Planta
Planta (1982)156:481M86
9 Springer-Verlag 1982
Changes in the non-structural carbohydrate content of cotton ( Gossypiumspp.) fibres at different stages of development Jean-Pierre Jaquet, Antony J. Buchala* and Hans Meier Institut de Biologie v6g6tale et de Phytochimie, Universit6 de Fribourg, CH-]700 Fribourg, Switzerland
Abstract. The neutral sugars (glucose, fructose,
and sucrose) and the sugar phosphates (glucose 6-phosphate, glucose 1-phosphate and fructose 6-phosphate) soluble in hot aqueous 80% methanol from the fibres of cotton - Gossypium arboreum L., G. barbadense L., and G. hirsutum L. - were determined at various stages of fibre development. In addition, the (1 ~3)-fl-D-glucan content was measured and in the case of G. arboreum the rate of (1 ~ 3)-fl-D-glucan and cellulose synthesis was determined with [14C]sucrose as the precursor. For each of the species a similar chronology was obtained for the changes in content of the various non-structural carbohydrates. At the early stages of secondary wall formation, glucose and fructose exhibited a maximum which was closely followed by a maximum in the (1--* 3)-fl-D-glucan content and in the sugar phosphates. On the other hand, the sucrose content increased regularly until fibre maturity. The rates of synthesis of (1 ~3)-fi-Dglucan and of cellulose were highest following the maximum in the (1 - , 3)-fl-D-glucan content, when the latter was being depleted.
Huwyler et al. (1979) deduced that a glucose-containing non-cellulosic polysaccharide was metabolized during the formation of the secondary cell wall. Later studies (Meier et al. 1981) showed that a (1--, 3)-fl-D-glucan underwent turnover during massive secondary cell wall synthesis. The (1 ---,3)-fl-D-glucan content shows a distinct maxim u m at the beginning of secondary wall formation. Conner et al. (1972) determined the levels of low molecular weight sugars in cotton fruits during development under various climatic conditions. They suggested that such sugars could be the precursors for cellulose synthesis, since the observed maxima occurred when the fibres had just attained their maximum length and when secondary thickening had begun. These values, however, are difficult to interpret in terms of the fibres since carpels and seeds were also included in the analyses. The present work was undertaken in an attempt to clarify the relationship between the low molecular weight sugars, the (1--*3)-fl-D-glucan content, and the synthesis of the latter and cellulose.
Key words: Callose - Carbohydrates (non-structural) - Cellulose - Cotton fibre - Gossypium.
Materials and methods
Introduction
Studies have been carried out on the variations in the structural polysaccharides of cotton fibres during their development (Huwyler etal. 1979; Meinert and Delmer 1977). From their results, * To whom correspondence should be addressed Abbreviations: DMSO = dimethyl-sulphoxide; DPA = days post anthesis; UDP-glucose = uridinediphosphoglucose
Plant material, Cotton plants (Gossypium arboreum L. and G. barbadense L.) were grown in a greenhouse at a temperature of 30 ~ C during the day and 20 ~ C at night. G. hirsutum L. (var. Stoneville No. 406) plants were grown in a growth chamber (K. Weiss, Giessen, F R G ) under day conditions (12.30 h) of 26 ~ C and 65% relative humidity and night conditions (11.30 h) of 20 ~ C and 95% relative humidity. For G. arboreum and G. barbadense the fruits at various stages of development were all harvested at the same time in order to eliminate any possible changes due to differences in weather conditions. The fruits from G. hirsutum plants were harvested when required, but always at the same time of day. Once harvested, the fruits were kept on ice and opened as soon as possible. The individual seed clusters were removed from their locules and studied intact for incorporation experiments.
0032-0935/82/0156/0481/$01.20
482 For extraction of the low molecular weight sugars, the seed clusters were unraveled and the fibres separated from the seeds by forceps on a petri dish. The seeds were counted and discarded, while the fibres and any cytoplasmic or vacuolar fluid were quantitatively transferred to a test tube containing hot 80% (v/v) methanol which was immediately brought to boiling to inactivate enzyme activity.
Extraction of low molecular weight sugars. The deactivated fibres were coarsely cut with scissors and exhaustively extracted with hot 80% aqueous methanol (5.15 ml) with intermediate filtration. All of the methanolic solutions were pooled, filtered (Whatman GF/A) and the final volume adjusted to 100 ml. The extracts were stored at 4 ~ until required for analysis. The fibres were quantitatively recovered and the dry weight determined after drying at 60 ~ C and 0.13 mbar for 24 h.
J.P. Jaquet et al. : Non-structural carbohydrates in cotton fibres or hydrolyzed in 2 M trifluoroacetic acid for 1 h at 120 ~ C in an autoclave and evaporated to dryness. The glucose content of the hydrolysates was estimated as described above. Both hydrolysis methods gave essentially similar results.
Sampling and reproductibility of the analyses. For each stage of development three similar fruits were chosen. The fibres from the seed clusters of each fruit were combined, except for those from G. arboreum, where one seed cluster from each fruit was used for the incorporation experiments. All the results are expressed on the basis of the fibres from one seed. Each analysis was carried out in triplicate for each of the three fruits at a particular stage of development, and the values given are the means after the elimination of erratic results. In general, good reproducibility was attained and the results can be considered to be representative.
Feeding of radioactive precursors. This was carried out as described earlier (Pillonel et al. 1980), except that the incubation medium was continuously aerated using an aquarium pump.
Results
Determination of the radioactivity in the soluble and insoluble high molecular weight products. The methods involving extrac-
Qualitative analysis of the low molecular weight sugars soluble in 80% methanol. For Gossypium arboreum, G. barbadense, and G. hirsutum, at all stages of fibre maturity, the sugars which could be solubilized in hot 80% aqueous methanol were qualitatively identical. In each case large amounts of glucose, fructose, and sucrose were observed, along with much smaller amounts of raffinose, stachyose, melibiose, galactinol, and myo-inositol. These sugars were identified by paper and thin-layer chromatography and by gas-liquid chromatography of their trimethylsilyl or trimethylsilyl/oxime derivatives. For the latter method some extracts were chromatographed on a column of Sephadex G-15, and the fractions enriched in the minor components were examined. The identities of the glucose and sucrose were further confirmed by testing them as substrates for glucose oxidase and invertase, respectively. Fractions corresponding to raffinose and myo-inositol were purified to chromatographic homogeneity by preparative paper chromatography. The former was treated with an 0~-galactosidase from green coffee beans to release galactose, and sucrose and the hydrolysate was further treated with yeast invertase. The galactose, glucose and fructose in the final hydrolysate were determined enzymatically with galactose dehydrogenase and the hexokinase methods, respectively, as described by Bergmeyer (1974); galactose, glucose, and fructose were found to be present in the ratio /.1:0.97:1.0. Since the oligosaccharide was resolved from umbelliferose (solvent A) its identity to raffinose was confirmed. The fraction corresponding to myo-inositol was treated with myo-inositol dehydrogenase but very little N A D was reduced. Further chromatography showed that the material was contaminated with an oligosaccharide chromatographically similar to meli-
tion with hot dimethyl sulphoxide (DMSO) or by acid hydrolysis for total incorporation, described by Pillonel et al. (1980) were used.
Chromatographic methods. Paper chromatography was carried out on Schleicher and Schfill No 2043b paper, preparative paper chromatography on Whatman 3MM paper, and thin-layer chromatography on Kieselgel 60 (Merck, Darmstadt, FRG) using the following solvents: (A) propanol-1 :ethyl acetate: water (7:1 : 2 v/v); (B) ethyl acetate: pyridine: water (8: 2:1 v/v) or (C) acetone:water (88:12 v/v). Detection was either with alkaline AgNO 3 (Trevelyan et al. 1950) or with 7-naphthol/ conc. H2SO 4 (Stahl 1969) where appropriate. Gas-liquid chromatography of the neutral sugars as their pertrimethylsilyl derivatives (non-reducing sugars) or as their pertrimethylsilyl oxime derivatives (reducing sugars and fructose) was performed on a Perkin-Elmer F-30 chromatograph with FID, using a column (2 m.2 mm i.d.) containing 2.5% OV-17 on AW DMCS Chromasorb G (10(~120 mesh) as described by Zfircher et al. (1975). Benzyl-fl-D-arabinopyranoside was used as the internal standard and correction factors were established for the sugars to be analyzed. Peak areas were measured using an Autolab Minigrator (Spectra Physics, Santa Clara, USA). Enzymic analyses. Sugar phosphates were determined by the methods described by Bergmeyer (1974), using glucose 6-phosphate dehydrogenase, phosphoglucoisomerase, and phosphoglucomutase supplied by Boehringer, Mannheim, FRG. Glucose was determined with glucose oxidase (GOD-Perid method, Boehringer, FGR) and galactose with galactose dehydrogenase (Bergmeyer 1974). Other analyses. The total sugar content of methanol extracts was also determined using the phenol/conc. H z S O 4 method (Dubois et al. 1956) with reference to a standard curve prepared for a mixture of glucose, fructose, and sucrose in the proportions usually found in the extracts. The (1 ~ 3)-fl-D-glucan content in the DMSO extracts (see above) was determined in an aliquot which, after dilution with water, was exhaustively dialyzed against distilled water and freeze-dried. The product was either suspended in 0.01 M citrate/phosphate buffer (pH 5.0) and treated with 50 p,g exo-fl1,3-glucanase (EC 3.2.1.58) from Basidiomycete sp. QM 806
J.P. Jaquet et al. : Non-structural carbohydrates in cotton fibres
483
biose, and total acid hydrolysis of the material gave inter alia glucose, galactose, and starting material. Quantitative analysis of the low molecular weight sugars soluble in 80% methanol. Preliminary studies confirmed the observations of Demol (1964), who showed that the weight of the cotton fibres increased with the height of the fruit on the plant and that the climatic conditions affected the translocation of sugars in the cotton plant. In the comparative study described in this communication, particular care was taken in the selection of the fruits which were all harvested at approximately the same height on the plant. The fruits of G. hirsuturn, from the growth chamber, were cultivated under controlled climatic conditions, while those of G. arboreum and G. barbadense were subject to some variations during their development, especially concerning illumination. The regularity of the changes in the sugar contents observed can be taken to indicate the success in the sampling, but it should be noted that occasional erratic results were eliminated. Treatment of the fibres with hot aqueous 80% (v/v) methanol solubilizes very little high molecular weight material, so that the dry weight obtained after extraction can be taken to indicate the extent of deposition of the cell walls. Normally, the phases corresponding to the periods of primary and secondary cell wall formation can be distinguished without difficulty. Curves showing the fibre weight at various stages of development are given in Figs. 1(c), 2(c), and 3(c) for G. arboreum, G. barbadense and G. hirsutum, respectively. It can be seen that secondary wall formation began relatively early (17-27 days post anthesis (DPA)) for G. hirsutum and somewhat later for G. arboreum (24-27 DPA) and G. barbadense (ca. 32 DPA). Only fructose, glucose, and sucrose were present in sufficient quantity in the methanol-soluble fractions for acceptable quantitative determination by gas-liquid chromatography (GLC). The histograms in Figs. l(a), 2(a), and 3(a) show the variations in the quantities of these neutral sugars. For each species a maximum was observed for the fructose and glucose contents, corresponding to the transition phase from primary to secondary wall formation or to the beginning of massive secondary cell wall formation. The sucrose content, on the other hand, gradually increased until fibre maturity and for G. arboreum and G. hirsutum then diminished. For most of the analyses the fructose content was slightly greater than that of the glucose, and that of the sucrose considerably smaller. Fructose 6-phosphate, glucose 6-phosphate,
i t
~
glucose
[]
fructose
9
sucrose
5
~
3
.~
2 0
k'~ fructose 6-phosphate I~glucose 6-phosphate
~
8
cell wall C----O 3--D
dry weight
(1 - 3 ) - 15- D -
glucan content
incorporation into ( 1 - 3 ) - , a n d (1 - 4 ) - B - D - glucan 60
50
1.0
40
90 . 8
30
98 , 6
20
20
.o.4
~'
10
10
' 0.2
~
0
0
P'
u
0
15
20
24
27
30
33
37
40
48
Days post anthesis
Fig. l a - e . Gossypium arboreum - a quantitative changes in the neutral sugars soluble in 80% methanol; b quantitative changes in the sugar phosphates soluble in 80% methanol; e changes in the dry weight of the cotton fibres and of their callose content, after treatment with 80% methanol, during development, and variation of the rate of incorporation of [~4C] sucrose into /~-glucan (cellulose+callose) by intact fibres of isolated seed clusters
and much smaller amounts of glucose l-phosphate were also detected in the methanol-soluble fraction. However, the amounts of the glucose 1-phosphate were normally insufficient for significant routine estimation. Maxima were observed for the three species [Figs. l(b), 2(b), and 3(b)] which occurred several days after the maxima observed for the monosaccharides. For G. arboreum and G. barbadense the delay observed was small - 7 and 3 d, respectively - while that for G. hirsutum, where the maximum itself was less well defined, was 15 d. For the three species there was usually two to three times more glucose 6-phosphate than fructose 6-phosphate. Determination of the (l~3)-~-D-glucan. It was previously shown (Pillonel et al. 1980) that treatment of cotton fibres with DMSO in the autoclave solubilizes essentially all of the (1 ~ 3)-/~-D-glucan.
484
J.P. Jaquet et al. : Non-structural carbohydrates in cotton fibres J~ glucose
8 6I 5 ~ a)1~1" ..... I
0~ "6
4
k~
3 2
9~
~
fit. ~
fructose 6-phosphate
24 t6
.~
12
[]
glucose 6-phosphate
1
-~
ii
cell wall dry weight (1- 3) - B- D- glucan
[
o
36 32 28
[]
fructose 6-phosphate
E
=- -= cell wall dry weight (1-3)-=3 D - glucan
E
Z
0.8
40
O.6
30;
0.4
20
0.2
10
0
0
9
17
27
32
37
42
48
r
Days post anthesis
Fig. 2a-e. Gossypium barbadense - a quantitative changes in the neutral sugars soluble in 80% methanol; b quantitative changes in the sugar phosphates soluble in 80% methanol; e changes in the dry weight of the cotton fibres and of their callose content, after treatment with 80% methanol, during development
glucose 6-phosphate
j__j
24 20 16
8
60
~
(b)
content
(c)
ca
sucrose
o
28
"6
9
d 9 . .~1 .....
o 3
[] fructose
content
(c) E
6O
1.0
50 m
O.8
40 ,im
0.6
30
0.4
20
0.2
10
0 21
E
26
32
36 39
45
56
~"
0
Days post anthesis
The material solubilized at the various stages o f d e v e l o p m e n t was h y d r o l y z e d and the glucose produced was determined. The values obtained are shown in Figs. l(c), 2(c), and 3(c) for G. arboreum, G. barbadense, and G. hirsutum, respectively. M a x i m a were obtained which occurred at the same time for that o f the monosaccharides (G. arboreurn), or which occurred slightly later (G. barbadense and G. hirsutum). The absolute a m o u n t s o f callose are low in the p r i m a r y cell walls and increase sharply at the transition period, marking the beginning o f secondary cell wall formation. After the m a x i m u m , the ( l ~ 3 ) - f l - D - g l u c a n content decreases slowly by a b o u t 50% until fibre maturity.
Rate of fl-gIucan synthesis. Pillonel et al. (1980) showed that the rate o f fl-glucan synthesis [cellulose and (1 --. 3)-fl-D-glucan], using sucrose as precursor, varied with fibre d e v e l o p m e n t and was at
Fig, 3a-e. Gossypium hirsutum - a quantitative changes in the neutral sugars soluble in 80% methanol; b quantitative changes in the sugar phosphates soluble in 80% methanol; e changes in the dry weight of the cotton fibres and of their callose content, after treatment with 80% methanol, during development a m a x i m u m at the middle o f massive secondary wall formation, but that the age at which the m a x i m a occurred depended to some extent on external climatic factors. The m a x i m u m appeared to occur later than the previously r e p o r t e d peak for the (1 ~ 3)-fl-D-glucan (Huwyler et al. 1979), but n o data were available for b o t h analysis f r o m the same series o f fruits. Such analyses are n o w rep o r t e d for G. arboreum using radioactive sucrose as the precursor. The rate o f synthesis o f the (1 ~ 3)-fl-D-glucan was obtained by measuring the radioactivity in the D M S O - s o l u b l e material (see above) and the rate o f cellulose synthesis by measuring the radioactivity in the residue after D M S O
J.P. Jaquet et al. : Non-structural carbohydrates in cotton fibres
485
treatment. As previously reported (Pillonel et al. 1980), incorporation into (1 ~ 3)-fl-D-glucan paralleled that into cellulose and at all fibre ages incorporation into the former was about twice that of the latter. Fig, l(c) shows the total incorporation (the sum of the two values obtained) into fl-glucan with respect to fibre development. A maximum was observed at 33 DPA, i.e., three days later than the maximum in the (1 ---,3)-fl-D-glucan content.
served for the monosaccharides fructose and glucose corresponding to the beginning of secondary wall formation. The amount of sucrose continued to build up slowly and was at a maximum at fibre maturity. A second maximum was observed for the (l~3)-fl-D-glucan content; this occurred closely after the maximum for the monosaccharides and just before the massive synthesis of the secondary wall. For G. arboreum, it was also established that the maximum rate of synthesis of /%glucan (cellulose and (1 ~3)-fl-D-glucan) occurred during this period of massive synthesis of the cell wall. The maxima observed for the sugar phosphates were in general less well defined, corresponding more or less in the case of G. arboreum and G. barbadense to the maximum for (1 -->3)-fl-D-glucan, while in the case of G. hirsutum a spread-out peak was observed, occurring later than the corresponding maximum in the (1 ~3)-fl-D-glucan. However, it should be noted that for G. hirsuturn the peak for the (1 ~ 3)-fl-Dglucan content is also less well defined than that for each of the other two species. These results indicate that during primary cell wall formation, sucrose, which is the transport sugar, is inverted and that glucose and fructose are stored. These reserves are used later for the formation of the secondary cell wall. Since there is a maximum in the (1 ~ 3)-fl-D-glucan content which might also have a reserve function (Meier 1981), following the maxima for the monosaccharides but before the period of massive secondary wall formation, it is possible that these monosaccharides are used in part for the synthesis of (1 ~3)-fl-Dglucan. This polysaccharide clearly undergoes a turnover during the formation of the secondary cell wall (Meier et al. 1981, Meier 1981), but it is not clear whether it is an obligatory or a facultarive intermediate in cellulose synthesis. The coincidence in the maxima for the sugar phosphates and the maximum for the rate of fl-glucan synthesis indicates that the former may be precursors for UDP-glucose which is known to be an excellent precursor for (1->3)-fl-D-glucan (Pillonel et al. 1980; Delmer et al. 1977) in cotton fibres. However, when UDP-glucose is fed to seed clusters or detached fibres the main product is sucrose. An interpretation of the variations of the sucrose level in the fibres during their development is difficult, since this is the main sugar transported via the phloem to the ovules and its concentration in the fibres depends on changes in its influx and efflux, which in turn depend on other factors external to the fruit. A knowledge of the concentrations of the various sugars in the different metabolic pools
Discussion
During the incorporation experiments in vitro, of the type previously described for cotton seed clusters (Pillonel et al. 1980) when sucrose was present in the incubation medium, there was a subsequent accumulation of glucose and fructose in the incubation medium, indicating the presence of an invertase in the fibres (unpublished observation). Buchala and Meier (unpublished data) have demonstrated the presence of cell surface located invertase for a Gossypium arboreum fibres, but the role of the enzyme is not clear. A possible function of the enzyme could involve regulation of cell wall synthesis by mediation of uptake of sucrose from the apoplast into the symplast. In order to understand the possible role of the enzyme, its effect on the amount of low molecular weight sugars and relationship to cell wall synthesis, a study of the low molecular weight sugars present in cotton fibers was undertaken. At the same time a comparative study of three species of cotton was carried out. Carpita and Delmer (1981) reported analyses for low molecular weight sugars in fibres from cotton (G, hirsutum) ovules cultured in vitro, but the study was directed toward the concentration and turnover of UDP-glucose, and no systematic study of the variations in the amounts of other sugars was reported. The sugar content of whole cotton (G. hirsutum) fruits was found to be influenced by the climatic conditions prevailing (Conner et al. 1972). Definite age-temperature relationships were established and they were associated with the cessation of fibre elongation and the initiation of secondary wall formation. These results are, however, difficult to interpret in terms of the fibres alone, although they may be quite pertinent to studies carried out with whole fruits, e.g., those described by Meier et al. (1981). The results obtained for the three different species of cotton show similar trends, although there were quite marked differences in the absolute amounts of a particular component from species to species. In each case, a first maximum was ob-
486
J.P. Jaquet et al. : Non-structural carbohydrates in cotton fibres
and of the enzymes which are active in the apoplast and the symplast is essential for an understanding of the synthesis of the cell wall fl-glucans. The results given are global for the apoplast plus the symplast, since even their separate analysis poses many technical problems. Work is at present being carried out to elucidate the metabolic pathway of sugars fed separately to the apoplast and symplast, respectively, of intact and detached cotton fibres.
Dubois, M., Gilles, K.A., Hamilton, J.K., Rebers, P.A., Smith, F. (1956) Colorimetric method for determination of sugars and related substances. Anal. Chem. 28, 350-356 Huwyler, H.R., Franz, G., Meier, H. (1979) Changes in the composition of cotton fibre cell walls during development. Planta 146, 635-642 Meier, H., Buchs, L., Buchala, A.J., Homewood, T. (1981) (I--* 3)-fl-D-Glucan (callose) is a possible intermediate in biosynthesis of cellulose of cotton fibres. Nature (London) 289, 821-822 Meier, H. (1981) Is callose a precursor of cellulose in cotton fibres? In: Cell walls '81, pp. 75-83, Robinson, D.G., Quader, H., eds. Wissenschaftl. Verlagsges, Stuttgart Meinert, M.C., Delmer, D.P. (1977) Changes in biochemical composition of the cell wall of the cotton fiber during development. Plant Physiol. 59, 1088-1097 Pillonel, Ch., Buchala, A.J., Meier, H. (1980) Glucan synthesis by intact cotton fibres fed with different precursors at the stages of primary and secondary wall formation. Planta 149, 306-312 Stahl, E. (1969) Thin-layer chromatography. Springer, New York Heidelberg Berlin Trevelyan, W.E., Procter, D.P., Harrison, J.S. (1950) Detection of sugars on paper chromatograms. Nature (London) 166, 444445 Zfircher, K., Hadorn, H., Strack, C. (1975) Gaschromatographische Zuckerbestimmung. Mitt. Lebensmittelchem. Hyg. 66, 92-116
The authors wish to thank Mrs. Th6r6se Homewood for help with the incorporation experiments and Dr. H. Hopf for a sample of umbelliferose. This project was supported by the Swiss National Science Foundation.
References Bergmeyer, H.U. (1974) Methoden der enzymatischen Analyse, pp. 1224-1287. Verlag Chemic, Weinheim Carpita, N.C., Delmer, D.P. (1981) Concentration and metabolic turnover of UDP-glucose in developing cotton fibers. J. Biol. Chem. 256, 308-315 Conner, J.W., Kreig, D.R., Gipson, J.R. (1972) Accumulation of simple sugars in developing cotton bolls as influenced by night temperatures. Crop Sci. 12, 752-754 Demol, J. (1964) Contribution fi l'6tude de l'influence du r6gime hydrique sur diff6rents facteurs de productivit6 chez Gossypium hirsutum L.. Bull. Inf. INEAC 13, 41-75 Delmer, D.P., Heiniger, U., Kulow, C. (1977) UDP-glucose: glucan synthetase in developing cotton fibers. Plant Physiol. 59, 713-718
Received 6 August; accepted 9 September 1982
Planta (1982)156:481M86
9 Springer-Verlag 1982
Changes in the non-structural carbohydrate content of cotton ( Gossypiumspp.) fibres at different stages of development Jean-Pierre Jaquet, Antony J. Buchala* and Hans Meier Institut de Biologie v6g6tale et de Phytochimie, Universit6 de Fribourg, CH-]700 Fribourg, Switzerland
Abstract. The neutral sugars (glucose, fructose,
and sucrose) and the sugar phosphates (glucose 6-phosphate, glucose 1-phosphate and fructose 6-phosphate) soluble in hot aqueous 80% methanol from the fibres of cotton - Gossypium arboreum L., G. barbadense L., and G. hirsutum L. - were determined at various stages of fibre development. In addition, the (1 ~3)-fl-D-glucan content was measured and in the case of G. arboreum the rate of (1 ~ 3)-fl-D-glucan and cellulose synthesis was determined with [14C]sucrose as the precursor. For each of the species a similar chronology was obtained for the changes in content of the various non-structural carbohydrates. At the early stages of secondary wall formation, glucose and fructose exhibited a maximum which was closely followed by a maximum in the (1--* 3)-fl-D-glucan content and in the sugar phosphates. On the other hand, the sucrose content increased regularly until fibre maturity. The rates of synthesis of (1 ~3)-fi-Dglucan and of cellulose were highest following the maximum in the (1 - , 3)-fl-D-glucan content, when the latter was being depleted.
Huwyler et al. (1979) deduced that a glucose-containing non-cellulosic polysaccharide was metabolized during the formation of the secondary cell wall. Later studies (Meier et al. 1981) showed that a (1--, 3)-fl-D-glucan underwent turnover during massive secondary cell wall synthesis. The (1 ---,3)-fl-D-glucan content shows a distinct maxim u m at the beginning of secondary wall formation. Conner et al. (1972) determined the levels of low molecular weight sugars in cotton fruits during development under various climatic conditions. They suggested that such sugars could be the precursors for cellulose synthesis, since the observed maxima occurred when the fibres had just attained their maximum length and when secondary thickening had begun. These values, however, are difficult to interpret in terms of the fibres since carpels and seeds were also included in the analyses. The present work was undertaken in an attempt to clarify the relationship between the low molecular weight sugars, the (1--*3)-fl-D-glucan content, and the synthesis of the latter and cellulose.
Key words: Callose - Carbohydrates (non-structural) - Cellulose - Cotton fibre - Gossypium.
Materials and methods
Introduction
Studies have been carried out on the variations in the structural polysaccharides of cotton fibres during their development (Huwyler etal. 1979; Meinert and Delmer 1977). From their results, * To whom correspondence should be addressed Abbreviations: DMSO = dimethyl-sulphoxide; DPA = days post anthesis; UDP-glucose = uridinediphosphoglucose
Plant material, Cotton plants (Gossypium arboreum L. and G. barbadense L.) were grown in a greenhouse at a temperature of 30 ~ C during the day and 20 ~ C at night. G. hirsutum L. (var. Stoneville No. 406) plants were grown in a growth chamber (K. Weiss, Giessen, F R G ) under day conditions (12.30 h) of 26 ~ C and 65% relative humidity and night conditions (11.30 h) of 20 ~ C and 95% relative humidity. For G. arboreum and G. barbadense the fruits at various stages of development were all harvested at the same time in order to eliminate any possible changes due to differences in weather conditions. The fruits from G. hirsutum plants were harvested when required, but always at the same time of day. Once harvested, the fruits were kept on ice and opened as soon as possible. The individual seed clusters were removed from their locules and studied intact for incorporation experiments.
0032-0935/82/0156/0481/$01.20
482 For extraction of the low molecular weight sugars, the seed clusters were unraveled and the fibres separated from the seeds by forceps on a petri dish. The seeds were counted and discarded, while the fibres and any cytoplasmic or vacuolar fluid were quantitatively transferred to a test tube containing hot 80% (v/v) methanol which was immediately brought to boiling to inactivate enzyme activity.
Extraction of low molecular weight sugars. The deactivated fibres were coarsely cut with scissors and exhaustively extracted with hot 80% aqueous methanol (5.15 ml) with intermediate filtration. All of the methanolic solutions were pooled, filtered (Whatman GF/A) and the final volume adjusted to 100 ml. The extracts were stored at 4 ~ until required for analysis. The fibres were quantitatively recovered and the dry weight determined after drying at 60 ~ C and 0.13 mbar for 24 h.
J.P. Jaquet et al. : Non-structural carbohydrates in cotton fibres or hydrolyzed in 2 M trifluoroacetic acid for 1 h at 120 ~ C in an autoclave and evaporated to dryness. The glucose content of the hydrolysates was estimated as described above. Both hydrolysis methods gave essentially similar results.
Sampling and reproductibility of the analyses. For each stage of development three similar fruits were chosen. The fibres from the seed clusters of each fruit were combined, except for those from G. arboreum, where one seed cluster from each fruit was used for the incorporation experiments. All the results are expressed on the basis of the fibres from one seed. Each analysis was carried out in triplicate for each of the three fruits at a particular stage of development, and the values given are the means after the elimination of erratic results. In general, good reproducibility was attained and the results can be considered to be representative.
Feeding of radioactive precursors. This was carried out as described earlier (Pillonel et al. 1980), except that the incubation medium was continuously aerated using an aquarium pump.
Results
Determination of the radioactivity in the soluble and insoluble high molecular weight products. The methods involving extrac-
Qualitative analysis of the low molecular weight sugars soluble in 80% methanol. For Gossypium arboreum, G. barbadense, and G. hirsutum, at all stages of fibre maturity, the sugars which could be solubilized in hot 80% aqueous methanol were qualitatively identical. In each case large amounts of glucose, fructose, and sucrose were observed, along with much smaller amounts of raffinose, stachyose, melibiose, galactinol, and myo-inositol. These sugars were identified by paper and thin-layer chromatography and by gas-liquid chromatography of their trimethylsilyl or trimethylsilyl/oxime derivatives. For the latter method some extracts were chromatographed on a column of Sephadex G-15, and the fractions enriched in the minor components were examined. The identities of the glucose and sucrose were further confirmed by testing them as substrates for glucose oxidase and invertase, respectively. Fractions corresponding to raffinose and myo-inositol were purified to chromatographic homogeneity by preparative paper chromatography. The former was treated with an 0~-galactosidase from green coffee beans to release galactose, and sucrose and the hydrolysate was further treated with yeast invertase. The galactose, glucose and fructose in the final hydrolysate were determined enzymatically with galactose dehydrogenase and the hexokinase methods, respectively, as described by Bergmeyer (1974); galactose, glucose, and fructose were found to be present in the ratio /.1:0.97:1.0. Since the oligosaccharide was resolved from umbelliferose (solvent A) its identity to raffinose was confirmed. The fraction corresponding to myo-inositol was treated with myo-inositol dehydrogenase but very little N A D was reduced. Further chromatography showed that the material was contaminated with an oligosaccharide chromatographically similar to meli-
tion with hot dimethyl sulphoxide (DMSO) or by acid hydrolysis for total incorporation, described by Pillonel et al. (1980) were used.
Chromatographic methods. Paper chromatography was carried out on Schleicher and Schfill No 2043b paper, preparative paper chromatography on Whatman 3MM paper, and thin-layer chromatography on Kieselgel 60 (Merck, Darmstadt, FRG) using the following solvents: (A) propanol-1 :ethyl acetate: water (7:1 : 2 v/v); (B) ethyl acetate: pyridine: water (8: 2:1 v/v) or (C) acetone:water (88:12 v/v). Detection was either with alkaline AgNO 3 (Trevelyan et al. 1950) or with 7-naphthol/ conc. H2SO 4 (Stahl 1969) where appropriate. Gas-liquid chromatography of the neutral sugars as their pertrimethylsilyl derivatives (non-reducing sugars) or as their pertrimethylsilyl oxime derivatives (reducing sugars and fructose) was performed on a Perkin-Elmer F-30 chromatograph with FID, using a column (2 m.2 mm i.d.) containing 2.5% OV-17 on AW DMCS Chromasorb G (10(~120 mesh) as described by Zfircher et al. (1975). Benzyl-fl-D-arabinopyranoside was used as the internal standard and correction factors were established for the sugars to be analyzed. Peak areas were measured using an Autolab Minigrator (Spectra Physics, Santa Clara, USA). Enzymic analyses. Sugar phosphates were determined by the methods described by Bergmeyer (1974), using glucose 6-phosphate dehydrogenase, phosphoglucoisomerase, and phosphoglucomutase supplied by Boehringer, Mannheim, FRG. Glucose was determined with glucose oxidase (GOD-Perid method, Boehringer, FGR) and galactose with galactose dehydrogenase (Bergmeyer 1974). Other analyses. The total sugar content of methanol extracts was also determined using the phenol/conc. H z S O 4 method (Dubois et al. 1956) with reference to a standard curve prepared for a mixture of glucose, fructose, and sucrose in the proportions usually found in the extracts. The (1 ~ 3)-fl-D-glucan content in the DMSO extracts (see above) was determined in an aliquot which, after dilution with water, was exhaustively dialyzed against distilled water and freeze-dried. The product was either suspended in 0.01 M citrate/phosphate buffer (pH 5.0) and treated with 50 p,g exo-fl1,3-glucanase (EC 3.2.1.58) from Basidiomycete sp. QM 806
J.P. Jaquet et al. : Non-structural carbohydrates in cotton fibres
483
biose, and total acid hydrolysis of the material gave inter alia glucose, galactose, and starting material. Quantitative analysis of the low molecular weight sugars soluble in 80% methanol. Preliminary studies confirmed the observations of Demol (1964), who showed that the weight of the cotton fibres increased with the height of the fruit on the plant and that the climatic conditions affected the translocation of sugars in the cotton plant. In the comparative study described in this communication, particular care was taken in the selection of the fruits which were all harvested at approximately the same height on the plant. The fruits of G. hirsuturn, from the growth chamber, were cultivated under controlled climatic conditions, while those of G. arboreum and G. barbadense were subject to some variations during their development, especially concerning illumination. The regularity of the changes in the sugar contents observed can be taken to indicate the success in the sampling, but it should be noted that occasional erratic results were eliminated. Treatment of the fibres with hot aqueous 80% (v/v) methanol solubilizes very little high molecular weight material, so that the dry weight obtained after extraction can be taken to indicate the extent of deposition of the cell walls. Normally, the phases corresponding to the periods of primary and secondary cell wall formation can be distinguished without difficulty. Curves showing the fibre weight at various stages of development are given in Figs. 1(c), 2(c), and 3(c) for G. arboreum, G. barbadense and G. hirsutum, respectively. It can be seen that secondary wall formation began relatively early (17-27 days post anthesis (DPA)) for G. hirsutum and somewhat later for G. arboreum (24-27 DPA) and G. barbadense (ca. 32 DPA). Only fructose, glucose, and sucrose were present in sufficient quantity in the methanol-soluble fractions for acceptable quantitative determination by gas-liquid chromatography (GLC). The histograms in Figs. l(a), 2(a), and 3(a) show the variations in the quantities of these neutral sugars. For each species a maximum was observed for the fructose and glucose contents, corresponding to the transition phase from primary to secondary wall formation or to the beginning of massive secondary cell wall formation. The sucrose content, on the other hand, gradually increased until fibre maturity and for G. arboreum and G. hirsutum then diminished. For most of the analyses the fructose content was slightly greater than that of the glucose, and that of the sucrose considerably smaller. Fructose 6-phosphate, glucose 6-phosphate,
i t
~
glucose
[]
fructose
9
sucrose
5
~
3
.~
2 0
k'~ fructose 6-phosphate I~glucose 6-phosphate
~
8
cell wall C----O 3--D
dry weight
(1 - 3 ) - 15- D -
glucan content
incorporation into ( 1 - 3 ) - , a n d (1 - 4 ) - B - D - glucan 60
50
1.0
40
90 . 8
30
98 , 6
20
20
.o.4
~'
10
10
' 0.2
~
0
0
P'
u
0
15
20
24
27
30
33
37
40
48
Days post anthesis
Fig. l a - e . Gossypium arboreum - a quantitative changes in the neutral sugars soluble in 80% methanol; b quantitative changes in the sugar phosphates soluble in 80% methanol; e changes in the dry weight of the cotton fibres and of their callose content, after treatment with 80% methanol, during development, and variation of the rate of incorporation of [~4C] sucrose into /~-glucan (cellulose+callose) by intact fibres of isolated seed clusters
and much smaller amounts of glucose l-phosphate were also detected in the methanol-soluble fraction. However, the amounts of the glucose 1-phosphate were normally insufficient for significant routine estimation. Maxima were observed for the three species [Figs. l(b), 2(b), and 3(b)] which occurred several days after the maxima observed for the monosaccharides. For G. arboreum and G. barbadense the delay observed was small - 7 and 3 d, respectively - while that for G. hirsutum, where the maximum itself was less well defined, was 15 d. For the three species there was usually two to three times more glucose 6-phosphate than fructose 6-phosphate. Determination of the (l~3)-~-D-glucan. It was previously shown (Pillonel et al. 1980) that treatment of cotton fibres with DMSO in the autoclave solubilizes essentially all of the (1 ~ 3)-/~-D-glucan.
484
J.P. Jaquet et al. : Non-structural carbohydrates in cotton fibres J~ glucose
8 6I 5 ~ a)1~1" ..... I
0~ "6
4
k~
3 2
9~
~
fit. ~
fructose 6-phosphate
24 t6
.~
12
[]
glucose 6-phosphate
1
-~
ii
cell wall dry weight (1- 3) - B- D- glucan
[
o
36 32 28
[]
fructose 6-phosphate
E
=- -= cell wall dry weight (1-3)-=3 D - glucan
E
Z
0.8
40
O.6
30;
0.4
20
0.2
10
0
0
9
17
27
32
37
42
48
r
Days post anthesis
Fig. 2a-e. Gossypium barbadense - a quantitative changes in the neutral sugars soluble in 80% methanol; b quantitative changes in the sugar phosphates soluble in 80% methanol; e changes in the dry weight of the cotton fibres and of their callose content, after treatment with 80% methanol, during development
glucose 6-phosphate
j__j
24 20 16
8
60
~
(b)
content
(c)
ca
sucrose
o
28
"6
9
d 9 . .~1 .....
o 3
[] fructose
content
(c) E
6O
1.0
50 m
O.8
40 ,im
0.6
30
0.4
20
0.2
10
0 21
E
26
32
36 39
45
56
~"
0
Days post anthesis
The material solubilized at the various stages o f d e v e l o p m e n t was h y d r o l y z e d and the glucose produced was determined. The values obtained are shown in Figs. l(c), 2(c), and 3(c) for G. arboreum, G. barbadense, and G. hirsutum, respectively. M a x i m a were obtained which occurred at the same time for that o f the monosaccharides (G. arboreurn), or which occurred slightly later (G. barbadense and G. hirsutum). The absolute a m o u n t s o f callose are low in the p r i m a r y cell walls and increase sharply at the transition period, marking the beginning o f secondary cell wall formation. After the m a x i m u m , the ( l ~ 3 ) - f l - D - g l u c a n content decreases slowly by a b o u t 50% until fibre maturity.
Rate of fl-gIucan synthesis. Pillonel et al. (1980) showed that the rate o f fl-glucan synthesis [cellulose and (1 --. 3)-fl-D-glucan], using sucrose as precursor, varied with fibre d e v e l o p m e n t and was at
Fig, 3a-e. Gossypium hirsutum - a quantitative changes in the neutral sugars soluble in 80% methanol; b quantitative changes in the sugar phosphates soluble in 80% methanol; e changes in the dry weight of the cotton fibres and of their callose content, after treatment with 80% methanol, during development a m a x i m u m at the middle o f massive secondary wall formation, but that the age at which the m a x i m a occurred depended to some extent on external climatic factors. The m a x i m u m appeared to occur later than the previously r e p o r t e d peak for the (1 ~ 3)-fl-D-glucan (Huwyler et al. 1979), but n o data were available for b o t h analysis f r o m the same series o f fruits. Such analyses are n o w rep o r t e d for G. arboreum using radioactive sucrose as the precursor. The rate o f synthesis o f the (1 ~ 3)-fl-D-glucan was obtained by measuring the radioactivity in the D M S O - s o l u b l e material (see above) and the rate o f cellulose synthesis by measuring the radioactivity in the residue after D M S O
J.P. Jaquet et al. : Non-structural carbohydrates in cotton fibres
485
treatment. As previously reported (Pillonel et al. 1980), incorporation into (1 ~ 3)-fl-D-glucan paralleled that into cellulose and at all fibre ages incorporation into the former was about twice that of the latter. Fig, l(c) shows the total incorporation (the sum of the two values obtained) into fl-glucan with respect to fibre development. A maximum was observed at 33 DPA, i.e., three days later than the maximum in the (1 ---,3)-fl-D-glucan content.
served for the monosaccharides fructose and glucose corresponding to the beginning of secondary wall formation. The amount of sucrose continued to build up slowly and was at a maximum at fibre maturity. A second maximum was observed for the (l~3)-fl-D-glucan content; this occurred closely after the maximum for the monosaccharides and just before the massive synthesis of the secondary wall. For G. arboreum, it was also established that the maximum rate of synthesis of /%glucan (cellulose and (1 ~3)-fl-D-glucan) occurred during this period of massive synthesis of the cell wall. The maxima observed for the sugar phosphates were in general less well defined, corresponding more or less in the case of G. arboreum and G. barbadense to the maximum for (1 -->3)-fl-D-glucan, while in the case of G. hirsutum a spread-out peak was observed, occurring later than the corresponding maximum in the (1 ~3)-fl-D-glucan. However, it should be noted that for G. hirsuturn the peak for the (1 ~ 3)-fl-Dglucan content is also less well defined than that for each of the other two species. These results indicate that during primary cell wall formation, sucrose, which is the transport sugar, is inverted and that glucose and fructose are stored. These reserves are used later for the formation of the secondary cell wall. Since there is a maximum in the (1 ~ 3)-fl-D-glucan content which might also have a reserve function (Meier 1981), following the maxima for the monosaccharides but before the period of massive secondary wall formation, it is possible that these monosaccharides are used in part for the synthesis of (1 ~3)-fl-Dglucan. This polysaccharide clearly undergoes a turnover during the formation of the secondary cell wall (Meier et al. 1981, Meier 1981), but it is not clear whether it is an obligatory or a facultarive intermediate in cellulose synthesis. The coincidence in the maxima for the sugar phosphates and the maximum for the rate of fl-glucan synthesis indicates that the former may be precursors for UDP-glucose which is known to be an excellent precursor for (1->3)-fl-D-glucan (Pillonel et al. 1980; Delmer et al. 1977) in cotton fibres. However, when UDP-glucose is fed to seed clusters or detached fibres the main product is sucrose. An interpretation of the variations of the sucrose level in the fibres during their development is difficult, since this is the main sugar transported via the phloem to the ovules and its concentration in the fibres depends on changes in its influx and efflux, which in turn depend on other factors external to the fruit. A knowledge of the concentrations of the various sugars in the different metabolic pools
Discussion
During the incorporation experiments in vitro, of the type previously described for cotton seed clusters (Pillonel et al. 1980) when sucrose was present in the incubation medium, there was a subsequent accumulation of glucose and fructose in the incubation medium, indicating the presence of an invertase in the fibres (unpublished observation). Buchala and Meier (unpublished data) have demonstrated the presence of cell surface located invertase for a Gossypium arboreum fibres, but the role of the enzyme is not clear. A possible function of the enzyme could involve regulation of cell wall synthesis by mediation of uptake of sucrose from the apoplast into the symplast. In order to understand the possible role of the enzyme, its effect on the amount of low molecular weight sugars and relationship to cell wall synthesis, a study of the low molecular weight sugars present in cotton fibers was undertaken. At the same time a comparative study of three species of cotton was carried out. Carpita and Delmer (1981) reported analyses for low molecular weight sugars in fibres from cotton (G, hirsutum) ovules cultured in vitro, but the study was directed toward the concentration and turnover of UDP-glucose, and no systematic study of the variations in the amounts of other sugars was reported. The sugar content of whole cotton (G. hirsutum) fruits was found to be influenced by the climatic conditions prevailing (Conner et al. 1972). Definite age-temperature relationships were established and they were associated with the cessation of fibre elongation and the initiation of secondary wall formation. These results are, however, difficult to interpret in terms of the fibres alone, although they may be quite pertinent to studies carried out with whole fruits, e.g., those described by Meier et al. (1981). The results obtained for the three different species of cotton show similar trends, although there were quite marked differences in the absolute amounts of a particular component from species to species. In each case, a first maximum was ob-
486
J.P. Jaquet et al. : Non-structural carbohydrates in cotton fibres
and of the enzymes which are active in the apoplast and the symplast is essential for an understanding of the synthesis of the cell wall fl-glucans. The results given are global for the apoplast plus the symplast, since even their separate analysis poses many technical problems. Work is at present being carried out to elucidate the metabolic pathway of sugars fed separately to the apoplast and symplast, respectively, of intact and detached cotton fibres.
Dubois, M., Gilles, K.A., Hamilton, J.K., Rebers, P.A., Smith, F. (1956) Colorimetric method for determination of sugars and related substances. Anal. Chem. 28, 350-356 Huwyler, H.R., Franz, G., Meier, H. (1979) Changes in the composition of cotton fibre cell walls during development. Planta 146, 635-642 Meier, H., Buchs, L., Buchala, A.J., Homewood, T. (1981) (I--* 3)-fl-D-Glucan (callose) is a possible intermediate in biosynthesis of cellulose of cotton fibres. Nature (London) 289, 821-822 Meier, H. (1981) Is callose a precursor of cellulose in cotton fibres? In: Cell walls '81, pp. 75-83, Robinson, D.G., Quader, H., eds. Wissenschaftl. Verlagsges, Stuttgart Meinert, M.C., Delmer, D.P. (1977) Changes in biochemical composition of the cell wall of the cotton fiber during development. Plant Physiol. 59, 1088-1097 Pillonel, Ch., Buchala, A.J., Meier, H. (1980) Glucan synthesis by intact cotton fibres fed with different precursors at the stages of primary and secondary wall formation. Planta 149, 306-312 Stahl, E. (1969) Thin-layer chromatography. Springer, New York Heidelberg Berlin Trevelyan, W.E., Procter, D.P., Harrison, J.S. (1950) Detection of sugars on paper chromatograms. Nature (London) 166, 444445 Zfircher, K., Hadorn, H., Strack, C. (1975) Gaschromatographische Zuckerbestimmung. Mitt. Lebensmittelchem. Hyg. 66, 92-116
The authors wish to thank Mrs. Th6r6se Homewood for help with the incorporation experiments and Dr. H. Hopf for a sample of umbelliferose. This project was supported by the Swiss National Science Foundation.
References Bergmeyer, H.U. (1974) Methoden der enzymatischen Analyse, pp. 1224-1287. Verlag Chemic, Weinheim Carpita, N.C., Delmer, D.P. (1981) Concentration and metabolic turnover of UDP-glucose in developing cotton fibers. J. Biol. Chem. 256, 308-315 Conner, J.W., Kreig, D.R., Gipson, J.R. (1972) Accumulation of simple sugars in developing cotton bolls as influenced by night temperatures. Crop Sci. 12, 752-754 Demol, J. (1964) Contribution fi l'6tude de l'influence du r6gime hydrique sur diff6rents facteurs de productivit6 chez Gossypium hirsutum L.. Bull. Inf. INEAC 13, 41-75 Delmer, D.P., Heiniger, U., Kulow, C. (1977) UDP-glucose: glucan synthetase in developing cotton fibers. Plant Physiol. 59, 713-718
Received 6 August; accepted 9 September 1982
