Atractyloside and carboxyatractyloside: A study of possible growth regulatory functions

S.-Afr.Tydskr.Plantk., 1990,56(5): 520-524

520

Atractyloside and carboxyatractyloside: A study of possible growth regulatory functions D.L. Fowlds, M.T. Smith and J. van Staden* UN/FRO Research Unit for Plant Growth and Development, Department of Botany, University of Natal, P.O. Box 375, Pietermaritzburg, 3200 Republic of South Africa Accepted 16 July 1990

The possible growth regulatory properties of the kaurenoid glycosides atractyloside and carboxyatractyloside, which show structural similarities to the gibberellins, were tested using the dwarf rice microdrop and lettuce hypocotyl bioassays. The glycosides did not show gibberellin-like activity, nor did they appear to be gibberellin antagonists. Die moontlike groeiregulerende eienskappe van die kaureno'fedglikosiede aktraktielosied en karboksi-aktraktielosied, wat strukturele verwantskappe met die gibberelliene vertoon, is ondersoek deur gebruik te maak van die dwerg rys-mikrodruppel- en slaaihipokotielbiotoetse. Die glikosiede het geen positiewe of antagonistiese gibberellienagtigeaktiwiteit getoon nie. Keywords: Atractyloside, bioassays, carboxyatractyloside,gibberellins, growth regulators *To whom correspondence should be addressed

Abbreviations ABA, abscisic acid; AMO-1618, (2- isopropy1-5-methy1-4-trimethy lammoniumchloride)-pheny l-l-piperidinium-carboxylate; ATS, atractyloside, 18-nor-ent-kaur-16-ene-15a-hydroxy-2f3-[[20-(3-methyl-1-oxybuty-1-)-3, 4-di-0-sulpho-a-D-glucopyranosyl] oxy] -19-oic acid; CATS, carboxyatractyloside, ent-kaur -16-ene-15a-hydroxy-2f3[[2-0-(3-methyl-1-oxybuty-1)-3, 4-di-0-sulpho-a-D-glucopyranosyl] -oxy-] -18,19-dioic acid; GA 3 , gibberellic acid; Paclobutrazol, (2RS 3RS)-1 ·-( 4-chlorophenyl)-4,4-dimethyl-2(l,2,4-triazol-1-yl)pentan-3-ol; Phosphon D, 2,4-dichlorobenzyl-tributhylphosphoniumchloride.

Introduction The gibberellins are 19- or 20-carbon tetracyclic diterpenoid compounds possessing an ent-gibberellane skeleton (Figure 1A) derived from ent-kaurene (Figure 1B) (Dalziel & Lawrence 1984; Jones & MacMillan 1984; Hedden 1987). Several other substances based on the ent-kaurene skeleton have been isolated from plants including the toxic glycosides ATS and CATS (Figure 1C). These compounds have been isolated from several plants, namely Atractylis gummifera L. (Danieli et al. 1972; Piozzi 1978), Callilepis laureola D.C. (Candy et al. 1977; Brookes 1979; Brookes et al. 1982), Xanthium strumarium L. (Craig et al. 1976; Cole et al. 1980), Wedelia glauca (Ort.) Hoffmann ex Hicken (Scheingart & Pomilio 1985) and Coffea arabica L. (Maier & Wewetzer 1978). While ATS was shown to be inactive, CATS (10- 3 M to lOs M) was found to be inhibitory in the wheat coleoptile bioassay and enhanced necrosis in tobacco and maize plants. The difference was attributed to the presence of two carboxyl groups in the latter glycoside (Cutler & Cole 1983). There is no information on an in vivo role for these compounds in

plants, although CATS has been implicated in the dormancy of the upper seeds of Xanthium (Cole et al. 1980; Cutler & Cole 1983). On the basis of structural homology these glycosides might be expected to show gibberellin-like activity, or possibly antigibberellin action by competition for a common binding site(s). Both ATS and CATS were examined in two gibberellin bioassays for possible growth regulatory properties. In addition the inhibitors of gibberellin biosynthesis AMO1618, paclobutrazol, and Phosphon D, were used in various concentrations with gibberellins to investigate possible synergistic or antagonistic effects. Materials and Methods Atractyloside (Sigma), CATS (Boehringer Mannheim), GA3 (Sigma), AMO-1618 (Calbiochem), Paclobutrazol (Culr.arR) (lCI) and Phosphon D (BDH) were used in this investigation.

OH

HO Me

,, "

to

Figure 1 The structure of (A) gibberellic acid or GA 3 ; (B) ent-kaurene; and (C) atractyloside (*in carboxyatractyloside this H is replaced by COOH).

521

S.Afr.J.Bot., 1990,56(5) CARBOXYATR A C TY LaSlO E

ATRACTYLOSIOE

i

G I BBERELLIN

(GA3>

60

~

...<:I:

+G A3

50

W

:I:

'"u.

40

< W

...

0

30

z

0

"w '"u.

20

a :I:

....

!

!

!

I

Ii

I

10

C!l Z W

...

9

0

3

La G N G IL

Figure 2 The effect of ATS and CATS, alone and response curve of the test material to GA3 is indicated.

III

combination with GA3 in the dwarf rice microdrop bioassay. The dose

The dwarf rice microdrop bioassay

This bioassay was based on that of Murakami (1968, 1970), with slight modifications (Harty 1986). Seeds of dwarf rice, Oryza sativa L. cultivar Tan-ginbozu were sterilized in 0.3% sodium hypochlorite containing 0.05% Tween 20, for 20 min. The seeds were thoroughly rinsed in tap water and then soaked in distilled water for 48 h at 32°C. The distilled water was changed after 24 h. After 48 h seeds were selected for uniform coleoptile emergence. Plastic repli-dishes were filled with 0.9% agar, pH 5.8, and two germinated seeds were planted per well. Ten seeds were used per treatment. The repli-dishes were kept in a moist environment under constant light and temperature (32°C). Two days after planting, the seedlings were treated with the test solutions. Droplets of 2 JJ.I were deposited in the axil of the first leaf of each seedling. Seedlings were then returned to the growth chamber for 3 days, after which measurements of the second leaf sheath were made. Experiments were repeated three times and the results were analysed statistically using a one-way analysis of variance. All test solutions were made up in 50% aqueous acetone and these were compared against a set of GA3 standards in the range of 0.1 to 1 000 ng per seedling. Aqueous acetone (50%) was used for the control. ATS and CATS were tested in the range of 0.002 to 400 ng per seedling. The lettuce hypocotyl bioassay

This bioassay was based on that of Frankland and Wareing (1960). Seeds of Lactuca sativa L. cultivar Supagreen were germinated in the dark at 25°C for 2 days. Seedlings with uniform radicles were then selected, placed in glass Petri dishes (5 cm) lined with filter paper and moistened with 1.2 ml of the appropriate test solutions. Ten seedlings were used per treatment. The Petri dishes were kept moist and were incubated at 28 :±: 1°C under continuous light. After 5 days the hypocotyllength was measured to the nearest mm. This assay was repeated two or four times for different treatments. The means and the 95% confidence intervals for the means

were converted into a percentage of the control and the results from all the experiments were averaged.

Results Dwarf rice microdrop bioassay

A linear response to GA3 was observed in the range of 0.2 to 1 000 ng per plant (Figure 2). At a concentration of 0.2 ng per plant and below the response was not significantly different from the water controls. All concentrations of ATS and CATS tested gave results that were not significantly different from the water controls. Combinations of ATS and CATS with GA3 (1: 1 to give a final GA3 concentration of 1 000 ng per plant), gave responses that were not significantly different from those of the equivalent GA3 concentration alone. The lettuce hypocotyl bioassay

GA3 at concentrations of 10-2 M and 10-4 M caused hypocotyl elongation that was significantly greater than those of the distilled water control (Figures 3 & 4). None of the concentrations of ATS and CATS tested gave hypocotyl growth significantly different from the water controls, although at 10-5 M CATS, the hypocotyls were shorter. This difference was not significant (Figure 3). Significant inhibition of seedling elongation was seen with the higher concentrations of the known gibberellin biosynthesis inhibitors tested, namely 10-3 M to 10-7 M AMO-1618 (not illustrated), 10-3 M Phosphon D and 3.4 X 10-6 M to 1.7 X 10-4 M Paclobutrazol (Figure 4). In the case of 3.4 x W- 5 M Paclobutrazol inhibition was 50% of the water controls. GA3 together with ATS, CATS or gibberellin biosynthesis inhibitors gave results that were significantly greater than those of the respective compounds alone_ Both 10-2 M and 104 M GA3 completely overcame the inhibition of seedling elongation caused by the lower concentrations of gibberellin biosynthesis inhibitors. However, at higher concentrations, namely 10-3 M AMO-1618 (results not shown), 10-3 M Phosphon D and 3.4 x 10-6 M and 1.7 x 10-4 M Paclobutrazol, inhibition was not completely reversed by

S.-Afr.Tydskr. Plantk., 1990,56(5)

522

200

T

150

100

-

::::

~ 8 50 ~ :::: (J

:;; OJ

~

~

~

~ ~ ~ 1<

~ 0 ::::

rif

f;;

'0 ...

?P.

J

f

~::::

Carboxyatractyloside

A tractyloside

Standards

. o

'"0

o

I

~

.,c:

.,c:

:::0 :::0

I

.S!

;;:

~

<

C!l

<'"

<'"

:::0

:::0

C!l

•

'"

I

I

0 ~

~

Figure 3 The effects of ATS and CATS, alone and in combination with GA3 , in the lettuce hypocotyl bioassay. The concentrations of ATS and CATS tested were: 10.7 M ~ ; 10.5 M 0 ; and 10. 3 M mm3 . Bars representthe 95% confidence limits .

either 10.2 M or 10-4 M GA3 (Figure 4). The responses to all concentrations of ATS and CATS tested, in combination with GA3, were not significantly different from the responses to GA3 alone. A combination of 10.3 MATS and 10-2 M GA3 resulted in marginally longer hypocotyls compared to the 10.2 M GA3 standard, but this difference was not significant.

in the lettuce hypocotyl bioassay are in keeping with a previous report (Leopold 1971), since the inhibition of seedling elongation could be completely overcome by GA3 application at the lower concentration of inhibitor tested. The results obtained with Paclobutrazol are comparable to those obtained in an early study with Ancymidol (Leopold 1971). Both these compounds inhibit the kaurene oxidase step in the gibberellin biosynthetic pathway and inhibited GA3-induced seedling elongation when given in combination with GA3, although Paclobutraxol was less inhibitory than Ancymidolon a molar basis. Lettuce seedling elongation was inhibited by the higher concentrations of the known gibberellin biosynthesis inhibitors. However, ATS and CATS had neither inhibitory nor enhancing properties on the seedlings. ATS and CATS

Discussion The rice microdrop bioassay was chosen since it is very specific for a gibberellic response and is not greatly affected by other growth-promoting or inhibiting compounds (Graebe & Ropers 1978). As can be seen, neither ATS nor CATS showed gibberellin-like activity, nor did they act as antigibberellins in this bioassay. The results with known gibberellin biosynthesis inhibitors

160 Standards

eJ

140

l

120

~

Paclobutrazol 3 1

Phosphon D

~.

J ::~ ~

100

!;:

~

Ii !~. Ii

Ii: I!::~: Il:

I~ i~

I

20

!;:!~

o :;; OJ

~

:::0 :::0

•'0 '" I

~

~

.,

c: 0

;;:

<'"

<'"

C!l

:::0

:::0

C!l

'"e I

• I

~

.,c: 0

;;:

C!l

<'"

<'"

:::0

:::0

'"0 I

~

C!l

• I

0 ~

Figure 4 The effects of Paclobutrazol and Phosphon D, alone and in combination with GA3 , in the lettuce hypocotyl bioassay. The concentrations of Paclobutrazol and Phosphon D tested were: 3.4 X 10.7 M Paclobutrazol ~ ;10. 5 M Phosphon D or 3.4 X 10. 6 M Paclobutrazol J:.:;:;:J; 10.3 M Phosphon D or 3.4 X 10.5 M Paclobutrazol ~; and 1.7 X 10.4 M Paclobutrazol~ . Bars represent the 95% confidence limit.

523

S.Afr.J.Bot., 1990,56(5)

also had no effect when applied in combination with GA3. Therefore, as with the rice microdrop bioassay, ATS and CATS showed neither gibberellin-like nor gibberellinantagonistic activity. This inactivity of ATS and CATS may be due to the fact that kaurenoid compounds and gibberellin precursors are inactive in gibberellin bioassays. However, it has previously been shown that these compounds are active in the dwarf rice microdrop bioassay (Murakami 1970). Kaurene and kaurenol were slightly active at 4 and 2 J..Lg/plant respectively. Kaurenoic acid and steviol (the aglycone of stevioside) show clear activity above 0.2 J..Lg/plant. The increase in the latter was equivalent to 0.5 M J..Lg/plant GA 3. The precursor, mevalonic acid was inactive. These kaurenoid precursors also showed activity in the dwarf maize (d-5 and an-l mutants) bioassay (Katsumi et al. 1964). In this bioassay kaurenol, kaurenoic acid and steviol, but not kaurene, were active in stimulating leaf sheath elongation, although these compounds were only 1-10% as active as GA3 at 100 J..Lg dosage levels (phinney et al. 1964). The biological activities of the kaurene derivatives and the gibbere11ins may reflect similarities in active sites. However, it is also possible that these compounds are metabolically related, such that kaurenoic acid, kaurenol and steviol are precursors that can be converted to gibberellin by the two maize mutants (Katsumi et al. 1964). This would involve a lag phase, therefore the lower activity. The other alternative as to why ATS and CATS were inactive in the two bioassays studied, is that these compounds cannot enter into the plant cell due to the sugar moieties. Steviol has been shown to be active in both the dwarf rice microdrop (Murakami 1970) and the dwarf maize (Katsumi et al. 1964) bioassays. Stevioside, and the closely related rebau dios ide, showed effects similar to gibberellins in the activation of amylase in cereal, and in the promotion of growth when applied to whole plants. In Japan these compounds have been patented as growth regulators (Handro & Ferreira 1989). However, there is little information on the biological activity of the glycosylated gibberellins and ATS and CATS. In the dwarf pea bioassay, under red light conditions, GAgglucoside showed low but significant activity of the same order as GAg. This indicates that uptake of a glycoside may not be a limiting factor. It has also been shown that GA 3-g1ucosides have a low potency in most bioassays studied, including the dwarf pea and the lettuce hypocotyl bioassays (Sembdner et al. 1970). On the other hand, the GArglucosides show high activity in the dwarf pea and rice bioassays only when applied via the roots. This suggests that this glucoside cannot be hydrolysed by the plant and therefore alternative hydrolysis, possibly microbial, must take place at the root surface or in the surrounding medium before uptake, in order for GAr glucosides to be effective (Sembdner et al. 1970). As the GAg-glucoside was active in leaf applications, penetration of the GArglucoside is probably not the limiting factor. In order to see if the GAg-glucoside was hydrolysed by the plant, dwarf peas were supplied with a f3-g1ucosidase inhibitor. From the results it was concluded that enzymatic hydrolysis of GAg-glucosides does in fact take place and therefore its measurable activity in the dwarf pea bioassay

must depend upon its hydrolysis (Sembdner et al. 1970). Therefore it may be assumed that the plant glucosidases cannot hydrolyse GArglucosides, perhaps due to stereochemical differences between 0(3) and 0(2)-glucosides. It also seems that glycosylation of GA leads to deactivation, which is reversible at least with the 0(2) -glycosides. Stevioside has sugars attached at carbons 13 and 19 and is presumably also hydrolysed by the plant glucosidases. In nine gibberellin bioassays it was found that, amongst other gibbere11ins, GAg and GAg-glucoside, exhibited a11round low activity at the 1 J..Lg level (Crozier et al. 1970). An attempt was made to correlate structure to the level of activity in the bioassays. The highly active gibberellins (1, 2, 3,4, 5, 6, 7, 22, 23) are all oxygenated at the C3 position, or have a C2 , 3 double bond. These structural features are exclusive to the tested gibberellins showing high activity with the exception of gibberellins 8, 26 and 27, which are relatively inactive. In addition to having a C 3 hydroxyl group these three gibberellins are also C 2 hydroxylated and this could account for their reduced biological activity (Crozier et al. 1970). This could be supported by the fact that stevioside, kaurene, kaurenol and kaurenoic acid are all active in bioassays but none contain C2 hydroxyl groups. ATS and CATS are 0(2)-glucosides, and on the basis of structural homology might be expected to behave as GAgglucoside. This means they would be readily hydrolysed and absorbed. However, as with GAg, the presence of tfie C2 hydroxyl group may reduce the biological activity of ATS and CATS to such an extent that this activity is negligible. It is suggested therefore that the lack of biological activity of ATS and CATS in plant tissue is a property of the structural and, as yet unknown, metabolic role of these compounds.

Acknowledgements The Foundation for Research and Development (FRD) is thanked for financial assistance.

References BROOKES, K.B. 1979. A chemical study of the rootstock of Callilepsislaureola . M.Sc. Thesis, Univ. of Nata!. BROOKES, K.B., CANDY, H.A. & PEGEL, K.H. 1982. Atractylosidesin Callilepislaureola (Asteraceae). S. Afr. J. Chern. 36: 65-68. CANDY, H.A., PEGEL,K.H., BROOKES, K.B. & RODWELL, M. 1977. The occurrence of atractyloside in Callilepis laureola. Phytochern. 16: 1308. COLE, R.J., STUART, B.P., LANSDEN,J.A. & COX, R.H. 1980. Isolation and redefinition of the toxic agent from the cocklebur (Xanlhiurnslrumarium).J. Agric. Food. Chern. 28: 1330-1332. CRAIG, J.e. (Jun.), MOLE, M.L., BILLETS, S. & EL-FERALY, F. 1976. Isolation and identification of the hypoglycemic agent, carboxyatractylate, fromXanlhiurn slrumarium. Phylochem. 15: 1178. CROZIER, A., KUO, C.C., DURLEY, R.C. & PHARIS, R.P. 1970. The biological activities of 26 gibberellins in nine plant bioassays. Can . J . BOI. 48: 868-877. CUTLER, H.G. & COLE, R.J. 1983. Carboxyactractyloside: A compound fromXanlhiurn slrumarium and Alraclylis gummifera with plant growth inhibiting properties. The probable 'Inhibitor A'. J. Nat. Prod. 46(5): 609-613.

524 DALZIEL,I. & LAWRENCE, D.K. 1984. Biochemical and biological effects ofkaurene oxidase inhibitors, such as Paclobutrazol.Monogr.Br.PI. Growth Reg. Gr. 11 : 43-57. DANIEU;B., BOMBARDEU, E., BONA11, A. & GABETIA, B. 1972. Structure of the diterpenoid carboxyatractyloside. Phytochem. 11: 3501-3504. FRANKLAND, B. & WAREING, P.F. 1960. Effects of gibberellic acid on hypocotyl growth of lettuce seedlings. Nature 185: 255 . GRAEBE, I.E. & ROPERS, H.I. 1978. Gibberellins- sites of formation. In: Plant growth substances, eds Letham, D.S., Goodwin, P.B. & Higgins, T.I.V.,pp. 71-80, Academic Press, London. HANDRO, W. & FERREIRA, C.M. 1989. Stevia rebaundiana (Bert.) Bertoni: Production of natural sweeteners. In: Biotechnology in agriculture and forestry. Medicinal and aromatic plants. II, ed. Bajaj, Y.P.S ., Vol. 7, Ch. 26, SpringerVerlag, Berlin Heidelberg. HARTY, A.R. 1986. Control of flower initiation in 'Eureka'lemon. M.Sc. Thesis, Univ. of Natal. HEDDEN, P. 1987. Gibberellins . In: Principles and practice of plant hormone analysis, eds. Rivier, L. & Crozier, A., Vol. I, pp. 9-110, Academic Press, London. JONES, R.L. & MACMILLAN, 1. 1984. Gibberellins.In: Advanced plant physiology, ed. Wilkins, M.B., pp . 21-52, Pitman Press, Bath.

S.-Afr.Tydskr.Plantk., 1990, 56(5) KATSUMI, M., PHINNEY, B.O., JEFFERIES, P.R . & HENRICK, G.A. 1964. Growth response of the d-5 and an-I mutants of maize to some kaurene derivatives. Science 144: 849-850. LEOPOLD,A.C. 1971. Antagonism of some gibberellin actions by a substituted pyrimidine. PI. Physiol. 48: 537-540. MAIER, H.G. & WEWETZER, H. 1978. Bestimmung von diterpenglykosiden im bohnenkafee. Z. Lebensm. Unters Forsch. 167: 105-107 . MURAKAMI, Y. 1968. A new rice seedling test for gibberellins, microdrop method, and its use for testing extracts of rice and morning glory. Bot. Mag . Tokyo 81: 33-43. MURAKAMI, Y. 1970. New rice seedling test for gibberellinsmicrodropmethod. lap. Agric.Res. Quart. 5: 5-9. PHINNEY, B.O., JEFFERIES, P.R., KATSUMI, M. & HENRICK, C.A. 1964. The biological activity of kaurene and related compounds. Pl. Physiol. 39 suppl. XXVII. PIOZZI, F. 1978. The chemistry of Atractyloside. In: Atractyloside. Chemistry, biochemistry and toxicology, eds. Sariti., R. & Luciani, S., pp. 13-32, Piccin Editore, Padova (Italy). SCHEINGART,C.D. & POMILIO, A.B. 1985 . Atractyloside, toxic compound from Wedeliaglauca.l. Nat. Prod. 47: 1046- 1047. SEMBDNER, G., WETLAND, I., SCHNEIDER, G., SCHREIBER, K. & FOCKE, I. 1970. Recent advances in the metabolism of gibberellins . In: Plant growth substances, ed. Carr, D.I., pp. 143-150, Springer-Verlag,Berlin.