link is not broken by culture. The data support the concept that fiber cells are held in a
258
latent state until the required hormonal stimulus arrives from the mother plant. The ability to initiate fibers evidently begins with the differentiation event and ends 2-3 d later. In-vitro-grown ovules apparently maintain an internal clock that signals when anthesis would occur in the intact plant, and the manipulations in these experiments did not disrupt this timing. In conclusion, our results provide information concerning a developmental scheme in which differentiating epidermal cells are programmed to become a particular terminal cell type but are latent until anthesis. Continued development seems related to phytohormonal stimulation which apparently occurs at anthesis in the intact plant. References Basra, A.S., Malik, C.P. (1984) Development of cotton fibers. Int. Rev. Cytol. 98, 65-113 Beasley, C.A., Ting, I.P. (1973) The effects of plant growth substances on in vitro fiber development from fertilized cotton ovules. Am. J. Bot. 60, 130-139 Beasley, C.A., Ting, I.P. (1974) Effects of plant growth substances on in vitro fiber development from unfertilized cotton ovules. Am. J. Bot. 61, 188-194
D.A. Graves and J.M. Stewart: Cotton fiber development DeLanghe, E.A.L. (1986) Lint development. In: Cotton physiology, pp. 325349, Mauney, J.R., Stewart, J.M., eds. Cotton Foundation, Memphis, Tenn., USA Graves, D.A. (1986) Protein expression during early cotton fiber development. Ph.D. Dissertation, University Microfilms, Ann Arbor, Mich., USA Graves, D.A., Stewart, J.M. (1988) Analysis of the protein constituency of developing cotton fibers. J. Exp. Bot. 39, 59-69 Marden, L.L., Stewart, J.M. (1984) Influence of inhibitors and atmospheric COz on cultured cotton ovules. (Abstr.) Proc. Beltwide Cotton Prod. Res. Conf., p. 57. National Cotton Council, Memphis, Tenn., USA Mauney, J.R. (1986) Vegetative growth and development of fruiting sites. In: Cotton physiology, pp. 11-28, Mauney, J.R., Stewart, J.M., eds. Cotton Foundation, Memphis, Tenn., USA Ramsey, J.C., Berlin, J.D. (1976) Ultrastructure of early stages of cotton fiber differentiation. Bot. Gaz. 137, 11-19 Stewart, J.M. (1975) Fiber initiation on the cotton ovule (Gossypium hirsutum L.). Am. J. Bot. 62, 723-730 Stewart, J.M. (1986) Integrated events in the flower and fruit. In: Cotton physiology, pp. 261-300, Mauney, J.R., Stewart, J.M., eds. Cotton Foundation, Memphis, Tenn., USA Stewart, J.M., Hsu, C.H. (1977) In-ovulo embryo culture and seedling development of cotton (Gossypium hirsutum L.). Planta 137, 113-117
Received 17 December 1987; accepted 17 March 1988