The elusive cytokinin biosynthetic pathway

Copyright © NISC Pty Ltd

South African Journal of Botany 2003, 69(3): 269–281 Printed in South Africa — All rights reserved

SOUTH AFRICAN JOURNAL OF BOTANY ISSN 0254–6299

Opinion Paper

The elusive cytokinin biosynthetic pathway NJ Taylor, WA Stirk and J van Staden* Research Centre for Plant Growth and Development, School of Botany and Zoology, University of Natal Pietermaritzburg, Private Bag X01, Scottsville 3209, South Africa * Corresponding author, e-mail: [email protected] Received 26 March 2003, accepted in revised form 14 April 2003

Although 40 years have passed since the first discovery of a naturally occurring cytokinin (CK) in plants, the biosynthetic pathway still remains unresolved. Numerous factors have contributed to the lack of success in this area, but most alarming has been the reliance upon unproven hypotheses and preconceived ideas. The greatest obstacle has been the identification and isolation of CK biosynthetic enzymes in plants, which have only recently been isolated and characterised in Arabidopsis. This has led to the classical isoprenoid CK biosynthetic pathway being partly challenged and the suggestion that multiple pathways might

exist in plants. CK biosynthetic mutants, which overproduce CKs as a result of IPT overexpression, are also starting to be identified and this should help overcome the past difficulties experienced in determining the site(s) of CK biosynthesis in plants. On the other hand, aromatic CK biosynthesis has seen no progress in the last few years and remains strictly speculative. This review aims to evaluate the current status of knowledge with respect to CK biosynthesis, to identify gaps in the knowledge and to suggest ways by which to move forward.

Abbreviations: ABA = abscisic acid, Ade = adenine, ADP = adenosine-5’-diphosphate, AMP = adenosine-5’-monophosphate, ATP = adenosine-5’-triphosphate, BA = benzyladenine, BA9G = benzyladenine-9-glucoside, BAR = benzyladenine riboside, CK = cytokinin, DHZ = dihydrozeatin, DHZMP = dihydrozeatin riboside-5’-monophosphate, DHZN9G = dihydrozeatin-N9-glucoside, DHZR = dihydrozeatin riboside, DMAPP = dimethylallyl pyrophosphate, DOXP = 1-deoxy-D-xylulose-5-phosphate, DXR = 1-deoxy-D-xylulose-5-phosphate reductoisomerase, GA = gibberellin, GA-3-P = gylceraldehyde-3-phosphate, HMGCoA = 3-hydroxy-methylglutaryl coenzyme A, HMGR = 3-hydroxy-methylglutaryl coenzyme A reductase, hoc = high shootorganogenic capacity, iP = isopentenyladenine, iPA = isopentenyladenosine, iPDP = isopentenyladenosine-5’-diphosphate, iPMP = isopentenyladenosine-5’-monophosphate, iPNG = isopentenyladenine-N-glucoside, IPP = isopentenyl pyrophosphate, ipt = isopentenyl transferase, iPTP = isopentenyladenosine-5’-triphosphate, MEP = methylerythritol 4-phosphate, 2MeS-oTOG = 2-methylthiopurine-ortho-topolin-O-glucoside, mT = meta-topolin, mTOG = meta-topolin-O-glucoside, mTROG = meta-topolin riboside-O-glucoside, MVA = mevalonic acid, oT = ortho-topolin, oTOG = ortho-topolin-O-glucoside, oTR = ortho-topolin riboside, pga = plant growth activator, PPFM = pink pigmented facultative methyltrophs, pTOG = para-topolin-Oglucoside, SHO = shooting, Z = zeatin, ZDP = trans-zeatin-riboside diphosphate, ZMP = trans-zeatin-riboside monophosphate, ZN7G = zeatin-N7-glucoside, ZN9G = zeatin-N9-glucoside, ZR = zeatin riboside, ZTP = trans-zeatin-riboside triphosphate

Introduction Hormone biosynthesis in general remains a stumbling block for plant scientists, with very few hormone biosynthetic pathways having being elucidated fully. Cytokinin (CK) biosynthesis remains firmly entrenched as a grey area, with little headway having been made with respect to the biosynthetic pathway in ‘normal’ (i.e. non-transformed) higher plants, since the first discovery of a naturally occurring CK in Zea

mays endosperm tissue in the 1960s (Miller 1961, Letham 1963). Numerous complicating factors have prohibited rapid advance in this area, which include the extremely low levels of endogenous CKs, the central role of the likely precursors (adenine and its nucleoside and nucleotide) in cellular metabolism, the existence of numerous (currently more than 40) native substances with more or less pronounced CK

270

activity (Letham and Palni 1983, Zañímalová et al. 1999) and most alarmingly, the reliance on incorrect or only partially correct assumptions concerning CK biosynthesis. Advances in modern techniques and instrumentation (such as LC-MS and GC-MS) have allowed the purification and identification of minute quantities of compounds and thus the difficulties experienced in the past have largely been resolved. With the advances in molecular biology and the completion of plant genomic sequences, the day should be rapidly approaching when we know the pathway(s) for CK biosynthesis in plants. Aspects yet to be proven remain points of intense debate and scrutiny. Naturally occurring CKs are adenine derivatives with at least one substituent at the N6-position and they are defined by their ability to promote cell division in cultured tissues in the presence of auxin (Miller et al. 1955, Skoog et al. 1965). According to the substituent at this position, CKs can be divided into three groups: (1) isoprenoid (zeatin (Z), N 6 -Δ2isopentenyladenine and its derivatives); (2) isoprenoidderived (dihydrozeatin (DHZ) and its derivatives); and (3) aromatic CKs (Zañímalová et al. 1999). This review will attempt to evaluate the current situation with regard to CK biosynthesis, to identify shortcomings in currently accepted theories and finally to suggest possible ways in which knowledge in this field can be furthered. In view of the differences between the isoprenoid and aromatic N6 side chains, it is unlikely that they have the same biosynthetic origin (Van Staden and Crouch 1996, Prinsen et al. 1997, Mok and Mok 2001) and they will therefore be considered separately. Biosynthesis of isoprenoid cytokinins Three different pathways for the biosynthesis of isoprenoid CKs in plants are currently proposed. Initially it was believed that tRNA was the main source of free CKs in plants (Swaminathan and Bock 1977). Subsequently, it was demonstrated that there was a direct route for CK biosynthesis that did not involve tRNA degradation (Taya et al. 1978). Finally, a second side chain precursor, other than isopentenyl pyrophosphate (IPP), was suggested such that trans-zeatin-riboside monophosphate (ZMP) synthesis can occur independently of isopentenyladenosine-5’-monophosphate (iPMP) (Åstot et al. 2000). Cytokinins in tRNA The first discovery of an N 6 -Δ2-isopentenyladenine (iP) residue in tRNA in the 1960s (Hall et al. 1966, Zachau et al. 1966) prompted the theory that CKs were derived from the degradation of tRNA. Modified bases, including isopentenylated bases, in tRNA have been found in virtually all organisms examined to date (Hall 1971, Murai 1994, Taller 1994). The biosynthesis of CKs associated with tRNA is relatively well understood, with the first step involving the post-transcriptional isopentenylation of Ade37, using IPP and unmodified tRNA as substrates. This reaction is catalysed by the enzyme Δ2-isopentenylpyrophosphate:tRNA-Δ2-isopentenyltransferase (EC 2.5.1.8) which has been partially purified from yeast (Kline et al. 1969), Escherichia coli (Bartz and Soll 1972) and maize (Holtz and Klämbt 1978). One of the

Taylor, Stirk and Van Staden

side chain methyl groups can be further modified by hydroxylation to yield the cis-isomer of Z, which is the major component of plant tRNA (Prinsen et al. 1997). The IPP for the biosynthesis of tRNA CKs is most probably derived from mevalonic acid (MVA), as Murai et al. (1975) demonstrated the incorporation of label from MVA into cis-zeatin riboside (cis-ZR) in tobacco callus tRNA. Although it has been suggested that CKs derived from tRNA could comprise up to 50% of the free CK pool in plants (Barnes et al. 1980, Maaβ and Klämbt 1981), it is unlikely that tRNA mediated CK biosynthesis is of great biological significance. Evidence to support the predominance of a direct route for CK biosynthesis is as follows: 1) tRNA in CK-dependent tissues contain CKs, yet an exogenous supply is still necessary for the growth of these tissues (Chen and Hall 1969, Burrows 1976). 2) The levels of free CKs in pea root tips exceed the amount present in tRNA by 27-fold (Short and Torrey 1972). It should be noted that the method these authors employed involved acid hydrolysis of tRNA, which destroys isopentenyladenosine (iPA) and thus the CK content in tRNA may have been underestimated (Letham and Palni 1983). 3) The low rates of tRNA turnover in plant tissues do not support the view that tRNA degradation contributes significantly to the pool of free CKs (Trewavas 1970, Hall 1973). 4) Certain free CKs in higher plants are not found in tRNA (e.g. aromatic CKs) and vice versa. cis-Z is the predominant CK in tRNA, but this CK exhibits little biological activity in the soybean callus bioassay (Van Staden and Drewes 1991) and the trans-isomer is the major free CK in higher plants (Mok and Mok 2001 and references therein). However, the existence of cis-trans-isomerase (Mok and Martin 1994) may support a function for tRNA as a supplementary source of free CKs in plants. This enzyme has been partially purified from Phaseolus vulgaris (Mok et al. 1992, Bassil et al. 1993). CK moieties in tRNA are always located at the strategic 37 position adjacent to the 3’-end anticodon (Sprinzl et al. 1989, 1991). This very specific location of CKs in tRNA, together with the distribution of tRNA species, indicates a specific role of CK moieties in tRNA function, which could possibly be proteosynthetic in nature, involving the enhancement of tRNA translational efficiency (Laten et al. 1978, Landick et al. 1990). In fact, Kamínek (1992) argued that plants have evolved efficient mechanisms to prevent tRNA-derived CKs from interfering with hormonal regulation. The reduced number of tRNA species containing CKs in higher plants and the predominance of the relatively inactive cis-isomer of Z in tRNA are believed to be two such mechanisms (Kamínek 1982). Due to the predominance of cis-isomers in some plant species and algae, their importance in plant growth and development has recently been questioned (Emery et al. 1998, 2000, Martin et al. 2001, Stirk et al. 2003 in review). Indications are that cis-isomers could be a transport form, have a regulatory role or could be required in an, as yet unidentified, developmental event (Martin et al. 2001). The finding of a cis-specific metabolic enzyme also suggests that tRNA degradation may not be the only source of cis-isomers

South African Journal of Botany 2003, 69(3): 269–281

and that there may be a direct pathway for the synthesis of cis-Z (Martin et al. 2001). Direct de novo CK synthesis Until recently, it was thought that the first step in the direct pathway of isoprenoid CK biosynthesis was the addition of dimethylallyl-pyrophosphate (DMAPP; an interconvertible isomer of IPP) to AMP to yield iPMP and the corresponding nucleoside (iPA; Figure 1). The enzyme catalysing this step in higher plants is believed to be Δ2-isopentenylpyrophosphate:5’-AMP-Δ2-isopentenyltransferase (EC 2.5.1.27; trivially named ‘cytokinin synthetase’ and under control of the ipt gene). Subsequently, the step-by-step conversion of iPMP to the base iP is expected to be catalysed by 5’nucleotidase (Chen and Kristopeit 1981a), followed by adenosine nucleosidase (Chen and Kristopeit 1981b). Z and ZR are subsequently formed from iP and iPA, respectively, by the rapid hydroxylation of one of the terminal side chain’s methyl groups, a reaction that is expected to involve a cytochrome P450 enzyme (Chen and Leisner 1984). CK biosynthesis gene Many aspects of this ‘classical’ pathway are still unresolved and rely extensively on circumstantial evidence. One of the most debatable areas remains the first step in CK biosynthesis and the existence of an ‘IPT ’ gene in higher plants. Most of our knowledge concerning this enzyme is derived from bacteria, as it was first discovered in the slime mold Dictyostelium discoideum (Taya et al. 1978) and has since also been found in Agrobacterium tumefaciens, where it is the product of the T-DNA gene 4 (ipt) (Akiyoshi et al. 1984, Barry et al. 1984, Morris et al. 1993). Consequently much of the work on CK biosynthesis in higher plants has, rather all to liberally in our opinion, made use of ipt-transformed plants (e.g. Stuchbury et al. 1979, Palni and Horgan 1983, Zhang et al. 1995, Åstot et al. 2000). Genes encoding an ipt have also been identified in other bacteria which include: ptz of Pseudomonas savastanoi (Powell and Morris 1986), ipt of Rhodococcus fascines (Crespi et al. 1992) and ipt (same as etz) of Erwinia harbicola (Lichter et al. 1995). In fact, due to the well-documented ability of bacteria to synthesise CKs and the dearth of information concerning a CK biosynthesising enzyme in plants at the time, Holland (1997) proposed that cytokinins are produced exclusively by external microbial symbionts of plants. This author identified PPFMs (pink pigmented facultative methyltrophs) as possible candidates for the synthesis of the majority of CKs in plants, due to their ubiquitous nature and high population numbers on all plants. The identification of a set of plant CK biosynthesis genes in Arabidopsis (designated AtIPT) (Kakimoto 2001, Takei et al. 2001) has negated the possibility that microbial symbionts are the sole source of CKs in higher plants. However, one cannot discount the possibility that rhizospheric bacteria may still make a contribution towards the endogenous levels of CKs in some plants (Kulaeva and Kusnetsov 2002, Sakakibara and Takei 2002), particularly considering the finding that plants can take up CKs from the soil (Arthur et al. 2001).

271

Activity of an IPT enzyme has been found in crude cellfree extracts from CK-autotrophic tobacco callus (Chen and Melitz 1979, Chen 1981, Chen 1982) and from immature maize kernels (Blackwell and Horgan 1994), but the enzyme was not purified to homogeneity in either of these two systems. Reasons for the lack of success in isolating this enzyme from plants have been attributed to the extreme instability of the enzyme and its low content in tissues (Chen and Melitz 1979, Prinsen et al. 1997). By approaching this problem from a novel angle, two separate laboratories in Japan have made great progress with respect to the identification and characterisation of this enzyme in plants. Through Arabidopsis database searches Kakimoto (2001) and Takei et al. (2001) have identified nine ipt homologs, designated AtIPT1 to AtIPT9. AtIPT2 and AtIPT9 resemble DMAPP:tRNA isopentenyltransferases and are distantly related to the AtIPT1, AtIPT3-8 group, which have greater homology to bacterial IPT (Figure 1). The involvement of the AtIPT1, AtIPT3 to AtIPT8 group of genes in CK biosynthesis was first demonstrated by Takei et al. (2001) who showed that E. coli transformants that express these genes excrete iP and Z into the culture media. Furthermore, Kakimoto (2001) showed that overexpression of AtIPT4 in Arabidopsis callus resulted in the regeneration of shoots in the absence of exogenously applied CK. However, these authors did not demonstrate whether AtIPT cDNAs could produce CKs in plants. More recently, using genetic approaches, mutants have been identified in Arabidopsis (Catterou et al. 2002, Sun et al. 2003) and Petunia hybrida (Zubko et al. 2002) that produce shoots from explants in the absence of exogenously applied CK. The petunia mutant SHO (shooting) (Zubko et al. 2002) and the Arabidopsis mutant PGA22 (plant growth activator) (Sun et al. 2003) phenotypes were both the result of elevated levels of iPA and its derivatives, particularly in root tissue. Molecular and genetic analysis identified PGA22 as the previously reported AtIPT8, which has IPT activity (Sun et al. 2003). Similarly, SHO was shown to have strong homology to the AtIPT genes from Arabidopsis (Zubko et al. 2002), suggesting that it too has IPT activity. This is an important finding as it is the first reported CK biosynthesis gene in a plant other than Arabidopsis and thereby provides some evidence for the universality of the IPT gene in plants. Expression of both the SHO and PGA22 in wild-type plants was very low, with the highest levels of expression detected in roots (Zubko et al. 2002, Sun et al. 2003), which is consistent with reports that CK biosynthesis occurs in roots (Van Staden and Davey 1979, Letham 1994, Davies 1995). However, SHO expression was also detected in other tissues, especially the leaves (Zubko et al. 2002). Unlike SHO and PGA22, the Arabidopsis mutant HOC (high shootorganogenic capacity) displayed elevated levels of most of the major CKs, although this was much more evident for Ztype CKs (Catterou et al. 2002). As yet, this gene has not been characterised, but it seems likely that it is not a biosynthetic gene, but rather a negative regulator of the CK biosynthetic pathway. The existence of an AtIPT gene family suggests that each gene may have a distinct physiological role, with expression varying spatially and temporally in response to environmen-

P

P

3

N

O

N9

7

N

ATP

P

P

P

OH

P

ATP

HO

O

NH2

P

P

N N9

7

OH

3

N O

N9

N

7

OH

CH2

1

?

ZTP ?

HO

O

N

1

O

N

CH2

iPTP

3

N

N

HO

O

1

N

C

C

P

HO

O O

N9

7

N

OH

ADP

3

N

NH2

CH3

CH2

CH3

CH3

OH

ADP

P

P

1

N 3

N

N

O

7

N9

N

OH

CH2

3

N O

N9

N

7

OH

CH2

C

O

CH 2

OH

C

P

P

MVA

Side chain precursor?

CH3

CH2

CH3

CH3

DMAPP

P

HO

N

N

O

7

N9

N

3

N

N

O

N9

N

7

OH

CH2

1

P

CH3

P

N 3

N O

NH2 N

7

N9

2

CH2 CH3

AMP

OH

HO

1

N

HO

CH2

N

1

3

N

N

O

7

N9

N

O

N9

N

7

OH

CH2

ZR

3

N

N

1

OH

CH2

iPA

HO

CH2

OH

HO

AMP

HO

O

1

metyrapone

CH3

IPT

C

C

IPT

OH

CH2

O

ZMP

HO

O

1

3

N

N

OH

iPMP

HO

O

1

3

CH

3

CH

MEP?

MVA MEP

C

P

ZDP ?

HO

O

N

1

N

1

iPDP

HO

O

?

P

P

AtlPT1, 3-8 PGA22, SHO

P

N

1

OH

OH

C

C

CH3

CH3

CH2

3

CH3

OH

3

N

N9

N

7

CH2

1

C

t-zeatin

N

1

N

7

N9

N

CH2

iP

3

N

NH N

1

C

CH3

CH2

CH3

CH3

OH

4

DMAPP

CH2

N9

C

c-zeatin

3

N

N

7

CH2

CH3

C

CH2

physiological significance?

N

1

N

mature tRNA

A A

OH

CH3

CH3

AtlPT2, 9

pre tRNA

A A

Figure 1: Proposed model for CK biosynthesis in plants (adapted from Kakimoto 2001, Sakakibara and Takei 2002). The block highlighted in grey indicates the model proposed by Kakimoto (2001), where CK biosynthesis is initiated by the addition of the isopentenyl side chain to ADP and ATP. The block highlighted in orange indicates the iPMP-independent pathway proposed by Åstot et al. (2000). The responsible enzymes (indicated by red numerals) are: 1) hydroxylase, 2) 5’-nucleotidase, 3) adenosine nucleosidase and 4) cis-trans-isomerase. The enzyme inhibitor, metyrapone, is indicated by a vertical bar. Question marks indicate proposed biosynthetic steps for which there is no experimental evidence

P

N

1

OH

Acetate/MVA

HO

O

272 Taylor, Stirk and Van Staden

South African Journal of Botany 2003, 69(3): 269–281

tal stimuli (Takei et al. 2001, Sakakibara and Takei 2002). Parallel pathways for CK biosynthesis may therefore operate in plants, just as for auxin biosynthesis (Slovin et al. 1999), and different pathways may operate in different species (Mok and Mok 2001). To date only the spatial expression of AtIPT8 and SHO has been studied (Zubko et al. 2002, Sun et al. 2003). By expanding this study to include the other eight AtIPT genes and different plant species much should be learnt about the sites of CK synthesis in the plant. Reaction products and substrates Even though genes for CK biosynthesis are beginning to be identified and characterised in plants (Kakimoto 2001, Takei et al. 2001, Zubko et al. 2002, Sun et al. 2003), the issue of the substrates used by the enzymes and the resulting reaction products remains unresolved. In fact, molecular analysis seems to have raised more questions about substrate specificity than it has answered. Kakimoto (2001) tested for enzyme substrate specificity in both crude and purified extracts from E. coli expressing the AtIPT4 gene. This author found that although the crude extract exhibited activity in the presence of AMP and DMAPP, the affinity-purified enzyme utilised ADP and ATP preferentially over AMP (Kakimoto 2001), indicating that AtIPT4 is a DMAPP:ATP/ADP isopentenyltransferase (Figure 1). Km values of AtIPT4 for ATP and DMAPP were estimated to be 18μM and 6.5μM, which is comparable to those of TZS (11.1μM for AMP and 8.2μM for DMAPP) (Morris et al. 1993). Initially, Takei et al. (2001) reported that the purified recombinant AtIPT1 from E. coli cells had DMAPP:AMP isopentenyltransferase activity and was inhibited in the presence of both ADP and ATP. As the Km values for AMP and DMAPP were very high, the properties of the enzyme were re-evaluated. It was subsequently conceded that AtIPT1 does indeed preferentially utilise ADP and ATP (Sakakibara and Takei 2002) and that the apparent inhibition caused by ATP and ADP in the IPT rapid assay of Blackwell and Horgan (1993) was probably the result of substrate competition between the radioisotope-labelled AMP and cold ATP or ADP. However, unlike AtIPT4, this enzyme was still able to use AMP as a substrate. The synthesis of CKs and the regulation of IPT activity is therefore linked to the energy status of the plant. This is of particular importance as many of the cellular events mediated by CKs (e.g. cell division) are energy-requiring processes and under optimal conditions the ATP:AMP ratio of a cell is approximately 100:1 (Hardie et al. 1998). Due to their instability, attempts to isolate other recombinant enzymes resulted in very low yields and thus their enzymatic properties could not be determined (Takei et al. 2001). The identification of the reaction products has also proved to be difficult, as the current procedure for the quantification of CK nucleotides involves an alkaline phosphatase treatment that removes phosphate groups, which interfere with identification by mass-spectrometry. This treatment indiscriminately cleaves phosphate groups and thus it has as yet not been possible to determine whether a mono-, dior tri-phosphate nucleotide (i.e. iPMP, isopentenyl-5’-diphosphate (iPDP) or isopentenyl-5’-triphosphate (iPTP)) is

273

formed as a result of AtIPT activity. However, it would appear that iPTP is formed as a result of the action of AtIPT4 on DMAPP and ATP (Kakimoto 2001). It is evident that before any progress can be made in this particular area of CK biosynthesis, methods need to be developed to quantify and identify these nucleotides. Induction of the PGA22 gene in Arabidopsis and the SHO gene in petunia resulted in an increase in iPMP and iPA levels in mutant plants as compared to wild type plants, with no apparent change in the levels of Z-type ribosides and ribotides or free bases (Zubko et al. 2002, Sun et al. 2003). However, Åstot et al. (2000) established that induction of the IPT gene from A. tumefaciens in Arabidopsis plants resulted in the accumulation of Z-type CKs, with no significant difference in the level of iP-type CKs. This difference in CK accumulation is most likely attributed to a difference in substrate specificity between the plant genes (Zubko et al. 2002, Sun et al. 2003) and the bacterial gene (Åstot et al. 2000) as these genes share only 12% homology (Sun et al. 2003). However, it also raises questions about the importance of hydroxylation in CK biosynthesis. The lack of iP-type CKs in tissues where CKs are actively being synthesised has been attributed to the rapid hydroxylation of one of the terminal side chain methyl groups to yield Z derivatives (Stuchbury et al. 1979, Palni and Horgan 1983, Chen and Leisner 1984). The general occurrence of such an enzyme is still uncertain, as reports confirming the incorporation of labelled iP or iPA into Z or ZR (Miura and Miller 1969, Miura and Hall 1973, Einset 1984) are matched by those unable to detect any significant incorporation of label (Mok et al. 1982, Capelle et al. 1983). It is therefore possible that hydroxylation may be species-specific, a supposition that is supported by the survey of Einset (1986). Further evidence to dispute the universality of hydroxylation is found in tzs-specific A. tumefaciens cells, where accumulation of Z in the culture medium was thought to be the result of rapid hydroxylation of iPMP/iPA/iP by a bacterial enzyme (Beaty et al. 1986, Heinemeyer et al. 1987). However, Agrobacterium, without the Ti plasmid, failed to demonstrate such activity as tRNA-derived iP was secreted into the medium. The possibility therefore exists that Z is not only derived from the hydroxylation of iP-type CKs, but it may be synthesised independently, implying that there could be several pathways for CK biosynthesis in plants. Far too many studies have relied on the work of Chen and Leisner (1984) and assumed that Z resulting from feeding experiments was a result of rapid hydroxylation of iP (e.g. Peterson and Miller 1976, Stuchbury et al. 1979, Burrows and Fuell 1981, Palni et al. 1983, Hocart and Letham 1990). The inclusion of a simple experiment, making use of an inhibitor of this reaction such as metyrapone (a specific inhibitor of cytochrome P450 enzymes), would have verified this assumption, but there have been few reports of such. In our opinion this has helped to perpetuate misconceptions regarding the CK biosynthetic pathway. Åstot et al. (2000) used elegant labelling studies to demonstrate the possible existence of an alternate pathway for CK biosynthesis, whereby ZMP is synthesised independently of iPMP (Figure 1). By simultaneously feeding 2H2O and [2H6]iPA they were able to demonstrate that the ZMP profile

274

Taylor, Stirk and Van Staden

was not dominated by [2H5] arising from [2H6]iPMP, but rather by a mass isotopomer cluster produced from in vivo deuterium labelling in a pathway not involving iPMP. The addition of metyrapone, which uncouples ZMP synthesis from iPMP, predictably resulted in a decline in ZMP levels in ipt-transgenic plants. However, deuterium-labelled ZMP was still detected under these conditions, indicating de novo synthesis of ZMP independent of iPMP. Furthermore, in the absence of metyrapone ZMP had a higher tracer/tracee ratio than iPMP and was synthesised 66 times faster, which is inconsistent with the view that iPMP is a precursor of ZMP, as a product cannot have a higher level of enrichment than its precursor (Åstot et al. 2000). Importantly, this iPMP-independent route of ZMP biosynthesis was also shown to be part of wild-type metabolism, as labelling of ZMP occurred in wild type tissue in the presence of metyrapone. However, in wild type plants both pathways appear to be important for plant function, with the relative contribution of each possibly varying in a tissue- and time-dependent manner. However, iPMP-dependent biosynthesis does seem to predominate in Arabidopsis and petunia plants in which the AtIPT8 and SHO genes have been overexpressed, as iP-type CKs increased, whilst there were only minor increases in Z-type ribosides and ribotides. However, the study of Åstot et al. (2000) failed to identify the side chain precursor that forms ZMP, which appears to be derived from the acetate/MVA pathway, with the most likely candidate being 4-pyrophosphate-2-methyltrans-but-2-enol. To complicate the issue further, Mok and Mok (2001) have suggested that CKs with specific side chain configurations may have separate origins, rather than having a common intermediate. In light of the findings of Åstot et al. (2000), this may warrant attention. Kakimoto (2001) has, however, hinted at the possibility that his findings and those of Åstot et al. (2000) need not be mutually exclusive and could be linked together to form an iPMP-independent pathway in which one of the methyl groups of iPTP and iPDP is hydroxylated to produce trans-zeatin riboside triphosphate (ZTP) and transzeatin riboside diphosphate (ZDP) respectively, followed by dephosphorylation to produce ZMP.

Biosynthesis of aromatic CKs Efforts to understand the biosynthesis of CKs have focussed entirely on isoprenoid CKs, with no attention given to aromatic CKs, aside from speculation. To date, nothing is known about the biosynthesis of aromatic CKs and no mediating enzymes have been identified (Mok and Mok 2001). In all likelihood this stems from the fact that for many years this group of CKs was considered to be synthetic. Aromatic CKs (both benzyladenine (BA) and topolin derivatives) have been identified in many plant tissues, including both macroalgae and axenic microalgae (Table 1). In our opinion this is an area of CK research that warrants attention as aromatic CKs clearly have important roles in plant growth and development. Although isoprenoid and aromatic CKs have overlapping spectra of biological activity it is apparent that they are not merely alternative forms of the same signal (Strnad 1997). Aromatic CKs appear to be involved in the regulation of metabolism and development of mature tissues rather than the stimulation of cell division (Kamínek et al. 1987) and they are also thought to be senescence-retarding factors (Strnad 1996). It has also been suggested that the levels of the two groups may be either directly or inversely linked, with an increase in isoprenoid CKs occurring as a consequence of elevated aromatic CKs rather than the result of direct hydrolysis of conjugates (Strnad 1997). The elucidation of the biosynthesis of this group of CKs is therefore fundamental to our understanding of the manner in which CKs regulate plant growth and development. Owing to the characteristic aromatic side chain, it is thought that the biosynthetic pathway would differ considerably from that of the isoprenoid CKs (Van Staden and Crouch 1996, Prinsen et al. 1997, Mok and Mok 2001). However, a potential ‘crossing point’ between the isoprenoid and aromatic side chain cannot be ignored (Zañimalová et al. 1999). Due to the nature of the benzylic side chain, it is supposed that a phenolic is a likely precursor (Strnad 1997). The amino acid phenylalanine (Figure 2) has been proposed as a starting compound with benzylaldehyde and/or the

Table 1: Some plant tissues containing endogenous aromatic cytokinins Plant tissue Mature poplar leaves Zantedeschia aethiopica fruits Old anise cell cultures Tomato crown gall tissue Poplar leaves Alfalfa Oil palm embryos, leaves and shoots Chenopodium rubrum (photoautotrophic cell suspension cultures) Macroalgae (31 species of Chlorophyta, Phaeophyta and Rhodophyta) Microalgae (9 species of Chlorophyta) Lettuce seeds

Type of cytokinin 6-(2-hydroxybenzylamino)purine-9-β-D-ribofuranoside 6-(2-hydroxybenzylamino)purine-9-β-D-ribofuranoside and its 2-methylthioglucofuranosyl derivative BAR BA and its glucoside derivatives Topolin derivatives oTR BA and topolin derivatives BA9G, oTOG, 2MeS-oTOG

Reference Horgan et al. 1973, 1975 Chaves das Neves and Pais 1980 Ernst et al. 1983 Nandi et al. 1989 Strnad et al. 1992, 1994, 1997 Goicoechea et al. 1995 Jones et al. 1996 Dolezal et al. 2002

oT, mTOG (main aromatics) and BA, mT, oTR, oTOG

Stirk et al. in review

BA, oT, mTOG (largest amounts), BAR, mT, mTROG, oTR, oTOG, pTOG BA, mT (main aromatics) and BAR, oT, mTOG, oTOG

Stirk et al. in review Unpublished

South African Journal of Botany 2003, 69(3): 269–281

NH

1

CH2

N

N

N 3

275

7

N9 benzyladenine

CH2

NH2

CH

phenylalanine COOH

Figure 2: An example of an aromatic CK, benzyladenine, and the potential side chain precursor

hydroxylated benzylaldehydes as immediate side chain precursors (Zañimalová et al. 1999). It is proposed, although not yet proven, that BA is the precursor for the topolins (Holub et al. 1998). Origin of the side chain for CK biosynthesis The origin of the isoprenoid side chain for CK biosynthesis appears to have been taken for granted, with very little conclusive evidence available to support the current theory that it originates exclusively from MVA. Although the deduced peptide sequence of bacterial ipt suggests it has a cytosolic origin, this is insufficient evidence to conclude that CK biosynthesis in plants is also cytosolic, as bacteria lack plastids. There are two biosynthetic routes for IPP/DMAPP biosynthesis in plants — the classical acetate/MVA pathway in the cytosol and the methylerythritol 4-phosphate (MEP) pathway in plastids (Figure 3). Sesquiterpenoids and sterols are predominantly synthesised in the cytosol via the acetate/MVA pathway, whereas monoterpenoids, diterpenoids, the phytol side-chain of chlorophyll, carotenoids, and the nonaprenyl side-chain of plastoquinone-9 are synthesised within plastids by the MEP pathway. There have been reports of exchange between the cytoplasmic and plastid IPP pools in plants (Arigoni et al. 1997, Nabeta et al. 1997, Estévez et al. 2000), but this cross-flow is likely to be limited and has been estimated to be less than 1% in intact plants under physiological conditions (Eisenreich et al.

2001). However, regulation of cross-flow may vary dramatically between species, cell types and/or the developmental stage of the plant (Rodríguez-Concepción and Boronat 2002). The first committed step in the acetate/MVA pathway is the irreversible reduction of 3-hydroxy-3-methylglutaryl coenzyme A (HMG-CoA), which is catalysed by the enzyme 3hydroxy-3-methylglutaryl coenzyme A reductase (HMGR) (Figure 3) (Goldstein and Brown 1990, Chappell 1995). This step is inhibited by mevastatin (compactin) and mevinolin (6α-methylcompacton, lovastatin) (Bach and Lichtenthaler 1982). In the MEP pathway IPP is formed via 1-deoxy-Dxylulose-5-phosphate (DOXP) from pyruvate and glyceraldehyde-3-phosphate (GA-3-P) (Lichtenthaler 1999). The source of GA-3-P lies in the Calvin Cycle, whilst pyruvate can either be derived from 3-phosphoglyceric acid, as a minor by-product of rubisco activity (Andrews and Kane 1991, Lichtenthaler 1999), or through uptake from the cytosol. The herbicide, fosmidomycin (an antibiotic produced by Streptomyces lavendulae), inhibits DOXP reductoisomerase (DXR) which catalyses the second step in the MEP pathway, i.e. the conversion of DOXP to MEP (Kuzuyama et al. 1998, Zeidler et al. 1998, Jomaa et al. 1999). This antibiotic can therefore be used to study the MEP pathway in a similar manner as mevastatin has been used in the mevalonate pathway. Although abscisic acid (ABA) and gibberellins (GA) were originally thought to originate in the acetate/MVA pathway, it is now apparent that they originate from the plastid-localised pathway of IPP biosynthesis (Milborrow and Lee 1998, Sponsel 2002 and references therein), although, unlike ABA, the evidence to support GA synthesis via this pathway is still circumstantial. That the side chain of isoprenoid CKs has a MVA origin has relied heavily on the evidence provided by inhibitory studies in tobacco BY-2 cell cultures (Crowell and Salaz 1992, Laureys et al. 1998, 1999, Miyazawa et al. 2002). These authors reported that the growth-retarding effect of the HMGR inhibitor, lovastatin, on tobacco cell cultures could be overcome by the addition of CKs, specifically Z, implying that CK biosynthesis was specifically inhibited at low concentrations of lovastatin. Measurement of CK levels in response to lovastatin-treatment confirmed that trans-Z and trans-ZR levels were indeed reduced (Laureys et al. 1998, 1999, Miyazawa et al. 2002). However, the levels of iPMP, iPA, iP, dihydrozeatin riboside (DHZR) and DHZ were not dramatically altered by this treatment (Laureys et al. 1998) and the levels of cis-Z and its riboside actually increased (Miyazawa et al. 2002). Significantly, the decrease in trans-Z and trans-ZR levels reported by Miyazawa et al. (2002) was not as pronounced as that reported by Laureys et al. (1998, 1999) and the addition of MVA in the presence of lovastatin could only restore Z and ZR levels to 10% of the control after 7h (Laureys et al. 1998). Miyazawa et al. (2002) attributed this small decrease in Z levels to a sub-lethal dose of lovastatin as these authors used 1μM whilst Laureys et al. (1998, 1999) used 10μM. These findings, taken together with the increase in cis-isomers in response to lovastatin, does raise the possibility that CKs may not be derived solely from the acetate/MVA pathway. Åstot et al. (2000) recently presented the most convinc-

276

Taylor, Stirk and Van Staden

Acetate/MVA pathway

MEP pathway

GA-3P + pyruvate DXS

Acetyl-CoA

CYTOPLASM

DXP

MEV

HMG-CoA

DXR

HMGR

MEP

MVA

FOSMIDO

Aromatic CKs?

PLASTID

CYTOKININS monoterpenes

DMAPP

FPP

IPP

?

polyprenols

IPP

IDI

chlorophyll plastoquinones

CKs?

DMAPP

GA

GGPP ABA

sequiterpenoides

Figure 3: The separation of the parallel pathways for isoprenoid biosynthesis in plant cells. The acetate/MVA pathway is localised in the cytoplasm in higher plants. The MEP pathway is localised in plastids. A box indicates the first intermediate specific to each pathway. Genes are encircled in blue. The red arrow indicates the postulated exchange of IPP between cytoplasmic and plastidic pools. Vertical bars denote enzyme inhibitors: FOSMIDO (fosmidomycin) and MEV (mevastatin)

ing evidence to support a mevalonate origin of the side chain for the iPMP-independent pathway, as in vivo deuterium labelling of ZMP was reduced in the presence of both metyrapone and mevastatin, as compared to the control which contained only metyrapone. However, the addition of two experiments to this study would have provided much clarity with respect to the biosynthesis of CKs. Firstly, mevastatin could have been fed in the absence of metyrapone. If no label was detected in iPMP, one could have concluded that in all likelihood the side chain precursor for the iPMP-dependent pathway was also derived from the acetate/MVA pathway.

Secondly, the use of an MEP pathway inhibitor (such as fosmidomycin) would have provided a clearer indication of the possibility of the side chain originating in this pathway. The distribution of the two pathways for IPP synthesis differs between organisms. Whilst the acetate/MVA pathway is utilised exclusively by archaebacteria, certain bacteria, yeasts, fungi, some protozoa and animals, the MEP pathway is used exclusively by bacteria and green algae. On the other hand, streptomycetes, some algae, mosses and liverworts, marine diatoms and higher plants appear to use both pathways (Eisenreich et al. 2001 and references therein). By

South African Journal of Botany 2003, 69(3): 269–281

comparing organisms that utilise different pathways for IPP synthesis it should be possible to gain some indication as to whether the side chain of CKs can be synthesised via the MEP pathway. Takei et al. (2001) and Sakakibara and Takei (2002) expressed the AtIPT gene family in E. coli and in the yeast Saccharomyces cerevisiae, and then quantified the levels of CKs in the culture medium. Whilst iP and Z accumulated in the culture medium of E. coli, only iP was detected in the culture medium of S. cerevisiae. This demonstrates that the AtIPT1 and AtIPT3-8 enzymes are capable of utilising IPP for the formation of CKs from both the acetate/MVA and MEP pathways. However, the fact that S. cerevisiae did not synthesise Z is also very interesting. The authors suggest two possible explanations for this anomaly. First, E. coli could contain a side chain donor, other than DMAPP, that results in the formation of ZMP or second, E. coli contains an enzyme capable of hydroxylase activity which will convert iP-type CKs to Z-type CKs. We believe that the first possibility deserves attention. Unfortunately, the S. cerevisiae data were only mentioned in a review article (Sakakibara and Takei 2002) and thus the full details of this experiment are eagerly awaited. The comparison of CK profiles of higher plants and algae (specifically Chlorophyta) has also revealed anomalies that could possibly be traced back to the origin of the side chain precursor, as isoprenoids are derived solely from the MEP pathway in Chlorophyta (Schwender et al. 2001). No transZR was detected in Chlorophyta and iP levels were significantly higher than trans-Z levels (Stirk et al. in review). A similar trend was observed in Chlamydomonas reinhardtii where iP constituted 90% of the total CKs quantified (Ivanova et al. 1992). cis-Isomers of Z also predominated over trans-isomers (Ördög et al. 2003 in review). This is contrary to the situation in higher plants where Z and its derivatives are the dominant CKs (Scott et al. 1980, Palmer et al. 1981) and trans-isomers predominate over cis-isomers (Letham and Palni 1983). The reasons for this divergence in the predominant CK species between higher plants and algae have not been addressed, as there have been very few investigations of CKs in algae. Most people would argue that this divergence in CK species was a result of reduced hydroxylase activity in algae, but this explanation seems too simplistic. We would therefore like to put forward the suggestion that this is a reflection of the different sources of IPP in algae and higher plants. In our estimation algae could prove to be an extremely useful system for studying CK biosynthesis. The study of Laureys et al. (1998) in tobacco BY-2 cell cultures also provides evidence for a dual origin of CKs in higher plants, as lovastatin, an inhibitor of the acetate/MVA pathway, effected a significant decrease in Z-type CKs only. The levels of iPMP, iPA and iP, although present in very low quantities, did not change significantly. Additionally, only Z could rescue lovastatin-induced inhibition of growth. Similarly, although Åstot et al. (2000) used ipt-transformed Arabidopsis plants, they found elevated ZMP, ZR and Z levels, but no significant change in the iPMP, iPA and iP pool sizes as a result of overexpression of this gene. Significantly, they were able to show that ZMP could be synthesised independently of iPMP.

277

One of the most vital pieces of evidence to support the possibility that CKs could be derived from the MEP pathway comes from the work of Davey and Van Staden (1981) and more recently Benková et al. (1999) who quantified CKs in chloroplasts. The whole spectrum of CKs have been found in chloroplasts isolated from tobacco leaves; these include free bases (iP), ribosides (iPA, ZR), ribotides (iPMP, ZMP and DHZMP) and N-glucosides (ZN9G, DHZN9G, ZN7G and iPNG). No Z or O-glucosides were detected. Whether these CKs are synthesised in the cytosol and then transported into the chloroplasts where they are interconverted or whether they are derived in situ in the chloroplast is still unknown. The presence of iPMP in chloroplasts is of particular interest, as at physiological pH this molecule is charged and does not easily penetrate membranes (Åstot et al. 2000). It must therefore be synthesised in the chloroplasts, but whether this is a result of de novo CK biosynthesis or interconversion awaits determination. Once again, in an organelle where the MEP pathway functions, iP-type CKs predominate over Ztype CKs. One manner by which the origin of the side chain could be resolved would be to overexpress key genes involved in the MEP and acetate/MVA pathways in callus tissue and then to assess if shoot formation occurs in the absence of exogenously applied CK (as achieved by Kakimoto (2001) and Sun et al. (2003)), and then to determine the CK profile by GC-MS. A similar strategy has been employed in Arabidopsis to ascertain if GA biosynthesis is plastidlocalised (Estévez et al. 2001). Alternatively, chloroplasts could be isolated and fed labelled pyruvate or GA-3-P and CKs analysed to assess if any have incorporated label. This could be confirmed by feeding label in the presence of fosmidomycin, which should result in a decrease in incorporation into CKs. Of utmost importance is the unequivocal demonstration of incorporation of label from the MEP or acetate/MVA pathways into CKs. As previously mentioned, nothing is known about the biosynthesis of aromatic CKs, particularly the nature of the side chain precursor. Although suggestions have been made there is no experimental evidence to support these suppositions. We would therefore like to propose that aromatic CKs may be derived from monoterpenoid precursors, which have a similar C6 ring structure. By definition, monoterpenoids are C10 compounds based on two isoprenoid (C5H10) units. Monoterpenoids are widely distributed in the plant kingdom, being found in higher plants as well as algae (Dev et al. 1982). They may also occur in any plant part but are most commonly distributed in the essential oils of plants (Mahmoud and Croteau 2001). Labelled monoterpenoids are incorporated into different metabolites such as monoterpenoids, higher terpenoids, amino acids and carbohydrates (Dev et al. 1982). Terpenoids undergo reactions such as cyclisation, hydroxylation and redox transformation (Mahmoud and Croteau 2001). If such a theoretical conversion of monoterpene precursors to aromatic CKs occurs in plants, then the topolins with the additional OH group are more likely to be the intermediary to BA and its derivatives. Monoterpenes are derived from the MEP pathway of IPP synthesis and are not formed irreversibly in plants. This could potentially represent the ‘crossing-point’ between the

278

isoprenoids and aromatic CK biosynthetic pathways alluded to by Zañimalová et al. (1999), as IPP/DMAPP would be the ultimate side chain donor in both instances. The biosynthetic pathways of isoprenoid and aromatic CKs could therefore be shown to be much more closely related than previously supposed. Conclusions and Future Prospects CK biosynthesis has been dogged by assumption and misconception since these compounds were first discovered in plants. As a result of the circumstances that led to their discovery and because of their impact on both cell division and protein synthesis, CKs were always thought to be closely associated with nucleic acids. For over a decade efforts focussed solely on the possibility that CKs were derived from tRNA and it was only in the late 1970s that a direct route of CK biosynthesis, independent of tRNA, was proposed. However, many of the original ideas concerning CK biosynthesis continued to plague this branch of CK research. Whilst significant advances have been made in the last three years, there is still much to be learnt, particularly concerning the nature of the substrates for the IPT enzyme. The continued use of genomics integrated with traditional biochemical and genetic approaches should allow the elucidation of the ‘elusive’ isoprenoid CK biosynthetic pathway. However, it must be emphasised that the use of any of these techniques alone will be insufficient and result in an incomplete picture of CK biosynthesis. Of prime importance is to establish the universality of the IPT gene in plants, which should become increasingly easier as more plant genomes are mapped, and by applying techniques similar to those used by Zubko et al. (2002) and Sun et al. (2003). Of equal importance is the determination of the origin of the side chain and the nature of the reaction products. By utilising the methods of Åstot et al. (2000) in plants where native genes are overexpressed, together with an inhibitor of the MEP pathway, this area of uncertainty should be resolved. The recently discovered predominance of Z cisisomers in certain plant species also begs explanation and alludes to the possibility that these cis-isomers may be synthesised via a direct route, independent of tRNA degradation. Finally, the most ignored area of CK biosynthesis deserves some attention, i.e. the biosynthesis of the aromatic CKs. These derivatives presumably play an important role in many plant species, yet research into this species of CK is minimal at the most. A starting point would be to examine the incorporation of label from previously proposed side chain precursors in cell free extracts, as was achieved with isoprenoid CKs, and possibly to identify mutants in which aromatic CKs are over-produced. Acknowledgements — We would like to thank the National Research Foundation and University of Natal Research Fund for financial assistance.

Taylor, Stirk and Van Staden

References Akiyoshi DE, Klee H, Amasino RM, Nester EW, Gordon MP (1984) T-DNA of Agrobacterium tumefaciens encodes an enzyme of cytokinin biosynthesis. Proceedings of the National Academy of Sciences of the USA 81: 5994–5998 Andrews TJ, Kane HJ (1991) Pyruvate is a by-product of catalysis by ribulosebisphosphate carboxylase/oxygenase. Journal of Biological Chemistry 266: 9447–9452 Arigoni D, Sagner S, Latzel C, Eisenreich W, Bacher A, Zenk MH (1997) Terpenoid biosynthesis from 1-deoxy-D-xylulose in higher plants by intramolecular skeletal rearrangement. Proceedings of the National Academy of Sciences of the USA 94: 10600–10605 Arthur GD, Jäger AK, Van Staden J (2001) Uptake of [3H]DHZ by tomato seedlings. South African Journal of Botany 67: 661–666 Åstot C, Dolezal K, Nordström A, Wang Q, Kunkel T, Moritz T, Chua N-H, Sandberg G (2000) An alternative cytokinin biosynthesis pathway. Proceedings of the National Academy of Sciences of the USA 97: 14778–14783 Bach TJ, Lichtenthaler HK (1982) Mevinolin, a highly specific inhibitor of microsomal 3-hydroxy-3-methyl-glutaryl-coenzyme A reductase of radish plants. Zeitschrift Naturforschung Teil C 37: 46–50 Barnes MF, Tien CL, Gray JS (1980) Biosynthesis of cytokinins by potato cell cultures. Phytochemistry 19: 409–412 Barry GF, Rogers SG, Fraley RT, Brand L (1984) Identification of a cloned cytokinin biosynthetic gene. Proceedings of the National Academy of Sciences of the USA 81: 4776–4780 Bartz L, Soll D (1972) N6-(Δ2-isopentenyl)adenosine: biosynthesis in vitro in transfer RNA by an enzyme purified from Escherichia coli. Biochemie 54: 31–39 Bassil NV, Mok DWS, Mok MC (1993) Partial purification of cistrans-isomerase of zeatin from immature seed of Phaseolus vulgaris L. Plant Physiology 102: 867–872 Beaty JS, Powell GK, Lica L, Regier DA, MacDonald EMS, Hommes NG, Morris RO (1986) Tzs, a nopaline Ti plasmid gene, from Agrobacterium tumefaciens associated with trans-zeatin biosynthesis. Molecular and General Genetics 203: 274–280 Benková E, Witters E, Van Dongen W, KoláÍ J, Motyka V, Brzobohatý B, Van Onckelen HA, Machá…ková I (1999) Cytokinins in tobacco and wheat chloroplasts. Occurrence and changes due to light/dark treatment. Plant Physiology 121: 245–251 Blackwell JR, Horgan R (1993) Cloned Agrobacterium tumefaciens ipt1 gene product, DMAPP:AMP isopentenyltransferase. Phytochemistry 34: 1477–1481 Blackwell JR, Horgan R (1994) Cytokinin biosynthesis by extracts of Zea mays. Phytochemistry 35: 339–342 Burrows WJ (1976) Mode of action of N, N’-diphenylurea: The isolation and identification of the cytokinins in the transfer RNA from tobacco callus grown in the presence of N, N’-diphenylurea. Planta 130: 313–316 Burrows WJ, Fuell KJ (1981) Cytokinin biosynthesis in cytokininautonomous and bacteria-transformed tobacco callus tissues. In: Guern J, Péaud-Lenoël C (eds) Metabolism and Molecular Activities of Cytokinins. Springer-Verlag, Berlin, pp 44–55 Capelle SC, Mok DWS, Kirchner SC, Mok MC (1983) Effects of thidiazuron on cytokinin autonomy and the metabolism of N6-(Δ2isopentenyl)[8-14C]adenosine in callus tissue of Phaseolus lunatus L. Plant Physiology 73: 796–802 Catterou M, Dubois F, Smets R, Vaniet S, Kichey T, Van Onckelen H, Sangwan-Norreel BS, Sangwan RS (2002) HOC: an Arabidopsis mutant overproducing cytokinins and expressing high in vitro organogenic capacity. The Plant Journal 30: 273–287 Chappell J (1995) The biochemistry and molecular biology of isoprenoid metabolism. Plant Physiology 107: 1–6

South African Journal of Botany 2003, 69(3): 269–281

Chaves das Neves HJ, Pais MSS (1980) A new cytokinin from the fruits of Zantedeschia aethiopica. Tetrahedron Letters 21: 4387–4390 Chen C-M (1981) Biosynthesis and enzymatic regulation of the interconversion of cytokinin. In: Guern J, Péaud-Lenoël C (eds) Metabolism and Molecular Activities of Cytokinins. SpringerVerlag, Berlin, pp 34–43 Chen C-M (1982) Cytokinin biosynthesis in cell-free systems. In: Wareing PF (ed) Plant Growth Substances 1982. Academic Press, London and New York, pp 155–163 Chen C-M, Hall RH (1969) Biosynthesis of N6-(Δ2-isopentenyl) adenosine in the transfer ribonucleic acid of cultured tobacco pith tissue. Phytochemistry 8: 1687–1695 Chen C-M, Kristopeit SM (1981a) Metabolism of cytokinins: dephosphorylation of cytokinin ribonucleotide by 5’-nucleotidase from wheat. Plant Physiology 67: 494–498 Chen C-M, Kristopeit SM (1981b) Metabolism of cytokinins: deribosylation of cytokinin ribonucleoside by adenine nucleotidases from wheat germ cells. Plant Physiology 68: 1020–1023 Chen C-M, Leisner SM (1984) Modification of cytokinins by cauliflower microsomal enzymes. Plant Physiology 75: 442–446 Chen C-M, Melitz DK (1979) Cytokinin biosynthesis in a cell-free system from cytokinin-autotrophic tobacco tissue cultures. FEBS Letters 107: 15–20 Crespi M, Messens E, Caplan AB, Van Montagu M, Desomer J (1992) Fasciation induction by the phytopathogen Rhodococcus fascians depends upon linear plasmid encoding a cytokinin synthase gene. EMBO Journal 11: 795–804 Crowell DR, Salaz MS (1992) Inhibition of growth of cultured tobacco cells at low concentrations of lovastatin is reversed by cytokinin. Plant Physiology 100: 2090–2095 Davey JE, Van Staden J (1981) Cytokinins in spinach chloroplasts. Annals of Botany 48: 243–246 Davies PJ (1995) Plant Hormones: Physiology, Biochemistry, and Molecular Biology. Kluwer Academic Press Dev S, Narula APS, Yadav JS (1982) CRC Handbook of Terpenoids: Monoterpenoids Vol 1. CRC Press Inc., Boca Raton, Florida Dolezal K, Åstot C, Hanuš J, Holub J, Peters W, Beck E, Strnad M, Sandberg G (2002) Identification of aromatic cytokinins in suspension cultured photoautotrophic cells of Chenopodium rubrum by capillary liquid chromatography/frit — fast atom bombardment mass spectrometry. Plant Growth Regulation 36: 181–189 Einset JW (1984) Conversion of N6-isopentenyladenine to zeatin by Actinidia tissues. Biochemical and Biophysical Research Communications 124: 470–474 Einset JW (1986) Zeatin biosynthesis from N6-(Δ2-isopentenyl)adenine in Actinidia and other woody plants. Proceedings of the National Academy of Sciences of the USA 83: 972–975 Eisenreich W, Rohdich F, Bacher A (2001) Deoxyxylulose phosphate pathway to terpenoids. Trends in Plant Science 6: 78–84 Emery RJN, Leport L, Barton JE, Turner NC, Atkins CA (1998) cisIsomers of cytokinins predominate Cicer arietinum throughout their development. Plant Physiology 117: 1515–1523 Emery RJN, Ma Q, Atkins CA (2000) The forms and sources of cytokinins in developing white lupine seeds and fruits. Plant Physiology 123: 1593–1604 Ernst D, Schäfer W, Oesterhelt D (1983) Isolation of a new naturally occurring cytokinin 6-benzylaminopurine riboside from an anise cell culture Pimpinella anisum L. Planta 195: 222–225 Estévez JM, Cantero A, Reindl A, Reichler S, León P (2001) 1Deoxy-D-xylulose-5-phosphate synthase, a limiting enzyme for plastidic isoprenoid biosynthesis in plants. The Journal of Biological Chemistry 276: 22901–22909 Estévez JM, Cantero A, Romero C, Kawaide H, Jiménez LF, Kuzuyama T, Seto H, Kamiya Y, León P (2000) Analysis of the expression of the CLA1, a gene that encodes the 1-deoxyxylu-

279

lose 5-phosphate synthase of the 2-C-Methyl-D-erythritol-4-phosphate pathway in Arabidopsis. Plant Physiology 124: 95–103 Goicoechea M, Dolezal K, Antolín MC, Strnad M, Sánchez-Diaz M (1995) Influence of mycorrhizae and Rhizobium of cytokinin content in drought-stressed alfalfa. Journal of Experimental Botany 46: 1543–1549 Goldstein JL, Brown MS (1990) Regulation of the mevalonate pathway. Nature 343: 425–430 Hall RH (1971) The Modified Nucleosides in Nucleic Acids. Columbia University Press, USA, pp 317–346 Hall RH (1973) Cytokinins as a probe of developmental processes. Annual Review of Plant Physiology 24: 415–444 Hall RH, Robins MJ, Stasiuk L, Thedford R (1966) Isolation of N6(γγ-dimethylallyl) adenosine from soluble ribonucleic acid. Journal of the American Chemical Society 88: 2614–2615 Hardie DG, Carling D, Carlson M (1998) The AMP-activated/SNF1 protein kinase subfamily: metabolic sensors of the eukaryotic cell. Annual Review of Biochemistry 67: 821–855 Heinemeyer W, Buchmann I, Tonge D, Windass JD, Alt-Moerbe J, Weiler EW, Botz T, Schröder J (1987) Two Agrobacterium tumefaciens genes for cytokinin biosynthesis: Ti plasmid-coded isopentenyltransferases adapted for function in prokaryotic or eukaryotic cells. Molecular and General Genetics 210: 156–164 Hocart CH, Letham DS (1990) Biosynthesis of cytokinin in germinating seeds of Zea mays. Journal of Experimental Botany 41: 1525–1528 Holland MA (1997) Occam’s razor applied to hormonology (Are cytokinins produced by plants?). Plant Physiology 115: 865–868 Holtz J, Klämbt D (1978) tRNA-isopentenyltransferase from Zea mays L. Characterization of the isopentenylation reaction of tRNA, oligo(A), and other nucleic acids. Hoppe-Seyler’s Zeitschrift fur Physiologische Chemie 259: 89–101 Holub J, Hanuš J, Hanke DE, Strnad M (1998) Biological activity of cytokinins derived from ortho- and meta-hydroxybenzyladenine. Plant Growth Regulation 26: 109–115 Horgan R, Hewett EW, Horgan JM, Purse JG, Wareing PF (1975) A new cytokinin from Populus x robusta. Phytochemistry 14: 1005–1008 Horgan R, Hewett EW, Purse JG, Wareing PF (1973) A new cytokinin from Populus robusta. Tetrahedron Letters 30: 2827–2828 Ivanova M, Todorov I, Pashankov P, Kostova L, Kamínek M (1992) Estimation of cytokinins in the unicellular green algae Chlamydomonas reinhardtii Dang. In: Kamínek M, Mok DWS, Zañímalová E (eds) Physiology and Biochemistry of Cytokinins in Plants. SPB Academic Publishing, The Hague, pp 483–485 Jomaa H, Weisner J, Sanderbrand S, Altincicek B, Weidermeyer C, Hintz M, Türbachova I, Eberl M, Zeidler J, Lichtenthaler H, Soldati D, Beck E (1999) Inhibitors of the nonmevalonate pathway of isoprenoid biosynthesis as antimalarial drugs. Science 285: 1573–1576 Jones LH, Martinková H, Strnad M, Hanke DE (1996) Occurrence of aromatic cytokinins in oil palm (Elaeis guineensis Jacq.). Journal of Plant Growth Regulation 15: 39–49 Kakimoto T (2001) Identification of plant cytokinin biosynthetic enzymes as dimethylallyl diphosphate:ATP/ADP isopentenyltransferase. Plant Cell Physiology 42: 677–685 Kamínek M (1982) Mechanisms preventing the interference of tRNA cytokinins in hormonal regulation. In: Wareing PF (ed) Plant Growth Substances 1982. Academic Press, London and New York, pp 215–224 Kamínek M (1992) Progress in cytokinin research. Tibtech 10: 159–164 Kamínek M, Vank T, Motyka V (1987) Cytokinin activities of N6benzyladenosine derivatives hydroxylated on the side-chain phenyl ring. Journal of Plant Growth Regulation 6: 113–120

280

Kline L, Fittler F, Hall R (1969) N6-(Δ2-isopentenyl)adenosine. Biosynthesis in transfer ribonucleic acid in vitro. Biochemistry 8: 4361–4371 Kulaeva ON, Kusnetsov VV (2002) Recent advances and horizons of the cytokinin studying. Russian Journal of Plant Physiology 49: 561–574 Kuzuyama T, Shimizu T, Takahashi S, Seto H (1998) Fosmidomycin, a specific inhibitor of 1-deoxy-D-xylulose 5-phosphate reductoisomerase in the nonmevalonate pathway for terpenoid biosynthesis. Tetrahedron Letters 39: 7913–7916 Landick R, Yanofski C, Choo K, Phung L (1990) Replacement of the Escerichia coli trp operon attenuation control codons alters operon expression. Journal of Molecular Biology 216: 25–37 Laten H, Gorman J, Bock RM (1978) Isopentenyladenosine deficient tRNA from an antisuppressor mutant of Saccharomyces cerevisiae. Nucleic Acid Research 5: 4329–4342 Laureys F, Dewitte W, Witters E, Van Montagu M, Inzé D, Van Onckelen H (1998) Zeatin is indispensable for the G2-M transition in tobacco BY-2 cells. FEBS Letters 426: 29–32 Laureys F, Smets R, Lenjou M, Van Bockstaele D, Inzé D, Van Onckelen H (1999) A low content in zeatin type cytokinins is not restrictive for the occurrence of G1/S transition in tobacco BY-2 cells. FEBS Letters 460: 123–128 Letham DS (1963) Zeatin: a factor inducing cell division from Zea mays. Life Science 8: 569–573 Letham DS (1994) Cytokinins as phytohormones: sites of biosynthesis, translocation and function of translocated cytokinin. In: Mok DWS, Mok MC (eds) Cytokinins, Activity and Function. CRC Press, Boca Raton, Florida, pp 57–80 Letham DS, Palni LMS (1983) The biosynthesis and metabolism of cytokinins. Annual Review of Plant Physiology 34: 163–197 Lichtenthaler HK (1999) The 1-deoxy-D-xylulose-5-phosphate pathway of isoprenoid biosynthesis in plants. Annual Review of Plant Physiology and Plant Molecular Biology 50: 47–65 Lichter A, Barash I, Valinsky L, Manulis S (1995) The genes involved in cytokinin biosynthesis in Erwinia herbicola pv. gypsophilae: characterization and role in gall formation. Journal Bacteriology 177: 4457–4465 Maaβ H, Klämbt D (1981) On the biogenesis of cytokinins in roots of Phaseolus vulgaris. Planta 151: 353–358 Mahmoud SS, Croteau RB (2001) Metabolic engineering of essential oil yield and composition in mint by altering expression of deoxyxylulose phosphate reductoisomerase and menthofuran synthase. Proceedings of the National Academy of Sciences of the USA 98: 8915–8920 Martin RC, Mok MC, Habben JE, Mok DWS (2001) A maize cytokinin gene encoding an O-glucosyltransferase specific to ciszeatin. Proceedings of the National Academy of Sciences of the USA 98: 5922–5926 Milborrow BV, Lee H-S (1998) Endogenous biosynthetic precursors of (+)-abscisic acid. VI. Carotenoids and ABA are formed by the ‘non-mevalonate’ triose-phosphate pathway in chloroplasts. Australian Journal of Plant Physiology 25: 507–512 Miller CO (1961) A kinetin-like compound in maize. Proceedings of the National Academy of Sciences of the USA 47: 170 Miller CO, Skoog F, Von Saltza MH, Strong FM (1955) Kinetin, a cell division factor from deoxyribonucleic acid. Journal of the American Chemical Society 78: 1329–1334 Miura GA, Hall RH (1973) trans-Ribosylzeatin. Its biosynthesis in Zea mays endosperm and the mycorrhizal fungus, Rhizopogon roseolus. Plant Physiology 51: 563–569 Miura GA, Miller CO (1969) 6-(γ,γ-Dimethylallylamino)purine as a precursor of zeatin. Plant Physiology 44: 372–376 Miyazawa Y, Kato H, Muranaka T, Yoshida S (2002) Amyloplast formation in cultured tobacco BY-2 cells requires a high cytokinin content. Plant and Cell Physiology 43: 1534–1541

Taylor, Stirk and Van Staden

Mok DWS, Mok MC (2001) Cytokinin metabolism and action. Annual Review of Plant Physiology and Plant Molecular Biology 52: 89–118 Mok MC, Martin RC (1994) Cytokinin metabolic enzymes. In: Mok DWS, Mok MC (eds) Cytokinins — Chemistry, Activity and Function. CRC Press, Boca Raton, Florida, pp 129–137 Mok MC, Martin RC, Mok DWS, Shaw G (1992) Cytokinin activity, metabolism and function in Phaseolus. In: Kamínek M, Mok DWS, Zañímalová E (eds) Physiology and Biochemistry of Cytokinins in Plants. SPB Academic Publishing, The Hague, pp 41–46 Mok MC, Mok DWS, Dixon SC, Armstrong DJ, Shaw G (1982) Cytokinin structure-activity relationships and the metabolism of N6-(Δ2-isopentenyl)adenosine-8-14C in Phaseolus callus tissues. Plant Physiology 70: 173–178 Morris RO, Blevins DG, Dietrich JT, Durley RC, Gelvin SB, Gray J, Hommes NG, Kamínek M, Mathews LJ, Meilan R, Reinbott TM, Sayavedra-Sato L (1993) Cytokinins in plant pathogenic bacteria and developing cereal grains. Australian Journal of Plant Physiology 20: 621–637 Murai N (1994) Cytokinin biosynthesis in tRNA and cytokinin incorporation into plant tRNA. In: Mok DWS, Mok MC (eds) Cytokinins: Chemistry, Activity, and Function. CRC Press, Boca Raton, Florida, pp 87–99 Murai N, Armstrong DJ, Skoog F (1975) Incorporation of mevalonic acid into ribosylzeatin in tobacco callus ribonucleic acid preparations. Plant Physiology 55: 853–858 Nabeta K, Kawae T, Saitoh T, Kikuchi T (1997) Synthesis of chlorophyll and carotene from 2H and 13C-labeled glycin in cultured cells of liverworts Heteroscyphus planus and Lophocolea heterophylla. Journal of the Chemical Society Perkin-Transactions 1: 261–267 Nandi SK, Letham DS, Palni LMS, Wong OC, Summons RE (1989) 6-Benzylaminopurine and its glycosides as naturally occurring cytokinins. Plant Science 61: 189–196 Ördög V, Stirk WA, Strnad M, Van Staden J (2003) Endogenous cytokinins in microalgae (Chlorophyta). Journal of Phycology, in review Palmer MV, Horgan R, Wareing PF (1981) Cytokinin metabolism in Phaseolus vulgaris L. III: Identification of endogenous cytokinins and metabolism of [8-14C]-dihydrozeatin in stems of decapitated plants. Planta 153: 297–302 Palni LMS, Horgan R (1983) Cytokinin biosynthesis in crown gall tissue of Vinca rosea: Metabolism of isopentenyladenine. Phytochemistry 22: 1597–1601 Palni LMS, Horgan R, Darrall NM, Stuchbury T, Wareing PF (1983) Cytokinin biosynthesis in crown gall tissue of Vinca rosea: the significance of nucleotides. Planta 159: 50–59 Peterson JB, Miller CO (1976) Cytokinins in Vinca rosea L. crown gall tumor tissue as influenced by compounds containing reduced nitrogen. Plant Physiology 57: 393–399 Powell GK, Morris RO (1986) Nucleotide sequence and expression of a Pseudomonas savastanoi cytokinin biosynthetic gene: homology with Agrobacterium tumefaciens tmr and tzs loci. Nucleic Acids Research 14: 2555–2565 Prinsen E, Kamínek M, Van Onckelen HA (1997) Cytokinin biosynthesis: a black box? Plant Growth Regulation 23: 3–15 Rodríguez-Concepción M, Boronat A (2002) Elucidation of the methylerythritol phosphate pathway for isoprenoid biosynthesis in bacteria and plastids. A metabolic milestone achieved through genomics. Plant Physiology 130: 1079–1089 Sakakibara H, Takei K (2002) Identification of cytokinin biosynthesis genes in Arabidopsis: A breakthrough for understanding the metabolic pathway and the regulation in higher plants. Journal of Plant Growth Regulation 21: 17–23 Schwender J, Gemünden C, Lichtenthaler HK (2001) Chlorophyta exclusively use the 1-deoxyxylulose 5-phosphate/2-C-methylery-

South African Journal of Botany 2003, 69(3): 269–281

thritol 4-phosphate pathway for the biosynthesis of isoprenoids. Planta 212: 416–423 Scott IM, Browning G, Eagles J (1980) Ribosylzeatin and zeatin in tobacco crown gall tissue. Planta 147: 269–273 Short KC, Torrey JG (1972) Cytokinins in seedling roots of pea. Plant Physiology 49: 155–160 Skoog F, Strong FM, Miller CO (1965) Cytokinins. Science 148: 532–533 Slovin JP, Bandurski RS, Cohen JD (1999) Auxin. In: Hooykaas PJJ, Hall MA and Libbenga KR (eds) Biochemistry and Molecular Biology of Plant Hormones. Elsevier Sciences, Amsterdam, The Netherlands, pp 35–42 Sprinzl M, Dank N, Nock S, Schön A (1991) Compilation of tRNA sequences and tRNA genes. Nucleic Acid Research 19: 2127–2171 Sprinzl M, Hartman T, Weber J, Blank J, Zeidler R (1989) Compilation of tRNA sequences and sequences of tRNA genes. Nucleic Acid Research 17(Suppl): r1–r172 Sponsel VM (2002) The deoxyxylulose phosphate pathway for the biosynthesis of plastidic isoprenoids: Early days in our understanding of the stages of gibberellin biosynthesis. Journal of Plant Growth Regulation 20: 332–345 Stirk WA, Novák O, Strnad M, Van Staden J (2003) Cytokinins in macroalgae. Functional Biological Sciences, in review Strnad M (1996) Enzyme immunoassays of N6-(meta-hydroxybenzyl)adenine cytokinins. Journal of Plant Growth Regulation 15: 179–188 Strnad M (1997) The aromatic cytokinins. Physiologia Plantarum 101: 674–688 Strnad M, Hanuš J, Vank T, Kamínek M, Ballantine JA, Fussell B, Hanke DE (1997) Meta-topolin, a highly active aromatic cytokinin from Poplar leaves (Populus x canadensis Moench cv. Robusta). Phytochemistry 45: 213–318 Strnad M, Peters W, Beck E, Kamínek M (1992) Immunodetection and identification of N6-(o-Hydroxybenzylamino) purine as a naturally occurring cytokinin in Populus x canadensis Moench cv. Robusta leaves. Plant Physiology 99: 74–80 Strnad M, Peters W, Hanuš J, Beck E (1994) Ortho-topolin-9-glucoside, and aromatic cytokinin from Populus x canadensis cv. Robusta leaves. Phytochemistry 37: 1059–1062 Stuchbury T, Palni LM, Horgan R, Wareing PF (1979) The biosynthesis of cytokinins in crown-gall tissue of Vinca rosea. Planta 147: 97–102 Sun J, Niu Q-W, Tarkowski P, Zheng B, Tarkowska D, Sandberg G, Chua N-H, Zuo J (2003) The Arabidopsis AtIPT8/PGA22 gene encodes an isopentenyl transferase that is involved in de novo cytokinin biosynthesis. Plant Physiology 131: 167–176

Edited by CH Bornman

281

Swaminathan S, Bock RM (1977) Isolation and identification of cytokinins from Euglena gracilis transfer ribonucleic acid. Biochemistry 16: 1355–1360 Takei K, Sakakibara H, Sugiyama T (2001) Identification of genes encoding adenylate isopentenyltransferase, a cytokinin biosynthesis enzyme, in Arabidopsis thaliana. Journal of Biological Chemistry 276: 26405–26410 Taller BJ (1994) Distribution, biosynthesis, and function of cytokinins in tRNA. In: Mok DWS, Mok MC (eds) Cytokinins: Chemistry, Activity, and Function. CRC Press, Boca Raton, Florida, pp 101–112 Taya T, Tanaka Y, Nishimura S (1978) 5’-AMP is a direct precursor of cytokinin in Dictyostelium discoideum. Nature 271: 545–547 Trewavas A (1970) The turnover of nucleic acids in Lemna minor. Plant Physiology 45: 742–751 Van Staden J, Crouch NR (1996) Benzyladenine and derivatives — their significance and interconversion in plants. Plant Growth Regulation 19: 153–175 Van Staden J, Davey JE (1979) The synthesis, transport and metabolism of endogenous cytokinins. Plant, Cell and Environment 2: 93–106 Van Staden J, Drewes FE (1991) The biological activity of cytokinin derivatives in the soybean callus bioassay. Plant Growth Regulation 10: 109–115 Zachau HG, Dutting D, Feldman H (1966) Nucleotide sequences of two serine-specific transfer ribonucleic acids. Angewandte Chemie International Edition 5: 422 Zañímalová E, Kamínek M, BÍezinová, Motyka V (1999) Control of cytokinin biosynthesis and metabolism In: Hooykaas PJJ, Hall MA, Libbenga KR (eds) Biochemistry and Molecular Biology of Plant Hormones. Elsevier Sciences, pp 141–160 Zeidler J, Schwender J, Müller C, Wiesner J, Weidemeyer C, Beck E, Jomaa H, Lichtenthaler HK (1998) Inhibition of the non-mevalonate 1-deoxy-D-xylulose-5-phosphate pathway of plant isoprenoid biosynthesis by fosmidomycin. Zeitschrift Naturforschung 53c: 980–986 Zhang R, Zhang X, Wang J, Letham DS, McKinney SA, Higgins TJV (1995) The effect of auxin on cytokinin levels and metabolism in transgenic tobacco tissue expressing an ipt gene. Planta 196: 84–94 Zubko E, Adams CJ, Macháèková I, Malbeck J, Scollan C, Meyer P (2002) Activation tagging identifies a gene from Petunia hybrida responsible for the production of active cytokinins in plants. The Plant Journal 29: 797–808