Planta (1984)160:242-249
P l a n t a 9 Springer-Verlag 1984
The stability and biological activity of cytokinin metabolites in soybean callus tissue L.M.S. Palni, M.V. Palmer and D.S. Letham Research School of Biological Sciences, Australian National University, P.O. Box 475, Canberra City, ACT 2601, Australia
Abstract. The activity, uptake and metabolism of cytokinin metabolites was determined in soybean (Glycine max (L.) Merr.) callus tissue. The following activity sequence was established: zeatin riboside (ZR) > zeatin (Z) > O-glucosides of Z, ZR and their dihydro derivatives > lupinic acid (an alanine conjugate of Z ) > 7- and 9-glucosides of Z which were almost inactive. The 7- and 9-glucosides and lupinic acid were taken up very slowly by the callus tissue and showed great metabolic stability, but some degradation to 7-glucosyladenine, 9-glucosyladenine and the 9-alanine conjugate of adenine occurred. Compared with its aglycone, O-glucosylZR exhibited slow uptake and greatly enhanced stability but gas chromatographic-mass spectrometric analysis showed that appreciable amounts were hydrolyzed to ZR in the tissue. Both ZR and O-glucosyl-ZR were metabolised extensively, with adenine, adenosine, and adenine nucleotide(s) as the major metabolites. A diversity of minor metabolites of ZR were identified, including O-glucosides, lupinic acid and dihydrolupinic acid. The metabolism of ZR was suppressed by 3-isobutyl-1methylxanthine. When compared with the soybean callus line normally used for cytokinin bioassays (cv. Acme, cotyledonary callus), related callus lines exhibited greatly differing growth responses to cytokinin; however, these were not reflected in marked differences in metabolism. Key words: Cytokinin (activity, stability) - Glycine (cytokinin metabolism) - Tissue culture (cytokinin metabolism). Abbrevh~tions: GC-MS = gas chromatography-mass spectrome-
try; HPLC=high-performance liquid chromatography; L A = lupinic acid; OGZR = O-fl-D-glucopyranosylzeatin riboside; TLC = thin-layer chromatography; IMX = 3-isobutyl-l-methylxanthine, Z = zeatin; ZR = zeatin riboSide
Introduction
Bioassays based on plant tissue cultures are frequently used to assess the cytokinin activity of plant extracts. Although cytokinin glucosides occur commonly in such extracts and may even be the dominant cytokinins present (see review by Letham and Palni 1983), there is very little information regarding the activity of these glucosides in tissue-culture bioassays. Furthermore, the activity of lupinic acid (LA), a naturally occurring alanine conjugate of zeatin, has not been defined in such assays. The activity of a cytokinin in a bioassay must depend to some extent on the effectiveness of the cytokinin per se at the functional site, but it will also be influenced by uptake, metabolic stability and the activity of any metabolites formed. However, the uptake and metabolic stabilities of naturally occurring cytokinin glucosides and alanine conjugates have not been determined in any bioassay system. Accordingly, in the present study, nine naturally occurring cytokinins, including glucosides and LA, were tested for activity in the soybean-callus bioassay. In complementary experiments using the same tissue type (cotyledonary callus of soybean cv, Acme), the uptake and metabolism of radioactive analogues of five of these cytokinins was assessed, 20 h after their application. The observed metabolism of zeatin riboside (ZR) was also compared with that of ZR in two other soybean callus lines which differed markedly from the normal Acme callus in their response to supplied ZR. One of these was an unusual, chlorophyll-containing callus line derived from normal Acme callus. Materials and methods Chemicals. Zeatin (Z), isopentenyladenine (IP) and their corresponding 9-fl-D-ribosides (ZR and IPA, respectively) were ob-
L.M.S. Palni et al. : Biological activity and metabolism of cytokinins
243
tained commercially (Sigma Chemical Co., St. Louis, Mo., USA, and Calbiochem-Behring Corp., La Jolla, Calif., USA); all other cytokinins were synthesized chemically by methods reported previously (Cowley et al. 1978; Duke et al. 1978, 1979). The penta-deuterium-labelled O-fl-D-glucopyranosylzeatin riboside ([2Hs]OGZR) was prepared according to Summons et al. (1979). 7-/~-D-Glucopyranosyladenine (adenine-7glucoside; Cowley et al. 1978) and 9-~-D-glucopyranosyladenine (adenine-9-gtucoside; Davoll et al. 1946) were synthesized according to the cited references. ]~-(6-Aminopurin-9-yl)alanine, the adenine-9-alanine conjugate, was synthesized by heating 3-(6-ehloropurin-9-yl)-N-trifluoroaeetylalanine (Duke et al. 1978) with ammonia at 100~ C in a sealed tube, and the trifluoroacetyl group was cleaved with 1 N NaOH. [8-3H]Zeatin riboside (9.04 GBqmmol-1), [8-~H and 2Hs]OGZR (614MBq mmo1-1) and [8-3H] LA (981 MBq mmol -I) were obtained by heating respective non-radioactive compounds with 3HzO under conditions known to result in selective exchange of the C-8 hydrogen of purine ring (Shelton and Clark 1967). [2,83H]Zeatin was prepared as described by Letham and Young (1971), and used for the enzymic synthesis (Entsch et al. 1979) of its 7- and 9-fl-D-glucopyranosyl derivatives, Z-7-glucoside (3.26 GBq mmo1-1) and Z-9-glucoside (6.4 GBq mmol-1). These were diluted with unlabeUed compounds to give the specific activities cited. All radiolabelled cytokinins were stored in 50% aqueous ethanol at - 2 0 ~ C and purified by thin-layer chromatography (TLC) and-or high-performance liquid chromatography (HPLC) prior to use. The product identities were established by TLC, HPLC, ultraviolet (UV) and mass spectrometry. 3-Isobutyl-l-methylxanthine (IMX) was obtained from Sigma. Three dyes used as TLC markers were: Meldola Blue (Gurr, Searle Diagnostic, High Wycombe, UK), Drimarene Brilliant Blue K-BL (Polysciences, Warrington, Pa., USA), and Rhodamine B (Hopkins and Williams, Chadwell Heath, Essex, UK).
kept in the light. The results presented are the average of four flasks for each cytokinin concentration tested. Metabolic studies. Aqueous solutions of radioactive cytokinins were filter-sterilised and added to autoclaved basal medium (without 6-benzylaminopurine and agar) to give the following concentrations (gM): Z-9-glucoside and ZR-O-glucoside (0.5), Z-7-glucoside (0.55), ZR (0.6), LA (0.7). Callus tissue (20 g) was divided into small pieces and transferred aseptically to each conical flask containing liquid medium (50 nal) with added radioactive cytokinin. The cultures were incubated on a reciprocating shaker (50 rpm) at 26~ C in the dark, except for Green tissue which was incubated in the light. During incubation the callus pieces remained essentially intact. After 20 h the tissue was removed by filtration, washed with water (20 ml), and extracted as reported by Palni and Horgan (1983). The washings were combined with filtered medium, and an aliquot was withdrawn for radioactivity determination to assess the uptake of supplied cytokinin by callus tissue. The tissue extract was passed through a column of insoluble polyvinylpyrrolidone (Calbiochem, San Diego, Calif., USA) followed by cation-exchange chromatography on a column of cellulose phosphate, as detailed by Palni et al. (1983)9 This yielded a basic fraction (which would include cytokinin bases, glycosides, and alanine conjugates), and an acidic-neutral fraction containing cytokinin nucleotides. The nucleotide fraction was treated with Escherichia coli alkaline phosphatase (Sigma), followed by extraction with four equal volumes of water-saturated butan-l-ol (Palni and Horgan 1983). Cytokinin O-glucosides were hydrolysed to their respective aglycones using sweet-almond/?-glucosidase (Boehringer Biochemica GmbH, Mannheim, FRG) as reported by Palni and Horgan (1983). N-Glucosides of Z and adenine were hydrolysed by heating at 95 ~ C for 60 min in 0.5 N HC1. Radioactivity determinations were done according to Gordon et al. (1974) using an LKB 1215 Rackbeta II liquid scintillation counter (Wallac Oy, Turku, Finland).
Callus cultures and bioassay. The soybean (Glycine max (L.) Merr. cv. Acme) cotyledonary callus and the chlorophyllus line derived from it (see Wang 1979) were a gift of Dr. T.L. Wang (John Innes Institute, Norwich, UK), and have been maintained in our laboratory for the last 2.5 years. These two callus lines will be termed "Normal" and "Green", respectively, throughout this paper. Root and cotyledonary callus (further referred to as "Root" and "Coty") were initiated from soybean cv. Clark ex 63 (Murrumbidgee Irrigation Authority, Leeton, N.S.W., Australia) by the procedure of Yeoman and Macleod (1977). All callus lines used were cytokinin-dependentand were maintained on basal medium (Miller 1968) solidified with agar (0.8%, w/v), and supplemented with ~-naphthaleneacetic acid (2 mg 1-1) and 6-benzylaminopurine (1 mg 1-1). All cultures were grown at 26~ in the dark, except for the Green line which was grown in light as described by Wang (1979). Callus tissue for metabolic experiments and bioassay was used 24 d after subculture. Cytokinin bioassays were performed according to Miller (1968) using cotyledonary (Normal) callus. Three callus explants, approx. 30 mg each, were placed in each 50-ml Erlenmeyer flask containing 25 ml of medium. Cytokinin solutions were prepared in dimethylsulfoxide (DMSO; Mallinckrodt Chemical Works, St. Louis, Mo. USA) and their concentrations determined by UV absorbance; test solutions of cytokinins were added to autoclaved medium in individual flasks just before gelation. The concentration of DMSO in the final medium was kept below 0.2% (v/v) and DMSO was also added to the control flasks. Callus fresh weights were determined after growth for 24 d at 26~ C in dark, except the Green callus which was
Thin-layer chromatography. Small aliquots (2% by vol.) of the basic fraction and the nucleotide-derived, butanol-soluble fraction were analysed initially by two-dimensional TLC on 0.25-mm-thick PF254 Silica Gel 60 plates (Merck, Darmstadt, FRG), developed first in butan-l-ol: acetic acid:water (12:3:5, by vol.) (system 1), and then in butan-l-ol:14 M ammonia: water (6:1:2, by vol., upper phase) (system 2). This procedure separates isopentenyIadenine,its riboside, O-glucosides of ZR, dihydro-Z and dihydro-ZR, Z-9-glucoside, adenine and adenosine from one another and from the following pairs of compounds which are not resolved (see Table 1) : LA and dihydrolupinic acid; Z-O-glucoside and Z-7-glucoside; ZR and dihydroZR; Z and dihydro-Z. The two-dimensional procedure was used to obtain tentative identification of the major metabolites, estimates of their levels and of the distribution of radioactivity in each extract. Subsequently larger aliquots of extracts were purified by preparative TLC (system 1) alone, for tIPLC analysis. Three marker dyes (Summons et al. 1981) were employed to locate zones on the TLC plates (Table 1); these zones were then scraped off and eluted in 40-60% aqueous m~thanol containing 6% acetic acid. In this way, five groups oficompounds were readily obtained: LA and dihydro-LA (below Meldola Blue); O-glucosides of Z, ZR, their dihydro derivatives and Z-7-glucoside (zone above Meldola Blue); Z-9-glucoside, adenine and adenosine (Drimarene Brilliant Blue zone + zone below); Z, dihydro-Z, ZR and dihydro-ZR (zone between Drimarene Brilliant Blue and Rhodamine B); and isopentenyladenine and isopentenyladenosine (zone above Rhodamine B). Recovery of radioactive Z, ZR, ZR-O-glucoside, Z~9-glucoside and Z-7-glucoside from Silica-gel TLC plates was greater than 95% ; the recovery of LA was 75%.
9
244
L.M.S. Palni et al. : Biological activity and metabolism of cytokinins
Table 1. Resolution of various cytokinins and marker dyes on TLC and HPLC Compound
Lupinic acid (LA) Dihydro-LA Meldola Blue Z-7-glucoside Z-O-glucoside Dihydro-Z-O-glucoside ZR-O-glucoside Dihydro-ZR-O-glucoside Z-9-glucoside Adenosine Adenine Drimarene Brilliant Blue Zeatin (Z) Dihydro-Z Zeatin riboside (ZR) Dihydro-ZR Rhodamine B Isopentenyladenine Isopentenyladenosine
Rf value (TLC)
Elution time (HPLC, rain)
System 1
System 2
System A
System B
System C
0.25 0.25 0.27 0.32 0.31 0.32 0.32 0.34 0.39 0.42 0.42 0.44 0.48 0.48 0.49 0.49 0.55 0.60 0.60
0.18 0.18 . 0.18 0.18 0.22 0.13 0.15 0.19 0.30 0.50 . 0.58 0.58 0.36 0.36 . 0.65 0.45
-
28.9 -
17.4 17.4
7.2 7.2
8.5 17.4 12.5 8.5 -
14.7 16.7 14.7 -
-
-
-
-
.
.
.
. 18.6 28.9 32.7 22.0 25.0 18.6 5.5 7.5
.
. 12.0 14.6 5.5 7.5 .
.
. -
. .
System D
. .
.
.
-
High-performance liquid chromatography. The
HPLC equipment and conditions were the same as used by Summons et al. (1980). The solvents used were mixtures of methanol and water containing 0.2 M acetic acid, the proportion of methanol being given as percentage by volume. Radioactive fractions which had been prepared by TLC (system 1 ; see Table 1 and inset, Fig. 3) were then analysed by HPLC using one or more of the following systems: C 8 radial compression column (8 mm diameter, 100mm long; Waters Associates, Milford, Mass., USA) eluted with a linear gradient of 10% to 60% methanol (system A) or 5% to 40% methanol (system B) in 30 min, or isocratic elution with 20% methanol (system C) at 3 ml m i n - 1 ; pBondapak phenyl column (3.9 mm diameter, 300 mm long; Water Associates) eluted with a linear gradient of 0% to 10% methanol in 20 rain at 2 ml min- 1 (system D). In a few cases only two H P L C steps were sufficient to resolve completely a complex TLC eluate into its components (Fig. 3).
Mass spectrometry. Per-trimethylsilyl derivatives were prepared by heating the sample in 10 gl of pyridine (Ajax Chemicals, Sydney, N.S.W., Australia): N,O-bis(trimethylsilyl)trifluoroacetamide (Pierce Chemical Co., Rockford, Ill., USA) (1 : 1, v/v) at 90~ for 10 min. The sample was introduced via a direct inlet probe, using a Vespel extension (du Pont Co., Wilmington, Del., USA) fitted to a Finnigan-4500 instrument (Finnigan Corp., Sunnyvale, Calif., USA). The mass spectrometer was scanned from m/e 530 to 730 0.1 s-1 and the electron-impact spectra were integrated over the desorption profile. Other details of equipment and conditions used have been described in Palni et al. (1983) and Summons et al. (1983).
Results
Activities of cytokinins in soybean callus bioassays. T h e f o l l o w i n g a c t i v i t y s e q u e n c e f o r Z a n d its m e tabolites was established for the Normal soybean callus bioassay: ZR > Z > (O-glucosides of Z, ZR
l
.r zR
c
8
~//[? i i Z
3oo
/ u
o
. ~ ~
OGDZ
//
?oozR
c
c
,oo
U . " !.
>Z-9-glucoside > LA >>ZR-O-glucoside > ZR. In the extreme case, 85% of the Z-7-glucoside taken up by the callus was recovered as such in the extract. In all cases where ZR or its O-glucoside was supplied, adenine nucleotide(s), adenine and adenosine were the major metabolites identified, accounting for 22-40% of the radioactivity taken up (Table 2). Unidentified neutral compounds comprised up to 29% of the radioactivity originally
246
L.M.S. Palni et al. : Biological activity and metabolism of cytokinins
ZR
{TMsih-z,
~n ] |
~-
~/"t
mB
D ~ 8 RB
"
7n- -
N
05
"0
g
h5
50
4_Hj D5 M? C 629 644
{TMSi)5- ZR
o
0~,
o
Mt
120
RF
Mt-CH 3
IO-
701 Z
0
5
10
,
DZ
15
D5- Mt 716
639
20
25
Elution time (min)
Fig. 3. Purification of basic metabolites of [8-SH]ZR extracted from Normal soybean callus. Preparative TLC (System 1) was used to separate groups of metabolites (inset), which were then resolved by HPLC. In this case the TLC zone between marker dyes DBB and RB was eluted and subjected to a "bulk HPLC" step (Ca radial compression column, eluted isocratically at 3 ml min -1 with 35% methanol). The resulting '~ peak" was further analysed by HPLC (system A) to resolve ZR, dihydro-ZR (DZR), Z and dihydro-Z (DZ). Bars indicate the position of marker dyes (inset): MB, Meldola Blue; DBB, Drimarerie Brilliant Blue; RB, Rhodamine B
supplied as ZR or its O-glucoside. Isopentenyladenine, its riboside or the nucleotide(s) were not found as metabolites in any of the experiments. Minor metabolites of ZR (comprising less than 3% of the radioactivity taken up) identified in all cases were Z, dihydro-Z, dihydro-ZR, O-glucosides of Z, ZR and their dihydro derivatives, LA, dihydro-LA, and nucleotide(s) of Z and dihydro-Z. The free bases and ribosides (including those derived from nucleotides after treatment with alkaline phosphatase) were identified by co-chromatography with authentic standards on TLC (system 1 followed by system 2) and HPLC (system A; Fig. 3). The O-glucosides ofcytokinins were identified by TLC in the same way followed by HPLC (system B), and confirmation of their identities was also obtained by HPLC (system A) of the corresponding aglycones following hydrolysis by fl-glucosidase. Identification of LA and dihydro-LA was based initially on co-chromatography in two TLC systems (1 and 2) and two HPLC systems (C and D, in which these two metabolites are not resolved from each other), and final separation of the two compounds by TLC on cellulose, developed in solvent system 2 (Rf of LA: 0.29, and dihydro-LA: 0.35). The enhanced stability of ZR in IMXtreated tissue was not reflected in lower levels of any identified metabolite(s). A marked feature of ZR-O-glucoside metabo-
,,,r
~00
....
I
i ........
I,, hli h
J. . . . . .
650
,i
....
, ....
r./~
, ....
,rmJl)l,,,,i Ih,,
, ....
I "
' ....
'"
'
700
Fig. 4. Molecular-ion regions of the electron-impact mass spectrum of tetra- and penta-trimethylsilylderivatives of penta-deuterium (Ds)-labelled ZR, isolated as a metabolite of [8-3H and 2H5]ZR-O-glucoside in Normal soybean callus tissue
lism in Normal soybean callus was its hydrolysis to ZR, identified by co-chromatography with the authentic compound on TLC and HPLC, which accounted for 5.3% of radioactivity taken up. The radioactivity also corresponded to the Rf of t r a n s ZR on Silica-gel plates developed in chloroformmethanol (9:1). To confirm this observation [8-3H and 2Hs]OGZR was diluted with [2Hs]OGZR to half its specific activity and supplied at a concentration of 2.2 gM to 100 g of Normal tissue as described earlier. Tissue extraction and purification of metabolites was performed 20 h later and HPLC-purified ZR was subjected to GC-MS analysis as a mixture of its tetra- and penta-trimethylsilyl (TMSi) derivatives. In addition to the molecular ions (M +.) for (TMSi) 4- and (TMSi)s-ZR and the corresponding M .+-CH5 ions (all from the deuterated, i.e., D s compound), ions at m/e 639 and 624 (Fig. 4) were also seen. These ions correspond to M .+ and M.+-CH~ ions for the unlabelled compound and probably represent very low levels of endogenous ZR in the callus tissue. The N-glucosides of Z, Z-7-glucoside in particular, exhibited high metabolic stability in the tissue (Table 2). However, 2.8 and 7.4% of radioactivity taken up as Z-7-glucoside and Z-9-glucoside, respectively, was shown to co-chromatograph with adenine-7-glucoside and adenine-9-glucoside by TLC on Silica-gel (systems ] and 2) and cellulose plates, and by HPLC (see Table 3 for details). When Z-7-glucoside, Z-9-glucoside (recovered from the tissue) and adenine-9-glucoside were subjected to acid hydrolysis the radioactivity was found to co-chromatograph with the respective bases (Z or adenine) on TLC system 1. However,
L.M.S. Palni et al. : Biological activity and metabolism of cytokinins
247
Table 3. TLC and HPLC data for the new cytokinin metabolites. TLC systems 1 and 2 are described in Materials and methods.
TLC system 3 : 0 . 2 5 mm cellulose (Serva Feinbiochemica, Heidelberg, FRG) plates run in solvent system 2. HPLC was done using a C 8 radial compression column (8 x 100 ram), eluted isocratically with 5% methanol containing 0.2 M acetic acid at 3 ml m i n - 1. Values in parentheses are for the parent cytokinins : a, Z-7-glucoside; b, Z-9-glucoside; c, lupinic acid Metabolite
Adenine-7-glucoside Adenine-9-glucoside Adenine-9-alanine
Rf value (TLC)
HPLC elution time (rain)
System 1
System 2
System 3
0.35 0.41 0.13
0.11 0.14 0.08
0.11 (0.28)" 0.12 (0.32) b 0.04 (0.14) c
in the case of adenine-7-glucoside sufficient radioactivity was not available for this purpose. When LA was supplied to soybean tissue, 14.5 and 1.2% of radioactivity taken up was identified as adenine-9-alanine and Z, respectively (details to be published elsewhere). Some TLC and HPLC data for adenine-9-alanine are given in Table 3. Discussion
At present, the bioassay which is used most frequently for assessment of cytokinin levels in plant tissue extracts is the soybean callus assay developed by Miller (1968). However, of the commonly used cytokinin bioassays, this is the one for which cytokinin structure-activity relationships have been least defined. From the integrated approach described above, we are able firstly to supplement the available structure-activity information for this system with data for several additional naturally occurring cytokinins, and secondly to attempt to explain these relationships by comparison with parallel studies of uptake and metabolism. In contrast to the 7- and 9-glucosides of Z, the side-chain O-glucosyl conjugates of Z and ZR did not differ markedly from the respective aglycones in their biological activity. Similar results have been reported for the radish-cotyledon, the tobacco-callus and the Amaranthus-betacyanin bioassays (Letham et al. 1983). Side-chain saturation of the O-glucosides markedly reduces biological activity in the radish-cotyledon and Amaranthus-betacyanin bioassays, but no such effect was found in the soybean callus assay, nor previously in the tobacco callus assay (Letham et al. 1983). Relative to Z, LA is essentially inactive in the tobacco-callus and radish-cotyledon bioassays. Hence it is of considerable interest that LA exhibits appreciable activity in the soybean callus bioassay and that a similar relative activity has recently been reported for this metabolite in the Amaranthus bioassay (Letham et al. 1983). Another interesting feature of the soybean callus assay in common with
7.70 (27.23) a 5.38 (14.68) 8 10.35 (39.43) ~
the soybean hypocotyl test (Newton et al. 1980) is the fact that ZR exhibits greater activity than Z. In most other bioassays including the tobacco callus assay, the converse applies (Conrad 1974; Letham et al. 1983; Schmitz et al. 1972). The present studies further establish the usefulness of the soybean assay in that it detects the complex of ubiquitous O-glucoside metabolites of Z at low concentrations and exhibits a linear dose-response relationship for them over a wide concentration range. The use of large callus explants, in which only the basal cells are in contact with agar medium, was considered undesirable for a short-term metabolism study which requires a rapid and uniform uptake of cytokinins. Accordingly, radiolabelled cytokinins were supplied at physiological concentrations to callus pieces in suspension. This procedure also minimises the generation of complex metabolic gradients within the tissue which would probably result from supplying cytokinins through agar medium, or in a small droplet on the callus surface. The cleavage of the N 6 side chain of cytokinins (particularly ZR and its O-glucoside) is the dominant form of cytokinin metabolism observed in soybean callus (Table 2). Hence a cytokinin oxidase system appears to operate in soybean callus tissue. The enhanced activity of ZR relative to Z noted above may be partly a consequence of greater susceptibility of Z to sidechain cleavage by cytokinin oxidase and indeed the radioactivity in adenine and its derivatives was 45-60% more when 3H-Z was supplied to soybean callus tissue in place of 3H-ZR (results not shown). This observation is consistent with published kinetic data on cytokinin oxidase from Vinca rosea crown-gall tissue (Scott et al. 1982). Zeatin-7-ghicoside, and to a lesser extent Z-9glucoside were very stable in soybean callus tissue, and Z was not detected as a metabolite. This finding is in accord with previous studies of the metabolism of 7- and 9-glucosides in radish cotyledons
248
L.M.S. Palni et al. : Biological activity and metabolism of cytokinins
and 7-glucosyl-BAP in tobacco cells (Letham and Palni 1983). However, Z was identified as a minor metabolite of LA in soybean callus tissue and this may account for the observed activity of LA. The biological activities of the four O-glucosides are similar to that of Z, particularly at high concentrations. In view of the large substituent at the N 6 position, these compounds are unlikely to be active per se (Letham 1978). Thus the high activity of these metabolites is most likely a consequence of their hydrolysis to the corresponding aglycones within the tissue. This view is supported by metabolism studies which indicate that Z-O-glucoside is hydrolyzed to Z in plant tissues (Palmer et al. 1981 ; Parker et al. 1978; van Staden and Papaphilippou 1977). In the present study, ZR was identified as a metabolite of ZR-O-glucoside by mass spectrometry (Fig. 4). Thus hydrolysis of an O-glucoside to the corresponding aglycone has now been shown unambiguously to occur within a plant tissue. Adenine, adenosine and adenine nucleotide(s) were also identified as metabolites of ZR-O-glucoside in soybean callus tissue. Since ZR-O-glucoside, like other O-glucosides, is resistant to cytokinin oxidase (Scott et al. 1982), the removal of the glucose moiety, possibly by the action of non-specific fl-glucosidases (Letham and Palni 1983), to yield ZR may well precede side-chain cleavage. In view of published data on cytokinin oxidase (Scott et al. 1982), adenine-9-glucoside, adenine-7-glucoside and adenine-9-alanine are potential metabolites of Z-9-glucoside, Z-7-glucoside and LA, respectively. Indeed, all three were identified, by comparison with authentic compounds, as minor metabolites in the soybean callus tissue. These are new cytokinin metabolites, except for the adenine7-glucoside which has been reported previously in mature cotyledons of intact radish seedlings (Letham et al. 1982). Thus the observed stability of the N-glucosides and LA may result from their compartmentation, and therefore protection against side-chain cleavage. In previous studies of soybean callus tissue, cytokinin conjugates have been isolated as major metabolites of exogenous bases. Thus 6-benzylaminopurine was converted mainly to 9-alanylbenzylaminopurine (Elliott and Thompson 1982) and Z to ZR and Z-O-glucoside (Horgan 1975; van Staden and Hutton 1982). 6-Benzylaminopurine is not a substrate for cytokinin oxidase (Scott et al. 1982) and this fact may account for the accumulation of 9-alanylbenzylaminopurine in soybean callus. However, the reported occurrence of ribosides and O-glucosides as major metabolites of Z is in contrast to our studies of Z and ZR metabolism in
which isoprenoid side-chain cleavage predominated. Limited evidence indicates that cytokinins are utilized actively in chloroplast development and that the concentrations required for this in tobacco callus cultures greatly exceed those needed for tissue growth (Kaminek et al. 1981). In soybean the growth response of Green callus was much less than that of Normal callus (Fig. 2), although the two lines did not exhibit any appreciable difference in ZR uptake and only a slight difference in degradative metabolism (Table 2). The observed differing responses of related callus lines (Fig. 2) to exogenous cytokinin will be of interest to those who use cytokinin bioassays, especially the relatively greater response from Root and Coty callus. However these differing responses to ZR cannot be rationalized in terms of altered cytokinin metabolism or uptake. Nevertheless, these parameters do provide an explanation for the differing activites of ZR, Z-7-glucoside, Z-9-glucoside, ZR-O-glucoside and LA, as discussed herein. The demonstration that ZR-O-glucoside slowly releases ZR in soybean tissue not only accounts for the activity of this glucoside, but also further establishes the view (Letham and Palni 1983) that ZR-O-glucoside and related glucosides are storage forms of Z and ZR. Similarly, the release of Z from LA observed in the present investigation offers a possible explanation of its weak activity, and also raises the possibility that this conjugate, of considerable stability in vivo, is a further storage form of cytokinin. We acknowledge gratefully the assistance of Mr. D.J. Pianca and Dr. R.E. Summons.
References Conrad, V.K. (1974) Ein sensibilisierter Amaranthus-Cytokinintest (AT74). Biochem. Physiol. Pflanz. 165, 531-535 Cowley, D.E., Duke, C.C., Liepa, A.J., MacLeod, J.K., Letham, D.S. (/978) The structure and synthesis of cytokinin metabolites. I. The 7- and 9-fl-o-glucofuranosides and pyranosides of zeatin and 6-benzylaminopurine. Aust. J. Chem. 31, 1095-1111 Davoll, J., Lythgoe, B., Todd, A.R. (1946) Experiments on the synthesis of purine nucleosides. XII. The configuration at the glycosidic centre in natural and synthetic pyrimidine and purine nucleosides. J. Chem. Soc. 2, 833-838 Duke, C.C., MacLeod, J.K., Summons, R.E., Letham, D.S., Paker, C.W. (1978) The structure and synthesis of cytokinin metabolites. II. Lupinic acid and O-fl-D-glucopyranosylzeatin from Lupinus angustifolius. Aust. J. Chem. 31, 1291-1301 Duke, C.C., Letham, D.S., Parker, C.W., MacLeod, J.K., Summons, R.E. (1979) The structure and synthesis of cytokinin metabolites. IV. The complex of O-glucosylzeatin derivatives formed in Populus species. Phytochemistry 18, 819-824 Elliott, D.C., Thompson, M.J. (1982) The identity of the major metabolite of benzylaminopurine in soybean cultures, and
L.M.S. Palni et al. : Biological activity and metabolism of cytokinins the inhibition of its formation by aminophylline. Plant Sci. Lett. 28, 29-38 Entsch, B., Parker, C.W., Letham, D.S., Summons, R.E. (1979) Preparation and characterisation, using high-performance liquid chromatography, of an enzyme forming glucosides of cytokinins. Biochim. Biophys. Acta 570, 124-139 Gordon, M.E., Letham, D.S., Parker, C.W. (1974) The metabolism and translocation of zeatin in intact radish seedlings. Ann. Bot. (London) 38, 809 825 Horgan, R. (1975) A new cytokinin metabolite. Biochem. Biophys. Res. Commun. 65, 358-363 Kaminek, M., Hada6ovS, V., Lugtinec, J. (1981) Plastids and the origin of cytokinin autonomy in tobacco tissue cultures. In: Metabolism and molecular activities of cytokinins, pp. 241-251, Guern, J., P6aud-Leno~l, C., eds. Springer, Berlin Heidelberg New York Letham, D.S., Young, H. (1971) The synthesis of radioisotopically labelled zeatin. Phytochemistry 10, 2077-2081 Letham, D.S. (1978) Cytokinins. In: Phytohormones and related compounds - a comprehensive treatise, vol. 1, pp. 205-263, Letham, D.S., Goodwin, P.B., Higgins, T.J.V., eds. Elsevier/North-Holland, Amsterdam Letham, D.S., Palni, L.M.S. (1983) The biosynthesis and metabolism of cytokinins. Annu. Rev. Plant Physiol. 34, 163-197 Letham, D.S., Palni, L.M.S., Tao, G.-Q., Gollnow, B.I., Bates, C. (1983) Regulators of cell division in plant tissues. XXIX. The activities of cytokinin glucosides and alanine conjugates in cytokinin bioassays. J. Plant Growth Regul. 2, 103-115 Letham, D.S., Tao, G.-Q., Parker, C.W. (1982) An overview of cytokinin metabolism. In: Plant growth substances 1982, pp. 143 153, Wareing, P.F., ed. Academic Press, London New York Miller, C.O. (1968) Naturally occurring cytokinins. In: Biochemistry and physiology of plant growth substances, pp. 33-45, Wightman, F., Setterfield, G., eds. Runge Press, Ottawa Newton, C., Morgan, C.B., Morgan, D.G. (1980) Evaluation of a bioassay for cytokinins using soybean hypocotyl sections. J. Exp. Bot. 31,721-729 Palmer, M.V., Scott, I.M., Horgan, R. (1981) Cytokinin metabolism in Phaseolus vulgaris L. II. Comparative metabolism of exogenous cytokinins by detached leaves. Plant Sci. Lett. 22, 187-195 Palni, L.M.S., Horgan, R. (1983) Cytokinin biosynthesis in crown gall tissue of Vinca rosea: metabolism of isopentenyladenine. Phytochemistry 22, 1597 1601 Palni, L.M.S., Summons, R.E., Letham, D.S. (1983) Mass spectrometric analysis of cytokinins in plant tissues. V. Identifi-
249
cation of the cytokinin complex of Datura innoxia crown gall tissue. Plant Physiol. 72, 858-863 Parker, C.W., Letham, D.S., Gollnow, B.I., Summons, R.E., Duke, C.C., MacLeod, J.K. (1978) Metabolism of zeatin by lupin seedlings. Planta 142, 239-251 Schmitz, R.Y., Skoog, F., Hecht, S.M., Bock, R.M., Leonard, N.J. (1972) Comparison of cytokinin activities of naturally occurring ribonucleosides and corresponding bases. Phytochemistry 11, 1603-16/0 Scott, I.M., McGaw, B.A., Horgan, R., Williams, P.E. (1982) Biochemical studies on cytokinins in Vinca rosea crown gall tissue. In: Plant growth substances 1982, pp. 165 174, Wareing, P.F., ed. Academic Press, London Shelton, K.R., Clark, J.M. (1967) A proton exchange between purines and water and its application to biochemistry. Biochemistry 6, 2735-2739 Summons, R.E., Duke, C.C., Eichholzer, J.V., Entsch, B., Letham, D.S., MacLeod, J.K., Parker, C.W. (1979) Mass spectrometric analysis of cytokinins in plant tissues. II. Quantitafion of cytokinins in Z e a m a y s kernels using deuterium labelled standards. Biomed. Mass Spectrom. 6, 407-413 Summons, R.E., Entsch, B., Letham, D.S., Gollnow, B.I., MacLeod, J.K. (1980) Metabolites of zeatin in sweetcorn kernels: purification and identifications using high-performance liquid chromatography and chemical-ionisation mass spectrometry. Planta 147, 422-434 Summons, R.E., Letham, D.S., Gollnow, B.I., Parker, C.W., Entsch, B., Johnson, L.P., MacLeod, J.K., Rolfe, B.G. (1981) Cytokinin translocation and metabolism in species of the leguminosae. In: Metabolism and molecular activites of cytokinins, pp. 56-65, Guern, J., P6aud-Leno~l, C., eds. Springer, Berlin Heidelberg New York Summons, R.E., Palni, L.M.S., Letham, D.S. (1983) Determination of intact zeatin nucleotide by direct chemical ionisation mass spectrometry. FEBS Lett. 151, 122-126 van Staden, J., Papaphilippou, A.P. (1977) Biological activity of O-/LD-glucopyranosylzeatin. Plant Physiol. 60, 649-650 van Staden, J., Hutton, M.J. (1982) Metabolism of [8-14C]zeatin in soybean callus. Z. Pflanzenphysiol. 106, 355-365 Wang, T.L. (1979) The sensitivity of soybean tissue cultures to the thymidine analogue, 5-bromodeoxyuridine. Plant Sci. Lett. 17, 123-128 Yeoman, M.M., MacLeod, A.J. (1977) Tissue (callus) cultures techniques. In: Plant tissue and cell culture, pp. 31-59, Street, H.E., ed. Blackwell, Oxford -
Received 18 July; accepted 2 November 1983
P l a n t a 9 Springer-Verlag 1984
The stability and biological activity of cytokinin metabolites in soybean callus tissue L.M.S. Palni, M.V. Palmer and D.S. Letham Research School of Biological Sciences, Australian National University, P.O. Box 475, Canberra City, ACT 2601, Australia
Abstract. The activity, uptake and metabolism of cytokinin metabolites was determined in soybean (Glycine max (L.) Merr.) callus tissue. The following activity sequence was established: zeatin riboside (ZR) > zeatin (Z) > O-glucosides of Z, ZR and their dihydro derivatives > lupinic acid (an alanine conjugate of Z ) > 7- and 9-glucosides of Z which were almost inactive. The 7- and 9-glucosides and lupinic acid were taken up very slowly by the callus tissue and showed great metabolic stability, but some degradation to 7-glucosyladenine, 9-glucosyladenine and the 9-alanine conjugate of adenine occurred. Compared with its aglycone, O-glucosylZR exhibited slow uptake and greatly enhanced stability but gas chromatographic-mass spectrometric analysis showed that appreciable amounts were hydrolyzed to ZR in the tissue. Both ZR and O-glucosyl-ZR were metabolised extensively, with adenine, adenosine, and adenine nucleotide(s) as the major metabolites. A diversity of minor metabolites of ZR were identified, including O-glucosides, lupinic acid and dihydrolupinic acid. The metabolism of ZR was suppressed by 3-isobutyl-1methylxanthine. When compared with the soybean callus line normally used for cytokinin bioassays (cv. Acme, cotyledonary callus), related callus lines exhibited greatly differing growth responses to cytokinin; however, these were not reflected in marked differences in metabolism. Key words: Cytokinin (activity, stability) - Glycine (cytokinin metabolism) - Tissue culture (cytokinin metabolism). Abbrevh~tions: GC-MS = gas chromatography-mass spectrome-
try; HPLC=high-performance liquid chromatography; L A = lupinic acid; OGZR = O-fl-D-glucopyranosylzeatin riboside; TLC = thin-layer chromatography; IMX = 3-isobutyl-l-methylxanthine, Z = zeatin; ZR = zeatin riboSide
Introduction
Bioassays based on plant tissue cultures are frequently used to assess the cytokinin activity of plant extracts. Although cytokinin glucosides occur commonly in such extracts and may even be the dominant cytokinins present (see review by Letham and Palni 1983), there is very little information regarding the activity of these glucosides in tissue-culture bioassays. Furthermore, the activity of lupinic acid (LA), a naturally occurring alanine conjugate of zeatin, has not been defined in such assays. The activity of a cytokinin in a bioassay must depend to some extent on the effectiveness of the cytokinin per se at the functional site, but it will also be influenced by uptake, metabolic stability and the activity of any metabolites formed. However, the uptake and metabolic stabilities of naturally occurring cytokinin glucosides and alanine conjugates have not been determined in any bioassay system. Accordingly, in the present study, nine naturally occurring cytokinins, including glucosides and LA, were tested for activity in the soybean-callus bioassay. In complementary experiments using the same tissue type (cotyledonary callus of soybean cv, Acme), the uptake and metabolism of radioactive analogues of five of these cytokinins was assessed, 20 h after their application. The observed metabolism of zeatin riboside (ZR) was also compared with that of ZR in two other soybean callus lines which differed markedly from the normal Acme callus in their response to supplied ZR. One of these was an unusual, chlorophyll-containing callus line derived from normal Acme callus. Materials and methods Chemicals. Zeatin (Z), isopentenyladenine (IP) and their corresponding 9-fl-D-ribosides (ZR and IPA, respectively) were ob-
L.M.S. Palni et al. : Biological activity and metabolism of cytokinins
243
tained commercially (Sigma Chemical Co., St. Louis, Mo., USA, and Calbiochem-Behring Corp., La Jolla, Calif., USA); all other cytokinins were synthesized chemically by methods reported previously (Cowley et al. 1978; Duke et al. 1978, 1979). The penta-deuterium-labelled O-fl-D-glucopyranosylzeatin riboside ([2Hs]OGZR) was prepared according to Summons et al. (1979). 7-/~-D-Glucopyranosyladenine (adenine-7glucoside; Cowley et al. 1978) and 9-~-D-glucopyranosyladenine (adenine-9-gtucoside; Davoll et al. 1946) were synthesized according to the cited references. ]~-(6-Aminopurin-9-yl)alanine, the adenine-9-alanine conjugate, was synthesized by heating 3-(6-ehloropurin-9-yl)-N-trifluoroaeetylalanine (Duke et al. 1978) with ammonia at 100~ C in a sealed tube, and the trifluoroacetyl group was cleaved with 1 N NaOH. [8-3H]Zeatin riboside (9.04 GBqmmol-1), [8-~H and 2Hs]OGZR (614MBq mmo1-1) and [8-3H] LA (981 MBq mmol -I) were obtained by heating respective non-radioactive compounds with 3HzO under conditions known to result in selective exchange of the C-8 hydrogen of purine ring (Shelton and Clark 1967). [2,83H]Zeatin was prepared as described by Letham and Young (1971), and used for the enzymic synthesis (Entsch et al. 1979) of its 7- and 9-fl-D-glucopyranosyl derivatives, Z-7-glucoside (3.26 GBq mmo1-1) and Z-9-glucoside (6.4 GBq mmol-1). These were diluted with unlabeUed compounds to give the specific activities cited. All radiolabelled cytokinins were stored in 50% aqueous ethanol at - 2 0 ~ C and purified by thin-layer chromatography (TLC) and-or high-performance liquid chromatography (HPLC) prior to use. The product identities were established by TLC, HPLC, ultraviolet (UV) and mass spectrometry. 3-Isobutyl-l-methylxanthine (IMX) was obtained from Sigma. Three dyes used as TLC markers were: Meldola Blue (Gurr, Searle Diagnostic, High Wycombe, UK), Drimarene Brilliant Blue K-BL (Polysciences, Warrington, Pa., USA), and Rhodamine B (Hopkins and Williams, Chadwell Heath, Essex, UK).
kept in the light. The results presented are the average of four flasks for each cytokinin concentration tested. Metabolic studies. Aqueous solutions of radioactive cytokinins were filter-sterilised and added to autoclaved basal medium (without 6-benzylaminopurine and agar) to give the following concentrations (gM): Z-9-glucoside and ZR-O-glucoside (0.5), Z-7-glucoside (0.55), ZR (0.6), LA (0.7). Callus tissue (20 g) was divided into small pieces and transferred aseptically to each conical flask containing liquid medium (50 nal) with added radioactive cytokinin. The cultures were incubated on a reciprocating shaker (50 rpm) at 26~ C in the dark, except for Green tissue which was incubated in the light. During incubation the callus pieces remained essentially intact. After 20 h the tissue was removed by filtration, washed with water (20 ml), and extracted as reported by Palni and Horgan (1983). The washings were combined with filtered medium, and an aliquot was withdrawn for radioactivity determination to assess the uptake of supplied cytokinin by callus tissue. The tissue extract was passed through a column of insoluble polyvinylpyrrolidone (Calbiochem, San Diego, Calif., USA) followed by cation-exchange chromatography on a column of cellulose phosphate, as detailed by Palni et al. (1983)9 This yielded a basic fraction (which would include cytokinin bases, glycosides, and alanine conjugates), and an acidic-neutral fraction containing cytokinin nucleotides. The nucleotide fraction was treated with Escherichia coli alkaline phosphatase (Sigma), followed by extraction with four equal volumes of water-saturated butan-l-ol (Palni and Horgan 1983). Cytokinin O-glucosides were hydrolysed to their respective aglycones using sweet-almond/?-glucosidase (Boehringer Biochemica GmbH, Mannheim, FRG) as reported by Palni and Horgan (1983). N-Glucosides of Z and adenine were hydrolysed by heating at 95 ~ C for 60 min in 0.5 N HC1. Radioactivity determinations were done according to Gordon et al. (1974) using an LKB 1215 Rackbeta II liquid scintillation counter (Wallac Oy, Turku, Finland).
Callus cultures and bioassay. The soybean (Glycine max (L.) Merr. cv. Acme) cotyledonary callus and the chlorophyllus line derived from it (see Wang 1979) were a gift of Dr. T.L. Wang (John Innes Institute, Norwich, UK), and have been maintained in our laboratory for the last 2.5 years. These two callus lines will be termed "Normal" and "Green", respectively, throughout this paper. Root and cotyledonary callus (further referred to as "Root" and "Coty") were initiated from soybean cv. Clark ex 63 (Murrumbidgee Irrigation Authority, Leeton, N.S.W., Australia) by the procedure of Yeoman and Macleod (1977). All callus lines used were cytokinin-dependentand were maintained on basal medium (Miller 1968) solidified with agar (0.8%, w/v), and supplemented with ~-naphthaleneacetic acid (2 mg 1-1) and 6-benzylaminopurine (1 mg 1-1). All cultures were grown at 26~ in the dark, except for the Green line which was grown in light as described by Wang (1979). Callus tissue for metabolic experiments and bioassay was used 24 d after subculture. Cytokinin bioassays were performed according to Miller (1968) using cotyledonary (Normal) callus. Three callus explants, approx. 30 mg each, were placed in each 50-ml Erlenmeyer flask containing 25 ml of medium. Cytokinin solutions were prepared in dimethylsulfoxide (DMSO; Mallinckrodt Chemical Works, St. Louis, Mo. USA) and their concentrations determined by UV absorbance; test solutions of cytokinins were added to autoclaved medium in individual flasks just before gelation. The concentration of DMSO in the final medium was kept below 0.2% (v/v) and DMSO was also added to the control flasks. Callus fresh weights were determined after growth for 24 d at 26~ C in dark, except the Green callus which was
Thin-layer chromatography. Small aliquots (2% by vol.) of the basic fraction and the nucleotide-derived, butanol-soluble fraction were analysed initially by two-dimensional TLC on 0.25-mm-thick PF254 Silica Gel 60 plates (Merck, Darmstadt, FRG), developed first in butan-l-ol: acetic acid:water (12:3:5, by vol.) (system 1), and then in butan-l-ol:14 M ammonia: water (6:1:2, by vol., upper phase) (system 2). This procedure separates isopentenyIadenine,its riboside, O-glucosides of ZR, dihydro-Z and dihydro-ZR, Z-9-glucoside, adenine and adenosine from one another and from the following pairs of compounds which are not resolved (see Table 1) : LA and dihydrolupinic acid; Z-O-glucoside and Z-7-glucoside; ZR and dihydroZR; Z and dihydro-Z. The two-dimensional procedure was used to obtain tentative identification of the major metabolites, estimates of their levels and of the distribution of radioactivity in each extract. Subsequently larger aliquots of extracts were purified by preparative TLC (system 1) alone, for tIPLC analysis. Three marker dyes (Summons et al. 1981) were employed to locate zones on the TLC plates (Table 1); these zones were then scraped off and eluted in 40-60% aqueous m~thanol containing 6% acetic acid. In this way, five groups oficompounds were readily obtained: LA and dihydro-LA (below Meldola Blue); O-glucosides of Z, ZR, their dihydro derivatives and Z-7-glucoside (zone above Meldola Blue); Z-9-glucoside, adenine and adenosine (Drimarene Brilliant Blue zone + zone below); Z, dihydro-Z, ZR and dihydro-ZR (zone between Drimarene Brilliant Blue and Rhodamine B); and isopentenyladenine and isopentenyladenosine (zone above Rhodamine B). Recovery of radioactive Z, ZR, ZR-O-glucoside, Z~9-glucoside and Z-7-glucoside from Silica-gel TLC plates was greater than 95% ; the recovery of LA was 75%.
9
244
L.M.S. Palni et al. : Biological activity and metabolism of cytokinins
Table 1. Resolution of various cytokinins and marker dyes on TLC and HPLC Compound
Lupinic acid (LA) Dihydro-LA Meldola Blue Z-7-glucoside Z-O-glucoside Dihydro-Z-O-glucoside ZR-O-glucoside Dihydro-ZR-O-glucoside Z-9-glucoside Adenosine Adenine Drimarene Brilliant Blue Zeatin (Z) Dihydro-Z Zeatin riboside (ZR) Dihydro-ZR Rhodamine B Isopentenyladenine Isopentenyladenosine
Rf value (TLC)
Elution time (HPLC, rain)
System 1
System 2
System A
System B
System C
0.25 0.25 0.27 0.32 0.31 0.32 0.32 0.34 0.39 0.42 0.42 0.44 0.48 0.48 0.49 0.49 0.55 0.60 0.60
0.18 0.18 . 0.18 0.18 0.22 0.13 0.15 0.19 0.30 0.50 . 0.58 0.58 0.36 0.36 . 0.65 0.45
-
28.9 -
17.4 17.4
7.2 7.2
8.5 17.4 12.5 8.5 -
14.7 16.7 14.7 -
-
-
-
-
.
.
.
. 18.6 28.9 32.7 22.0 25.0 18.6 5.5 7.5
.
. 12.0 14.6 5.5 7.5 .
.
. -
. .
System D
. .
.
.
-
High-performance liquid chromatography. The
HPLC equipment and conditions were the same as used by Summons et al. (1980). The solvents used were mixtures of methanol and water containing 0.2 M acetic acid, the proportion of methanol being given as percentage by volume. Radioactive fractions which had been prepared by TLC (system 1 ; see Table 1 and inset, Fig. 3) were then analysed by HPLC using one or more of the following systems: C 8 radial compression column (8 mm diameter, 100mm long; Waters Associates, Milford, Mass., USA) eluted with a linear gradient of 10% to 60% methanol (system A) or 5% to 40% methanol (system B) in 30 min, or isocratic elution with 20% methanol (system C) at 3 ml m i n - 1 ; pBondapak phenyl column (3.9 mm diameter, 300 mm long; Water Associates) eluted with a linear gradient of 0% to 10% methanol in 20 rain at 2 ml min- 1 (system D). In a few cases only two H P L C steps were sufficient to resolve completely a complex TLC eluate into its components (Fig. 3).
Mass spectrometry. Per-trimethylsilyl derivatives were prepared by heating the sample in 10 gl of pyridine (Ajax Chemicals, Sydney, N.S.W., Australia): N,O-bis(trimethylsilyl)trifluoroacetamide (Pierce Chemical Co., Rockford, Ill., USA) (1 : 1, v/v) at 90~ for 10 min. The sample was introduced via a direct inlet probe, using a Vespel extension (du Pont Co., Wilmington, Del., USA) fitted to a Finnigan-4500 instrument (Finnigan Corp., Sunnyvale, Calif., USA). The mass spectrometer was scanned from m/e 530 to 730 0.1 s-1 and the electron-impact spectra were integrated over the desorption profile. Other details of equipment and conditions used have been described in Palni et al. (1983) and Summons et al. (1983).
Results
Activities of cytokinins in soybean callus bioassays. T h e f o l l o w i n g a c t i v i t y s e q u e n c e f o r Z a n d its m e tabolites was established for the Normal soybean callus bioassay: ZR > Z > (O-glucosides of Z, ZR
l
.r zR
c
8
~//[? i i Z
3oo
/ u
o
. ~ ~
OGDZ
//
?oozR
c
c
,oo
U . " !.
>Z-9-glucoside > LA >>ZR-O-glucoside > ZR. In the extreme case, 85% of the Z-7-glucoside taken up by the callus was recovered as such in the extract. In all cases where ZR or its O-glucoside was supplied, adenine nucleotide(s), adenine and adenosine were the major metabolites identified, accounting for 22-40% of the radioactivity taken up (Table 2). Unidentified neutral compounds comprised up to 29% of the radioactivity originally
246
L.M.S. Palni et al. : Biological activity and metabolism of cytokinins
ZR
{TMsih-z,
~n ] |
~-
~/"t
mB
D ~ 8 RB
"
7n- -
N
05
"0
g
h5
50
4_Hj D5 M? C 629 644
{TMSi)5- ZR
o
0~,
o
Mt
120
RF
Mt-CH 3
IO-
701 Z
0
5
10
,
DZ
15
D5- Mt 716
639
20
25
Elution time (min)
Fig. 3. Purification of basic metabolites of [8-SH]ZR extracted from Normal soybean callus. Preparative TLC (System 1) was used to separate groups of metabolites (inset), which were then resolved by HPLC. In this case the TLC zone between marker dyes DBB and RB was eluted and subjected to a "bulk HPLC" step (Ca radial compression column, eluted isocratically at 3 ml min -1 with 35% methanol). The resulting '~ peak" was further analysed by HPLC (system A) to resolve ZR, dihydro-ZR (DZR), Z and dihydro-Z (DZ). Bars indicate the position of marker dyes (inset): MB, Meldola Blue; DBB, Drimarerie Brilliant Blue; RB, Rhodamine B
supplied as ZR or its O-glucoside. Isopentenyladenine, its riboside or the nucleotide(s) were not found as metabolites in any of the experiments. Minor metabolites of ZR (comprising less than 3% of the radioactivity taken up) identified in all cases were Z, dihydro-Z, dihydro-ZR, O-glucosides of Z, ZR and their dihydro derivatives, LA, dihydro-LA, and nucleotide(s) of Z and dihydro-Z. The free bases and ribosides (including those derived from nucleotides after treatment with alkaline phosphatase) were identified by co-chromatography with authentic standards on TLC (system 1 followed by system 2) and HPLC (system A; Fig. 3). The O-glucosides ofcytokinins were identified by TLC in the same way followed by HPLC (system B), and confirmation of their identities was also obtained by HPLC (system A) of the corresponding aglycones following hydrolysis by fl-glucosidase. Identification of LA and dihydro-LA was based initially on co-chromatography in two TLC systems (1 and 2) and two HPLC systems (C and D, in which these two metabolites are not resolved from each other), and final separation of the two compounds by TLC on cellulose, developed in solvent system 2 (Rf of LA: 0.29, and dihydro-LA: 0.35). The enhanced stability of ZR in IMXtreated tissue was not reflected in lower levels of any identified metabolite(s). A marked feature of ZR-O-glucoside metabo-
,,,r
~00
....
I
i ........
I,, hli h
J. . . . . .
650
,i
....
, ....
r./~
, ....
,rmJl)l,,,,i Ih,,
, ....
I "
' ....
'"
'
700
Fig. 4. Molecular-ion regions of the electron-impact mass spectrum of tetra- and penta-trimethylsilylderivatives of penta-deuterium (Ds)-labelled ZR, isolated as a metabolite of [8-3H and 2H5]ZR-O-glucoside in Normal soybean callus tissue
lism in Normal soybean callus was its hydrolysis to ZR, identified by co-chromatography with the authentic compound on TLC and HPLC, which accounted for 5.3% of radioactivity taken up. The radioactivity also corresponded to the Rf of t r a n s ZR on Silica-gel plates developed in chloroformmethanol (9:1). To confirm this observation [8-3H and 2Hs]OGZR was diluted with [2Hs]OGZR to half its specific activity and supplied at a concentration of 2.2 gM to 100 g of Normal tissue as described earlier. Tissue extraction and purification of metabolites was performed 20 h later and HPLC-purified ZR was subjected to GC-MS analysis as a mixture of its tetra- and penta-trimethylsilyl (TMSi) derivatives. In addition to the molecular ions (M +.) for (TMSi) 4- and (TMSi)s-ZR and the corresponding M .+-CH5 ions (all from the deuterated, i.e., D s compound), ions at m/e 639 and 624 (Fig. 4) were also seen. These ions correspond to M .+ and M.+-CH~ ions for the unlabelled compound and probably represent very low levels of endogenous ZR in the callus tissue. The N-glucosides of Z, Z-7-glucoside in particular, exhibited high metabolic stability in the tissue (Table 2). However, 2.8 and 7.4% of radioactivity taken up as Z-7-glucoside and Z-9-glucoside, respectively, was shown to co-chromatograph with adenine-7-glucoside and adenine-9-glucoside by TLC on Silica-gel (systems ] and 2) and cellulose plates, and by HPLC (see Table 3 for details). When Z-7-glucoside, Z-9-glucoside (recovered from the tissue) and adenine-9-glucoside were subjected to acid hydrolysis the radioactivity was found to co-chromatograph with the respective bases (Z or adenine) on TLC system 1. However,
L.M.S. Palni et al. : Biological activity and metabolism of cytokinins
247
Table 3. TLC and HPLC data for the new cytokinin metabolites. TLC systems 1 and 2 are described in Materials and methods.
TLC system 3 : 0 . 2 5 mm cellulose (Serva Feinbiochemica, Heidelberg, FRG) plates run in solvent system 2. HPLC was done using a C 8 radial compression column (8 x 100 ram), eluted isocratically with 5% methanol containing 0.2 M acetic acid at 3 ml m i n - 1. Values in parentheses are for the parent cytokinins : a, Z-7-glucoside; b, Z-9-glucoside; c, lupinic acid Metabolite
Adenine-7-glucoside Adenine-9-glucoside Adenine-9-alanine
Rf value (TLC)
HPLC elution time (rain)
System 1
System 2
System 3
0.35 0.41 0.13
0.11 0.14 0.08
0.11 (0.28)" 0.12 (0.32) b 0.04 (0.14) c
in the case of adenine-7-glucoside sufficient radioactivity was not available for this purpose. When LA was supplied to soybean tissue, 14.5 and 1.2% of radioactivity taken up was identified as adenine-9-alanine and Z, respectively (details to be published elsewhere). Some TLC and HPLC data for adenine-9-alanine are given in Table 3. Discussion
At present, the bioassay which is used most frequently for assessment of cytokinin levels in plant tissue extracts is the soybean callus assay developed by Miller (1968). However, of the commonly used cytokinin bioassays, this is the one for which cytokinin structure-activity relationships have been least defined. From the integrated approach described above, we are able firstly to supplement the available structure-activity information for this system with data for several additional naturally occurring cytokinins, and secondly to attempt to explain these relationships by comparison with parallel studies of uptake and metabolism. In contrast to the 7- and 9-glucosides of Z, the side-chain O-glucosyl conjugates of Z and ZR did not differ markedly from the respective aglycones in their biological activity. Similar results have been reported for the radish-cotyledon, the tobacco-callus and the Amaranthus-betacyanin bioassays (Letham et al. 1983). Side-chain saturation of the O-glucosides markedly reduces biological activity in the radish-cotyledon and Amaranthus-betacyanin bioassays, but no such effect was found in the soybean callus assay, nor previously in the tobacco callus assay (Letham et al. 1983). Relative to Z, LA is essentially inactive in the tobacco-callus and radish-cotyledon bioassays. Hence it is of considerable interest that LA exhibits appreciable activity in the soybean callus bioassay and that a similar relative activity has recently been reported for this metabolite in the Amaranthus bioassay (Letham et al. 1983). Another interesting feature of the soybean callus assay in common with
7.70 (27.23) a 5.38 (14.68) 8 10.35 (39.43) ~
the soybean hypocotyl test (Newton et al. 1980) is the fact that ZR exhibits greater activity than Z. In most other bioassays including the tobacco callus assay, the converse applies (Conrad 1974; Letham et al. 1983; Schmitz et al. 1972). The present studies further establish the usefulness of the soybean assay in that it detects the complex of ubiquitous O-glucoside metabolites of Z at low concentrations and exhibits a linear dose-response relationship for them over a wide concentration range. The use of large callus explants, in which only the basal cells are in contact with agar medium, was considered undesirable for a short-term metabolism study which requires a rapid and uniform uptake of cytokinins. Accordingly, radiolabelled cytokinins were supplied at physiological concentrations to callus pieces in suspension. This procedure also minimises the generation of complex metabolic gradients within the tissue which would probably result from supplying cytokinins through agar medium, or in a small droplet on the callus surface. The cleavage of the N 6 side chain of cytokinins (particularly ZR and its O-glucoside) is the dominant form of cytokinin metabolism observed in soybean callus (Table 2). Hence a cytokinin oxidase system appears to operate in soybean callus tissue. The enhanced activity of ZR relative to Z noted above may be partly a consequence of greater susceptibility of Z to sidechain cleavage by cytokinin oxidase and indeed the radioactivity in adenine and its derivatives was 45-60% more when 3H-Z was supplied to soybean callus tissue in place of 3H-ZR (results not shown). This observation is consistent with published kinetic data on cytokinin oxidase from Vinca rosea crown-gall tissue (Scott et al. 1982). Zeatin-7-ghicoside, and to a lesser extent Z-9glucoside were very stable in soybean callus tissue, and Z was not detected as a metabolite. This finding is in accord with previous studies of the metabolism of 7- and 9-glucosides in radish cotyledons
248
L.M.S. Palni et al. : Biological activity and metabolism of cytokinins
and 7-glucosyl-BAP in tobacco cells (Letham and Palni 1983). However, Z was identified as a minor metabolite of LA in soybean callus tissue and this may account for the observed activity of LA. The biological activities of the four O-glucosides are similar to that of Z, particularly at high concentrations. In view of the large substituent at the N 6 position, these compounds are unlikely to be active per se (Letham 1978). Thus the high activity of these metabolites is most likely a consequence of their hydrolysis to the corresponding aglycones within the tissue. This view is supported by metabolism studies which indicate that Z-O-glucoside is hydrolyzed to Z in plant tissues (Palmer et al. 1981 ; Parker et al. 1978; van Staden and Papaphilippou 1977). In the present study, ZR was identified as a metabolite of ZR-O-glucoside by mass spectrometry (Fig. 4). Thus hydrolysis of an O-glucoside to the corresponding aglycone has now been shown unambiguously to occur within a plant tissue. Adenine, adenosine and adenine nucleotide(s) were also identified as metabolites of ZR-O-glucoside in soybean callus tissue. Since ZR-O-glucoside, like other O-glucosides, is resistant to cytokinin oxidase (Scott et al. 1982), the removal of the glucose moiety, possibly by the action of non-specific fl-glucosidases (Letham and Palni 1983), to yield ZR may well precede side-chain cleavage. In view of published data on cytokinin oxidase (Scott et al. 1982), adenine-9-glucoside, adenine-7-glucoside and adenine-9-alanine are potential metabolites of Z-9-glucoside, Z-7-glucoside and LA, respectively. Indeed, all three were identified, by comparison with authentic compounds, as minor metabolites in the soybean callus tissue. These are new cytokinin metabolites, except for the adenine7-glucoside which has been reported previously in mature cotyledons of intact radish seedlings (Letham et al. 1982). Thus the observed stability of the N-glucosides and LA may result from their compartmentation, and therefore protection against side-chain cleavage. In previous studies of soybean callus tissue, cytokinin conjugates have been isolated as major metabolites of exogenous bases. Thus 6-benzylaminopurine was converted mainly to 9-alanylbenzylaminopurine (Elliott and Thompson 1982) and Z to ZR and Z-O-glucoside (Horgan 1975; van Staden and Hutton 1982). 6-Benzylaminopurine is not a substrate for cytokinin oxidase (Scott et al. 1982) and this fact may account for the accumulation of 9-alanylbenzylaminopurine in soybean callus. However, the reported occurrence of ribosides and O-glucosides as major metabolites of Z is in contrast to our studies of Z and ZR metabolism in
which isoprenoid side-chain cleavage predominated. Limited evidence indicates that cytokinins are utilized actively in chloroplast development and that the concentrations required for this in tobacco callus cultures greatly exceed those needed for tissue growth (Kaminek et al. 1981). In soybean the growth response of Green callus was much less than that of Normal callus (Fig. 2), although the two lines did not exhibit any appreciable difference in ZR uptake and only a slight difference in degradative metabolism (Table 2). The observed differing responses of related callus lines (Fig. 2) to exogenous cytokinin will be of interest to those who use cytokinin bioassays, especially the relatively greater response from Root and Coty callus. However these differing responses to ZR cannot be rationalized in terms of altered cytokinin metabolism or uptake. Nevertheless, these parameters do provide an explanation for the differing activites of ZR, Z-7-glucoside, Z-9-glucoside, ZR-O-glucoside and LA, as discussed herein. The demonstration that ZR-O-glucoside slowly releases ZR in soybean tissue not only accounts for the activity of this glucoside, but also further establishes the view (Letham and Palni 1983) that ZR-O-glucoside and related glucosides are storage forms of Z and ZR. Similarly, the release of Z from LA observed in the present investigation offers a possible explanation of its weak activity, and also raises the possibility that this conjugate, of considerable stability in vivo, is a further storage form of cytokinin. We acknowledge gratefully the assistance of Mr. D.J. Pianca and Dr. R.E. Summons.
References Conrad, V.K. (1974) Ein sensibilisierter Amaranthus-Cytokinintest (AT74). Biochem. Physiol. Pflanz. 165, 531-535 Cowley, D.E., Duke, C.C., Liepa, A.J., MacLeod, J.K., Letham, D.S. (/978) The structure and synthesis of cytokinin metabolites. I. The 7- and 9-fl-o-glucofuranosides and pyranosides of zeatin and 6-benzylaminopurine. Aust. J. Chem. 31, 1095-1111 Davoll, J., Lythgoe, B., Todd, A.R. (1946) Experiments on the synthesis of purine nucleosides. XII. The configuration at the glycosidic centre in natural and synthetic pyrimidine and purine nucleosides. J. Chem. Soc. 2, 833-838 Duke, C.C., MacLeod, J.K., Summons, R.E., Letham, D.S., Paker, C.W. (1978) The structure and synthesis of cytokinin metabolites. II. Lupinic acid and O-fl-D-glucopyranosylzeatin from Lupinus angustifolius. Aust. J. Chem. 31, 1291-1301 Duke, C.C., Letham, D.S., Parker, C.W., MacLeod, J.K., Summons, R.E. (1979) The structure and synthesis of cytokinin metabolites. IV. The complex of O-glucosylzeatin derivatives formed in Populus species. Phytochemistry 18, 819-824 Elliott, D.C., Thompson, M.J. (1982) The identity of the major metabolite of benzylaminopurine in soybean cultures, and
L.M.S. Palni et al. : Biological activity and metabolism of cytokinins the inhibition of its formation by aminophylline. Plant Sci. Lett. 28, 29-38 Entsch, B., Parker, C.W., Letham, D.S., Summons, R.E. (1979) Preparation and characterisation, using high-performance liquid chromatography, of an enzyme forming glucosides of cytokinins. Biochim. Biophys. Acta 570, 124-139 Gordon, M.E., Letham, D.S., Parker, C.W. (1974) The metabolism and translocation of zeatin in intact radish seedlings. Ann. Bot. (London) 38, 809 825 Horgan, R. (1975) A new cytokinin metabolite. Biochem. Biophys. Res. Commun. 65, 358-363 Kaminek, M., Hada6ovS, V., Lugtinec, J. (1981) Plastids and the origin of cytokinin autonomy in tobacco tissue cultures. In: Metabolism and molecular activities of cytokinins, pp. 241-251, Guern, J., P6aud-Leno~l, C., eds. Springer, Berlin Heidelberg New York Letham, D.S., Young, H. (1971) The synthesis of radioisotopically labelled zeatin. Phytochemistry 10, 2077-2081 Letham, D.S. (1978) Cytokinins. In: Phytohormones and related compounds - a comprehensive treatise, vol. 1, pp. 205-263, Letham, D.S., Goodwin, P.B., Higgins, T.J.V., eds. Elsevier/North-Holland, Amsterdam Letham, D.S., Palni, L.M.S. (1983) The biosynthesis and metabolism of cytokinins. Annu. Rev. Plant Physiol. 34, 163-197 Letham, D.S., Palni, L.M.S., Tao, G.-Q., Gollnow, B.I., Bates, C. (1983) Regulators of cell division in plant tissues. XXIX. The activities of cytokinin glucosides and alanine conjugates in cytokinin bioassays. J. Plant Growth Regul. 2, 103-115 Letham, D.S., Tao, G.-Q., Parker, C.W. (1982) An overview of cytokinin metabolism. In: Plant growth substances 1982, pp. 143 153, Wareing, P.F., ed. Academic Press, London New York Miller, C.O. (1968) Naturally occurring cytokinins. In: Biochemistry and physiology of plant growth substances, pp. 33-45, Wightman, F., Setterfield, G., eds. Runge Press, Ottawa Newton, C., Morgan, C.B., Morgan, D.G. (1980) Evaluation of a bioassay for cytokinins using soybean hypocotyl sections. J. Exp. Bot. 31,721-729 Palmer, M.V., Scott, I.M., Horgan, R. (1981) Cytokinin metabolism in Phaseolus vulgaris L. II. Comparative metabolism of exogenous cytokinins by detached leaves. Plant Sci. Lett. 22, 187-195 Palni, L.M.S., Horgan, R. (1983) Cytokinin biosynthesis in crown gall tissue of Vinca rosea: metabolism of isopentenyladenine. Phytochemistry 22, 1597 1601 Palni, L.M.S., Summons, R.E., Letham, D.S. (1983) Mass spectrometric analysis of cytokinins in plant tissues. V. Identifi-
249
cation of the cytokinin complex of Datura innoxia crown gall tissue. Plant Physiol. 72, 858-863 Parker, C.W., Letham, D.S., Gollnow, B.I., Summons, R.E., Duke, C.C., MacLeod, J.K. (1978) Metabolism of zeatin by lupin seedlings. Planta 142, 239-251 Schmitz, R.Y., Skoog, F., Hecht, S.M., Bock, R.M., Leonard, N.J. (1972) Comparison of cytokinin activities of naturally occurring ribonucleosides and corresponding bases. Phytochemistry 11, 1603-16/0 Scott, I.M., McGaw, B.A., Horgan, R., Williams, P.E. (1982) Biochemical studies on cytokinins in Vinca rosea crown gall tissue. In: Plant growth substances 1982, pp. 165 174, Wareing, P.F., ed. Academic Press, London Shelton, K.R., Clark, J.M. (1967) A proton exchange between purines and water and its application to biochemistry. Biochemistry 6, 2735-2739 Summons, R.E., Duke, C.C., Eichholzer, J.V., Entsch, B., Letham, D.S., MacLeod, J.K., Parker, C.W. (1979) Mass spectrometric analysis of cytokinins in plant tissues. II. Quantitafion of cytokinins in Z e a m a y s kernels using deuterium labelled standards. Biomed. Mass Spectrom. 6, 407-413 Summons, R.E., Entsch, B., Letham, D.S., Gollnow, B.I., MacLeod, J.K. (1980) Metabolites of zeatin in sweetcorn kernels: purification and identifications using high-performance liquid chromatography and chemical-ionisation mass spectrometry. Planta 147, 422-434 Summons, R.E., Letham, D.S., Gollnow, B.I., Parker, C.W., Entsch, B., Johnson, L.P., MacLeod, J.K., Rolfe, B.G. (1981) Cytokinin translocation and metabolism in species of the leguminosae. In: Metabolism and molecular activites of cytokinins, pp. 56-65, Guern, J., P6aud-Leno~l, C., eds. Springer, Berlin Heidelberg New York Summons, R.E., Palni, L.M.S., Letham, D.S. (1983) Determination of intact zeatin nucleotide by direct chemical ionisation mass spectrometry. FEBS Lett. 151, 122-126 van Staden, J., Papaphilippou, A.P. (1977) Biological activity of O-/LD-glucopyranosylzeatin. Plant Physiol. 60, 649-650 van Staden, J., Hutton, M.J. (1982) Metabolism of [8-14C]zeatin in soybean callus. Z. Pflanzenphysiol. 106, 355-365 Wang, T.L. (1979) The sensitivity of soybean tissue cultures to the thymidine analogue, 5-bromodeoxyuridine. Plant Sci. Lett. 17, 123-128 Yeoman, M.M., MacLeod, A.J. (1977) Tissue (callus) cultures techniques. In: Plant tissue and cell culture, pp. 31-59, Street, H.E., ed. Blackwell, Oxford -
Received 18 July; accepted 2 November 1983
