Interactions between sterol biosynthesis genes ... - Wiley Online Library

The Plant Journal (2002) 31(1), 61±73

Interactions between sterol biosynthesis genes in embryonic development of Arabidopsis Kathrin Schrick1,* Ulrike Mayer1, Gottfried Martin1, Catherine Bellini2, Christine Kuhnt3, JuÈrgen Schmidt3 and Gerd JuÈrgens1 1 ZMBP (Zentrum fuÈr Molekular Biologie der P¯anzen), UniversitaÈt TuÈbingen, Auf der Morgenstelle 3, D-72076 TuÈbingen, Germany, 2Laboratoire de Biologie Cellulaire, INRA, F-78026 Versailles, France, and 3Institut fuÈr P¯anzenbiochemie, Weinberg 3, D-06120 Halle/Saale, Germany Received 27 January 2002; revised 11 March 2002; accepted 21 March 2002 *For correspondence (fax: +49 7071 295797; email: [email protected])

Summary The sterol biosynthesis pathway of Arabidopsis produces a large set of structurally related phytosterols including sitosterol and campesterol, the latter being the precursor of the brassinosteroids (BRs). While BRs are implicated as phytohormones in post-embryonic growth, the functions of other types of steroid molecules are not clear. Characterization of the fackel (fk) mutants provided the ®rst hint that sterols play a role in plant embryogenesis. FK encodes a sterol C-14 reductase that acts upstream of all known enzymatic steps corresponding to BR biosynthesis mutants. Here we report that genetic screens for fklike seedling and embryonic phenotypes have identi®ed two additional genes coding for sterol biosynthesis enzymes: CEPHALOPOD (CPH), a C-24 sterol methyl transferase, and HYDRA1 (HYD1), a sterol C-8,7 isomerase. We describe genetic interactions between cph, hyd1 and fk, and studies with 15azasterol, an inhibitor of sterol C-14 reductase. Our experiments reveal that FK and HYD1 act sequentially, whereas CPH acts independently of these genes to produce essential sterols. Similar experiments indicate that the BR biosynthesis gene DWF1 acts independently of FK, whereas BR receptor gene BRI1 acts downstream of FK to promote post-embryonic growth. We found embryonic patterning defects in cph mutants and describe a GC±MS analysis of cph tissues which suggests that steroid molecules in addition to BRs play critical roles during plant embryogenesis. Taken together, our results imply that the sterol biosynthesis pathway is not a simple linear pathway but a complex network of enzymes that produce essential steroid molecules for plant growth and development. Keywords: sterols, embryogenesis, Arabidopsis, GC±MS, 15-azasterol, brassinosteroids.

Introduction Sterols play multi-faceted roles in eukaryotic biology, from structural roles in membranes (Bloch, 1983) to roles as signalling molecules (Edwards and Ericsson, 1999). Whereas mammalian and fungal cells generally contain one major sterol, cholesterol and ergosterol, respectively, plants have a characteristically complex sterol mixture referred to collectively as phytosterols (Hartmann, 1998; Patterson et al., 1993). The most abundant sterol found in plant cells is sitosterol, comprising 50±80% of the total sterols, followed by campesterol. Sitosterol, as the major component of plant membranes, appears to be important for membrane ¯uidity and permeability (Schuler et al., 1991). The only plant steroid molecules that have been assigned signalling functions so far are the campesterolã 2002 Blackwell Science Ltd

derived brassinosteroids (BRs) (Sakurai and Fujioka, 1997). BRs are phytohormones that have established roles in post-embryonic growth, including cell elongation and division (Clouse and Sasse, 1998). Signalling roles for the diverse array of plant sterols, including the most abundant species sitosterol, have not been described. A model for the plant sterol biosynthesis pathway is derived from sequence similarity to enzymes from other organisms, biochemical studies using puri®ed enzymes, heterologous expression (Bach and Benveniste, 1997; Benveniste, 1986) and mutant isolation (Altmann, 1999; Bishop and Yokota, 2001) (Figure 1). A branch in the pathway leads to the production of the two major endproducts, sitosterol and campesterol. This branch is 61

62

Kathrin Schrick et al.

controlled by the activity of C-28 sterol methyl transferase, and the ratio of campesterol to sitosterol is an important

Figure 1. A model for the sterol biosynthesis pathway in Arabidopsis from the ®rst cyclic precursor cycloartenol. A branch in the pathway results in biosynthesis the two major endproducts, sitosterol and campesterol, which is the precursor of the brassinosteroids (BRs). Enzymes are depicted in bold and corresponding genes identi®ed by mutant phenotypes in italics. CPH C-24 sterol methyl transferase (also termed SMT1), FK sterol C-14 reductase and HYD1 sterol C-8,7 isomerase steps and the corresponding genes are shown in boxes.

factor for plant growth (Schaeffer et al., 2001). Three Arabidopsis BR biosynthesis mutants cause a generic dwarf plant phenotype due to de®ciency in D7 sterol C-5 desaturase, D7 sterol C-7 reductase and C-24 reductase: dwf7/ste1 (Choe et al., 1999a; Gachotte et al., 1995; Husselstein et al., 1999), dwf5 (Choe et al., 2000) and dwf1/dim (Choe et al., 1999b; Klahre et al., 1998), respectively. Although dwf7/ste1, dwf5 and dwf1/dim mutants show reduced levels of both sitosterol and campesterol, their elongation defects can be rescued by BRs. This has led to the interpretation that their morphological defects are the consequence of a depletion of suf®cient amounts of campesterol for BR synthesis. Mutations that speci®cally affect the BR biosynthesis pathway cause a similar dwarf phenotype (Altmann, 1999; Bishop and Yokota, 2001). However, none of the BR dwarf mutants are reported to display defects in the embryo. Thus BRs are crucial for post-embryonic growth but there is no evidence for their role in embryogenesis. A role for steroid molecules in plant embryogenesis remained elusive until the recent description of two sterol biosynthesis mutants in Arabidopsis: fackel (fk) (Jang et al., 2000; Schrick et al., 2000) and sterol methyl transferase 1 (smt1) (Diener et al., 2000). Genetic screens for embryonic pattern formation mutants led to the isolation of fk mutants (Mayer et al., 1991). Identi®cation of the corresponding gene revealed that FK encodes a sterol C-14 reductase (Jang et al., 2000; Schrick et al., 2000). In genetic screens for short root mutants, smt1 mutants were identi®ed and shown to correspond to the gene for C-24 sterol methyl transferase, also referred to as sterol methyl transferase 1 (SMT1) (Diener et al., 2000). Strikingly, both the C-24 sterol methyl transferase (SMT1) and C-14 reductase (FK) steps are upstream of all BR dwarf mutants identi®ed so far (Figure 1). Moreover, fk and smt1 mutants both show embryonic defects (Diener et al., 2000; Jang et al., 2000; Schrick et al., 2000), suggesting that the sterol biosynthesis pathway produces steroid molecules in addition to BRs that are critical for embryonic growth and development (Clouse, 2000). To identify additional genes in the sterol biosynthesis pathway, we performed a genetic screen for fk-like phenotypes on a population of seedling mutants (Mayer et al., 1991; C. Bellini, unpublished results). Here we show the molecular characterization of six new mutants representing two genetic complementation groups, CEPHALOPOD (CPH) and HYDRA1 (HYD1). Point mutations in both genes map to sterol biosynthesis enzymes: CPH encodes a C-24 sterol methyl transferase while HYD1 encodes a sterol 8,7-isomerase (Grebenok et al., 1998; Souter et al., 2002). Both cph and hyd1 mutants, like fk mutants, exhibit patterning defects at the seedling stage. These defects can be traced back to abnormal cell divisions and expansions in the late globular to early heart stage ã Blackwell Science Ltd, The Plant Journal, (2002), 31, 61±73

Sterol biosynthesis genes in embryonic development embryo. To elucidate the relationship between the CPH, HYD1 and FK sterol biosynthesis enzymes, we performed

Figure 2. Seedling phenotypes and molecular characterization of the cph and hyd1 alleles. (a) Seedlings were germinated on 1.7% agar/water for 9 days. WT, wildtype control. cph-T357 and cph-GXIII display extremely variable seedling phenotypes. cph-GXIII seedlings represent a weak allele and display longer roots (arrow). hyd1-R216 and hyd1-E508 seedlings display similar variable phenotypes with shorter roots. Bars = 0.5 mm. (b) Positional cloning of CPH. A recombinant F2 population was used to map CPH to the top of chromosome 5, north of marker R89998. CAPS markers for determining recombination break points and numbers of recombinant chromosomes are indicated. Genetic distances between markers are given in cM. Three sequenced P1 clones from the CPH-containing region of 750 kb are shown (bold). The 2.8 kb fragment contains 13 exons (boxes) encoding 336 amino acids. Mutations identi®ed in the genomic DNA of cph alleles are marked and the corresponding amino acid positions are numbered. The cph-T357 allele contains an amino acid exchange in the putative sterol binding site YEYGWG conserved in the S. cerevisiae ERG6 protein. At, Arabidopsis thaliana; Sc, S. cerevisiae. (c) HYD1 has four exons (boxes) encoding 223 amino acids within a 1.1 kb fragment. Amino acid exchanges identi®ed in the hyd1 alleles are numbered. The hyd1-E508 allele encodes an amino acid change in a box of unknown function, WKEYSKGDSRY, which is conserved in the mouse C-8,7 sterol isomerase/emopamil binding protein.

ã Blackwell Science Ltd, The Plant Journal, (2002), 31, 61±73

63

double mutant analysis with the corresponding mutants. We utilized a speci®c inhibitor of the FK C-14 reductase to examine post-embryonic phenotypes of the different combinations. Our results indicate that, while HYD1 and FK act together, CPH functions independently to promote both embryonic and seedling growth. The sterol pro®le of cph mutants does not show a signi®cant reduction in campesterol, suggesting that the embryonic phenotypes cannot simply be attributed to a lack in BRs. We propose that the sterol biosynthesis pathway is not a simple linear

Figure 3. Seedling patterning defects and post-embryonic growth in cph and hyd1 mutants. (a) Seedlings grown on rich medium for 14 days. Wild-type (WT) seedlings show emergence of the ®rst two leaves. cph-G301 àpical' seedlings (asterisk) and `double root' seedlings (arrows) indicate patterning defects. (b) The CLV1±GUS reporter marks the shoot apical meristem and the vasculature in wild-type (WT) and cph seedlings. Note the enlarged meristem and multiple vascular strands in cph. (c) Patterning defects in hyd1 seedlings: àpical' (asterisk), `double root' (arrows) and multiple apices (asterisks). (d) Vegetative stage plants at 30 days: wild-type (WT), cph, hyd1 and fk. hyd1 and fk mutants display a similar phenotype, while cph mutants are larger in comparison. (e) Reproductive stage plants at 60 days: wild-type (WT), and cph and hyd1 variable phenotypes. Bars = 0.5 mm (a±d), 1 cm (e).

64


pathway but a more complex interplay of enzymatic activities leading to the production of sterols that play as yet unidenti®ed roles in plant development. Results and discussion Genetic screens to identify fk-like mutants Mutations affecting the BR biosynthesis pathway in Arabidopsis display a generic dwarf phenotype that is expressed post-embryonically and can be rescued by exogenous application of BRs (Altmann, 1999; Bishop and Yokota, 2001). These dwarf phenotypes were helpful in placing corresponding genes in the same biological pathway prior to their cloning. In contrast to previously identi®ed BR dwarf mutants, fk mutants were found to exhibit embryonic and seedling patterning defects (Jang et al., 2000; Schrick et al., 2000). FK encodes a sterol C-14 reductase that acts upstream of all biosynthesis steps characterized by BR dwarf phenotypes (Figure 1). To identify additional genes in this upstream branch of the sterol biosynthesis pathway, we screened an EMS-mutagenized seedling population (Mayer et al., 1991; C. Bellini, unpublished results) for fk-like phenotypes: short hypocotyl and root, and malformed cotyledons. Allelism tests veri®ed six fk-like mutants that were not allelic to fk, de®ning two new complementation groups that we designated CPH and HYD1. The hyd1 mutants showed allelism to the hyd1-1 mutant previously identi®ed (Topping et al., 1997). Four alleles of CPH were isolated from the Landsberg erecta (Ler) ecotype; cph-G301, cphG213 and cph-T357 show similar seedling phenotypes, while cph-GXIII shows a slightly weaker phenotype, as evidenced by longer roots (Figure 2a). The hyd1-E508 and hyd1-R216 alleles show nearly identical seedling phenotypes (Figure 2a), although they originate from different ecotypes, Columbia (Col) and Ler, respectively. CPH and HYD1 encode sterol biosynthesis enzymes Molecular characterization revealed that both cph and hyd1 mutations affect genes in the upstream branch of the sterol biosynthesis pathway (Figure 1). A map-based cloning approach was applied to isolate the CPH gene, which was found to encode a C-24 sterol methyl transferase (Figure 2b). CPH is identical (with minor sequence differences, see below) to SMT1 (Diener et al., 2000), isolated previously from the Col ecotype. The CPH gene isolated from the Ler ecotype differs from the Col sequence (accession number AAG28462) at three nucleotide positions, two of which represent differences at the amino acid level (Val to Ala and Ser to Asn at positions 119 and 162, respectively) that presumably do not alter protein function. The cph mutations result in splice site ablations

(cph-G301, cph-GXIII), an early stop codon (cph-G213) and an amino acid exchange (cph-T357) in the predicted sterolbinding pocket that is conserved in the S. cerevisiae sterol methyl transferase ERG6 (Marshall and Nes, 1999) (Figure 2b). As the cph-GXIII allele showed a weaker seedling phenotype compared to the other alleles, we characterized this change further using RT±PCR. The cphGXIII transcript showed a one-base deletion of a G at the beginning of the last exon. The AA to AG splice site mutation prior to this last 16 amino acid exon leads to aberrant usage of the adjacent G for splicing, resulting in a single-base frameshift. Thus the predicted cph-GXIII mutant protein contains a C-terminal exon with an altered amino acid composition that apparently retains partial function. The HYD1 gene encodes a sterol C-8,7 isomerase (Souter et al., 2002). Both hyd1-E508 and hyd1-R216 mutations result in amino acid exchanges in conserved residues of the predicted HYD1 protein (Figure 2c). The amino acid exchange in hyd1-E508 lies within a sequence box of unknown function that is completely conserved in the mouse sterol C-8,7 isomerase protein. The cDNA corresponding to the HYD1 sterol C-8,7 isomerase was previously isolated by its complementation of the erg2 mutant from S. cerevisiae (Grebenok et al., 1998), consistent with its function as a sterol C-8,7 isomerase. Phenotypic characterization reveals patterning and postembryonic defects in cph and hyd1 mutants Patterning defects similar to those found in fk seedlings (Schrick et al., 2000) were observed in cph and hyd1 mutants. Examples included seedlings with apical (cotyledon) tissue at the expense of root tissue, which we termed àpical', abnormal seedlings with two embryonic roots termed `double root', and seedlings with multiple apices (Figure 3a,c). Expression of a CLV1±GUS marker showed that the shoot apical meristem is typically expanded in cph seedlings, and that the vascular strands are multiple and/ or not tightly bundled as in wild-type (Figure 3b). cph and hyd1, like fk mutants, exhibit a high degree of variability as illustrated by the seedling phenotypes in Figure 2(a). To explain this, we envisage that, in mutant embryos, cell division and/or polarity decisions are randomized such that embryos in which early decisions fail give rise to the most severe seedling phenotypes. In general, cph seedlings and plants displayed weaker post-embryonic phenotypes, as exempli®ed by longer roots in comparison to hyd1 or fk, whereas hyd1 mutants showed post-embryonic phenotypes more similar to that of fk (Figures 2a and 3d; Schrick et al., 2000). However, cph seedlings were not viable on soil, in contrast to smt1 mutants isolated in the Wassilewskija (WS) ecotype (Diener et al., 2000). We investigated whether this is due ã Blackwell Science Ltd, The Plant Journal, (2002), 31, 61±73

Sterol biosynthesis genes in embryonic development to an ecotype difference in gene expression of other sterol methyl transferases. In Arabidopsis, C-24 sterol methyl transferase is encoded by a single gene, whereas the related C-28 sterol methyl transferase is encoded by two genes, SMT2;1 and SMT2;2 (also called SMT2 and SMT3) (Diener et al., 2000). RT±PCR experiments con®rmed that both SMT2 genes are expressed in the Ler and Col ecotypes (data not shown). Thus the phenotypic difference between cph and smt1 mutants may be due to some other difference in their genetic backgrounds. In contrast, both hyd1 alleles, regardless of ecotype background, were not viable on soil. Nonetheless, cph and hyd1 mutants, like fk mutants, could be propagated on rich medium to reach the vegetative stage (Figure 3d). About 60% of cultured cph and 10% of hyd1 plants reached the reproductive stage (Figure 3e), although ¯owers showed reduced fertility. Embryonic phenotypes of cph and hyd1 mutants In embryogenesis of Arabidopsis, cells of the proliferating embryo acquire speci®c cell fates in an integrated manner to form the three-dimensional pattern evident in the seedling (Schrick and Laux, 2001). The embryo undergoes characteristic shape changes, from globular to heart to torpedo, brought about by different orientations of cell divisions and directional cell expansions. The cph and hyd1 mutants display abnormal embryonic phenotypes (Figure 4) similar to those of fk mutants (Jang et al., 2000; Schrick et al., 2000). The earliest deviations from the wildtype were observed at the late globular to early heart stages (Figure 4e,m), when both cph and hyd1 embryos failed to elongate cells in the centre of the embryo. Both cph and hyd1 mutants exhibited a delay in development, as compared to wild-type (Figure 4a±f,i±p). Sections through cph embryos revealed cell division and expansion defects (Figure 4e,f) like those seen in fk embryos (Schrick et al., 2000). We did not observe a different embryonic phenotype for the cph-GXIII allele, which showed a weak post-embryonic phenotype, suggesting that there is a more stringent requirement for CPH function during embryogenesis. A patterning defect in cph embryos was supported by the presence of multiple or expanded shoot apical meristems, as detected by a CLV1±GUS reporter that marked a single shoot apical meristem in wild-type (Figure 4d,h). Expression of CLV1±GUS in fk embryos gave similar results (G. Martin and G. JuÈrgens, unpublished results). It was previously shown that fk embryos exhibit multiple, misplaced or expanded domains of STM expression (Schrick et al., 2000), indicating a defect in positioning of the shoot apical meristem. Thus cph and fk mutants share a common patterning defect in the embryo. This suggests a cell-signalling defect that may involve unidenti®ed steroid signals that are lacking in ã Blackwell Science Ltd, The Plant Journal, (2002), 31, 61±73

65

the mutants (Schrick et al., 2000). However, it is also possible that abnormal steroid products in fk and cph mutants interfere with crucial signalling processes in embryogenesis. GC±MS analysis of cph tissues reveals a different sterol pro®le than that of fk mutants Consistent with a block at the C-14 reductase step, fk mutants accumulate sterol molecules that have the steroid nucleus structure of D8,14 sterols (Figure 1; Schrick et al., 2000). Interestingly, these D8,14 sterols have side-chain modi®cations at the C-24 that are equivalent to those in campesterol and sitosterol. The structures of these molecules reveal that not only upstream modi®cations, such as the C-24 methyl transferase step, but also downstream modi®cations, such as the C-28 methyl transferase step, occur in fk mutants. To examine whether downstream modi®cations also occur in cph mutants, we performed GC±MS analysis on two different cph tissues: seedlings and callus cells (Table 1). These tissues showed different sterol pro®les for wild-type as well as cph samples, suggesting general differences in sterol biosynthesis or metabolism in differentiated seedlings versus actively dividing callus cells. A block at the CPH C-24 methyl transferase step is expected to lead to accumulation of the biosynthetic intermediate cycloartenol (Figure 1). Indeed, we detected elevated levels of cycloartenol of up to 40.8% of the total sterols (cph-G213; Table 1) in cph callus cell cultures. Both cph tissues showed an abnormal accumulation of cholesterol and 24-methylenecholesterol, which was more pronounced in seedlings, 42.6% and 44.6% of total sterols for cph-G213 and cph-T357, respectively (Table 1), and higher than the level previously reported for an smt1 allele (36% of total sterols) (Diener et al., 2000). Seedlings from the cph-GXIII allele showed accumulation of these sterols to a lesser extent (12.6% of total sterols), consistent with its weaker seedling phenotype. In cph-G213 and cph-T357 seedlings, the elevated levels of cholesterol were accompanied by a signi®cant accumulation of pollinastanol and 29-norcycloartanol, which are thought to be biosynthetic precursors of cholesterol that, like cholesterol, lack a C-24 methyl addition. Surprisingly, we detected the accumulation of 24methylenecycloartenol, the product of C-24 methyltransferase, in cph-GXIII seedlings and callus cells (11.4% and 18.9% of total sterols, respectively). This may be explained by partial function of the weak cph-GXIII allele coupled with attenuated activity of C-4 demethylase, the next enzyme in the pathway. It is possible that CPH is partially required for C-4 demethylase activity, an idea that ®ts with the ®nding that callus cells from the strong alleles cphG213 and cph-T357 also accumulate 24-methylenecycloar-

66


Figure 4. Embryonic defects in cph and hyd1 mutants. (a±d) Wild-type and (e±h) cph. (a,e) Early heart and (b,f) torpedo stages from sectioned embryos. cph embryos appear delayed in development and exhibit cell expansion defects. (c,g) Transverse sections of embryonic hypocotyls are shown. Note additional cell layers and abnormal cell morphologies in cph (g). (d,h) The CLV1±GUS reporter marks the shoot apical meristem in wild-type (d) and cph (h) embryos. Note two distinct domains of expression in cph (h). (i±p) Whole-mount preparations of wild-type (i±l) and hyd1 (m±p) embryos visualized with Nomarski optics. (i,m) Early heart, (j,n) heart, (k,o) torpedo, and (l,p) bent-cotyledon stages. hyd1 embryos are delayed in development and show clear defects in cell expansion (m±p). (q±t) Embryos from doubly heterozygous plants having cph hyd1 or cph fk combinations. (q) Four ovules from a single silique of a cph fk heterozygous plant. The putative double mutant (arrow) is tiny in comparison to the single mutants (asterisks). (r) cph hyd1 double mutant, and (s,t) cph fk double mutants. Note the enlarged suspensors (arrows) and radial symmetry in the apical portion of the embryos (r±t). Bars = 40 mm (a±c, e±g, i±p), 0.5 mm (d,h,q), 20 mm (r±t).

tenol, albeit to a lesser extent (5.5% and 10.2% of total sterols, respectively). Assuming that these are null alleles, the appearance of 24-methylenecycloartenol suggests that other enzymes such as the C-28 sterol methyltransferases (SMT2.1 and SMT2.2) can partially substitute for CPH C-24 methyltransferase in its absence.

Figure 5. Effects of 15-azasterol on seedling growth. Seedlings were germinated on media containing varying concentrations of 15-azasterol. (a) The sum of hypocotyl and root lengths (cm) at 21 days after germination was plotted against the concentration of 15azasterol for wild-type (WT), bri1, twd1, dwf5 and two dwf1 alleles, dwf16 and dwf1-2. Data bars represent at least 50 seedlings in duplicate experiments giving the calculated standard deviations. Note that twd, dwf5 and dwf1 seedlings display a signi®cant length decrease at 0.10 mM 15-azasterol in comparison to WT and bri1. (b) Wild-type (WT), bri1 and dwf1 seedlings were germinated on 0 (15-azasterol-free control), 0.10 and 1.0 mM 15-azasterol. Whereas bri1 seedlings appear to be less sensitive to 0.10 mM treatment, dwf1 seedlings are more sensitive to 0.10 mM 15azasterol compared to wild-type. (c) cph, hyd1 and fk mutants were germinated on 0 (15-azasterol-free control) (top row) and 1.0 mM concentrations of 15-azasterol (bottom row). The phenotypic variability of the mutants is indicated, with the weakest to the strongest phenotypes from left to right. cph seedlings are sensitive to 1.0 mM 15-azasterol while hyd1 and fk seedlings appear insensitive. Bars = 0.5 mm.

As expected for an upstream block in the sterol biosynthesis pathway, cph extracts from both tissues showed a reduction in the end-product sitosterol. This effect was enhanced in seedlings, which showed a reduction from 80% of the total sterols in wild-type to 25% in cph-G213 (Table 1). However, the other major end-product campesterol and downstream biosynthetic intermediates such as isofucosterol and sitostanol were not signi®cantly ã Blackwell Science Ltd, The Plant Journal, (2002), 31, 61±73

Sterol biosynthesis genes in embryonic development

67

Table 1. Mass spectral analysis of sterols from cph tissues shows abnormal accumulations of cholesterol, 24-methylenecholesterol, pollinastanol, 29-norcycloartenol, cycloartenol and 24-methylenecycloartenol Seedlings Sterola

RT (min) M+ (m/z) WT

Cholesterol 7.63 24-Methylcholesta-5,22-dien-3b-ol 7.93 5a-Cholest-7-en-3b-ol 8.06 Pollinastanol 8.20 24-Methylenecholesterol 8.37 Campesterol 8.48 Campestanol 8.5 29-Norcycloartanol 8.63 Stigmasterol 8.70 Sitosterol 9.30 Sitostanol 9.40 Isofucosterol 9.40 Cycloeucalenol 9.58 Cycloartenol 9.74 24-Methylenecycloartenol 10.35

368b 380b 428 442 380b 382b 444 456 394b 396b 458 456 468 468 482

Callus cells

cph-G213 cph-T357 cph-GXIII WT

1.3 33.9 0.8 1.4 0.3 1.6 ± 7.3 0.6 8.7 13.8 18.0 0.2 (+) ± 2.0 0.5 (+) 80.8 25.1 0.9 0.7 ± ± 0.4 0.5 ± ± 0.4 0.8

36.2 1.1 1.1 3.3 8.4 16.7 (+) 2.2c 2.2c 29.3 0.9 ± 0.8 ± ±

10.4 0.5 0.2 1.3 2.1 12.3 0.2 ± 0.4 57.7 0.1d 0.1d ± 3.4 11.4

± ± ± ±

0.3

5.4 (+) ± 0.7 88.9 0.8d 0.8d ± ± 3.9

cph-G213 cph-T357 cph-GXIII ± ±

4.5 0.2

0.6 7.2 (+) ± (+) 40.1 1.1d 1.1d ± 40.8 5.5

± ± ±

5.6

1.0 8.9 (+) ± (+) 49.8 1.1d 1.1d ± 23.4 10.2

± ± ±

3.4

0.4 8.4 (+) ± (+) 59.2 0.2d 0.2d ± 9.5 18.9

Percentages of the relative sterol composition are shown. Sterols were determined as acetates by GC±MS; b [M-HOAc]+ ±, not present; (+), traces; WT, wild-type. cph-G213, cph-T357 and cph-GXIII alleles are indicated. Sterols are listed according to the retention time (RT) and the mass (M) is for each sterol is shown. c Stigmasterol and 29-norcycloartanol represent overlapping peaks. d Isofucosterol and sitostanol represent overlapping peaks. a

Table 2. Seedling genotypes from doubly heterozygous parents show absence of the double mutant class for cph hyd1 and cph fk combinations Parent plant

Seedling progeny

% heterozygous (expected)

% WT (expected)

CPH/cph HYD1/hyd1

hyd1/hyd1 (n = 24) cph/cph (n = 98) fk/fk (n = 81) cph/cph (n = 26) fk/fk (n = 19) hyd1/hyd1 (n = 20)

CPH/cph HYD1/hyd1 CPH/cph FK/fk HYD1/hyd1 FK/fk

CPH/CPH HYD1/HYD1 CPH/CPH FK/FK HYD1/HYD1 FK/FK

CPH/cph FK/fk HYD1/hyd1 FK/fk

54 55 64 73 53 55

(67) (67) (67) (67) (50) (50)

46 45 36 27 16 15

% Double mutant (expected) (33) (33) (33) (33) (25) (25)

cph/cph hyd1/hyd1 cph/cph fk/fk hyd1/hyd1 fk/fk

0 (0) 0 (0) 0 (0) 0 (0) 32 (25) 30 (25)

Crosses were performed to generate heterozygous plants with the given double mutant combinations. The alleles used were cph-T357 fkT293, cph-G213 hyd1-R216, and hyd1-E508 fk-5D8. Genotyping of parent plants and seedling progeny was done using PCR and restriction digest to detect allele-speci®c polymorphisms. The expected percentages that best ®t the data are shown in parentheses. n, number of seedling progeny examined with the given genotype. The genotype for the second locus is indicated as heterozygous, wild-type (WT) or double mutant.

reduced. In contrast, the sterol pro®les of fk mutants exhibit a reduction in both sitosterol and campesterol, as well as several downstream biosynthetic intermediates (Schrick et al., 2000). Therefore the cph and fk mutant phenotypes may be attributed to the absence of sitosterol or steroid products not detected by our analysis, and/or the deleterious effects of abnormal steroid molecules. cph extracts did not accumulate D8,14 sterols which are found in fk mutants, suggesting that the defects are not speci®cally caused by these biosynthetic intermediates. In addition, the lack of D8,14 sterols in cph mutants indicates that the Cã Blackwell Science Ltd, The Plant Journal, (2002), 31, 61±73

14 reductase step mediated by FK appears to occur in the absence of CPH. Thus our sterol analyses suggest that downstream sterol modi®cations occur in cph as well as in fk mutants. Double mutant analysis with cph, hyd1 and fk combinations We examined genetic interactions between CPH, HYD1 and FK to test the possibility that the corresponding sterol biosynthesis enzymes function independently of one

68


Table 3. Embryonic phenotypes from doubly heterozygous plants: detection of a double mutant class for cph hyd1 and cph fk combinations

Parent plant

Total embryos

% Single mutant phenotype (expected)

% Double mutant phenotype (expected)

CPH/cph HYD1/hyd1 CPH/CPH HYD1/hyd1 CPH/cph HYD1/HYD1 CPH/cph FK/fk CPH/CPH FK/fk CPH/cph FK/FK HYD1/hyd1 FK/fk HYD1/hyd1 FK/FK HYD1/HYD1 FK/fk

897 323 411 821 529 582 1069 551 615

39 23 23 40 25 21 44 26 25

4 0 0 4 0 0 0 0 0

(38) (25) (25) (38) (25) (25) (44) (25) (25)

(6) (0) (0) (6) (0) (0) (0) (0) (0)

Crosses were performed to generate heterozygous plants with the given double mutant combinations. The alleles used were cph-T357 fkT293, cph-G213 hyd1-R216, and hyd1-E508 fk-5D8. Genotyping of parent plants and seedling progeny was done using PCR and restriction digest to detect allele-speci®c polymorphisms. The expected percentages that best ®t the data are shown in parentheses. F1 plants were grown to reproductive stage. The percentages of embryos with single mutant and novel double mutant phenotypes are shown. No mutant embryos were detected in wild-type plants from each set.

another. The three genes are unlinked on different chromosomes, leading to a Mendelian expectation of 1/ 16 (6%) for the double mutant and 3/16 (19%) for each single mutant. Allele-speci®c polymorphisms were utilized to genotype doubly heterozygous parent plants containing cph hyd1, cph fk or hyd1 fk (Tables 2 and 3). If the sterol biosynthesis pathway is linear as in Figure 1, the expectation is that the double mutant should display the same phenotype as the most severe single mutant. However, as the sterol pro®les of cph and fk tissues suggest partial blocks at the corresponding enzymatic steps (Table 1; Schrick et al., 2000), an additive phenotype could be expected for cph fk double mutants. Our analysis revealed that only plants harbouring hyd1 fk showed viable double mutant seedling progeny exhibiting an fk-like phenotype. By contrast, no viable double mutant seedlings were found for cph hyd1 or cph fk, suggesting an additive phenotype for these combinations (Table 2). Embryonic lethality of the cph hyd1 and cph fk double mutants could explain our inability to recover viable seedlings with these genotypes. To test this possibility, we examined embryos from plants genotyped as doubly heterozygous for the different combinations (Table 3). Only plants harbouring cph hyd or cph fk mutations segregated a novel embryonic phenotype that, from its frequency (4% each for cph hyd and cph fk), approximates the expectation for the double mutant (6%). Figure 4(q) shows four ovules from a plant heterozygous for cph and fk. The putative double mutant is easily recognized by its distinct small size, indicating an early arrest in development. cph hyd1 and cph fk embryos showed similar phenotypes: In addition to their small size, the arrested double mutant embryos exhibited radial symmetry in the apical portion of the embryo and an abnormal suspensor

(Figure 4r±t). The embryos appear to arrest at the globular stage, prior to cellularization of the endosperm, a stage at which maternal sterol stores become less accessible. However, the arrested embryos are grossly malformed compared to a wild-type globular embryo, indicating that sterols, if harnessed from the mother tissue, are not suf®cient for properly oriented cell divisions and expansions. Moreover, the suspensor is often multi-cellular at its girth, indicating that the embryo lacks a normal apical± basal pattern. The novel embryonic phenotype of cph hyd1 and cph fk double mutants, together with absence of viable cph hyd1 and cph fk genotypes among the seedling progeny, suggest that HYD1 and FK function independently of CPH in promoting growth and cell division during embryogenesis. 15-azasterol studies to examine post-embryonic cph, hyd1 and fk combinations What is the relationship between CPH, HYD1 and FK gene functions in post-embryonic development? As the cph hyd1 and cph fk double mutant combinations were embryo-lethal, we were unable to analyse their postembryonic phenotypes genetically. To circumvent this problem, we utilized the anti-mycotic agent 15-aza-24methylene-D-homocholesta-8,14-dien-3b-ol (15-azasterol), which has been shown to be a strong speci®c inhibitor of sterol C-14 reductase in plant cell cultures (Schmitt et al., 1980). Thus the FK C-14 reductase is predicted to be a speci®c target of the drug. Wild-type seedlings germinated on agar medium containing varying concentrations of 15azasterol showed a reduction in both hypocotyl and root elongation, with a maximal response at a dose of 1 mM (Figure 5a). Other wild-type responses included reduced ã Blackwell Science Ltd, The Plant Journal, (2002), 31, 61±73

Sterol biosynthesis genes in embryonic development

Figure 6. FK and HYD1 act independently of CPH to modify essential steroid products. (a) A model for the action of the three sterol biosynthesis genes. While FK and HYD1 act in a sequential manner, CPH acts independently to produce steroids that are essential for cell division and growth during embryogenesis and post-embryonic development. (b) The proposed structure of a cph fk sterol that accumulates in arrested double mutant embryos. The steroid nucleus has the structure of D8,14 sterols (double bonds at C-8 and C-14) that accumulate in fk mutants (Schrick et al., 2000). The C-24 in the side-chain lacks a methyl addition, like C-24 in cholesterol or cycloartenol (shown above), two sterols that accumulate in cph mutants (Table 1). For comparison, sitosterol has a double bond at C-5 and two methyl additions at C-24.

leaf expansion and an overall dwarfed appearance (Figure 5b). As expected, fk mutant seedlings were not additionally affected by 15-azasterol treatment (Figure 5c). We observed that hyd1 seedlings were similarly insensitive to 15-azasterol treatment at all concentrations tested. In contrast, cph seedlings showed hypersensitivity to the drug at a dose of 1 mM, as re¯ected by necrosis and reduced leaf outgrowth (Figure 5c). Thus the cph hyd1 and cph fk double mutant lethality was mimicked by application of 15-azasterol to post-embryonically germinating cph mutant seedlings. We conclude that the HYD1 and FK genes act independently of CPH to promote post-embryonic growth.

FK and HYD1 act sequentially whereas CPH functions independently In the experiments described above, we have characterized the phenotypes of new cph and hyd1 alleles and shown that they represent lesions in sterol biosynthesis enzymes that, like the FK C-14 reductase, act in the upstream sterol biosynthesis pathway. If this pathway is linear (Figure 1), double mutant phenotypes should resemble the most severe single mutant phenotype. This is true for the hyd1 fk double mutants that were found to exhibit seedling phenotypes indistinguishable from either single mutant. This suggests that HYD1 and FK depend on each other for function. The FK C-14 reductase step immediately ã Blackwell Science Ltd, The Plant Journal, (2002), 31, 61±73

69

precedes the HYD1 C-8,7 isomerase step (Figure 1), and, furthermore, both enzymes are predicted to be integral membrane proteins of the endoplasmic reticulum (ER) (Grebenok et al., 1998; Schrick et al., 2000), suggesting that the two enzymes may interact directly with one another. In this respect, it is interesting that GC±MS analysis of hyd1 sterol extracts shows a sterol pro®le that is qualitatively similar to that found in fk mutants (K. Schrick and S. Fujioka, unpublished results), suggesting that FK is not functional in the absence of HYD1. Thus it may not be surprising that hyd1 seedlings are as insensitive as fk seedlings to 15-azasterol (Figure 5c). In addition, our observation that hyd1 or fk mutations in combination with cph lead to the same double mutant phenotype is consistent with the idea that HYD1 and FK act together. The CPH protein, unlike HYD1 and FK, does not contain multiple transmembrane domains and therefore is not predicted to be an integral membrane protein of the ER. In fact, its functional homologue in yeast, ERG6 (Diener et al., 2000), has been identi®ed as a major component of intracellular lipid particles (Athenstaedt et al., 1999), which may be transiently associated with the ER. Thus CPH action may be spatially different than that of HYD1 and FK. Taken together, the novel phenotype of cph hyd1 and cph fk double mutants (Figure 4r,s) and the deleterious effects of 15-azasterol on germinating cph seedlings (Figure 4c) indicate that HYD1 and FK function independently of CPH in the sterol biosynthesis pathway (Figure 6a). We propose that the sterol pro®les of the fk cph double mutants, if they could be measured, would re¯ect a combination of defects in fk and cph, giving abnormal sterol products that, unlike wild-type or single mutant endproducts, are unable to support cell division and growth (Figure 6b). The missing end-products in the double mutant may play structural roles in membrane integrity (Ourisson, 1994) and/or act as important signalling molecules in development. Alternatively, or in addition, it is possible that the abnormal double mutant sterols are deleterious to cell growth and/or signalling processes. The relationship to BR biosynthesis and BRI1-mediated BR signalling The CPH, HYD1 and FK sterol biosynthesis steps are predicted to act upstream of all of the known BR biosynthesis genes to date (Figure 1). To explore the relationship between the components of the upstream and BR dwarf branches, we analysed genetic interactions between fk and dwf1 (Choe et al., 1999b; Klahre et al., 1998) mutants. Examination of the seedling progeny from plants heterozygous for both mutations showed that the fk dwf1 double mutant is rarely seedling viable: Only 2% (1/65) of fk seedlings were also dwf1, although the Mendelian

70


expectation is 25%. Thus the fk dwf1 double mutant displays a phenotype that is different from either single mutant. We were unable to detect a novel embryonic phenotype among 475 embryos derived from a plant doubly heterozygous for dwf1 and fk, suggesting that fk dwf1 double mutants are either gametophytically lethal or display an embryonic phenotype that escaped our detection. These ®ndings show that FK and DWF1 act independently of each other in the sterol biosynthesis pathway. To further test this idea, we examined 15azasterol sensitivity, and thus absence of FK C-14 reductase function, in dwf1 and in another BR biosynthesis dwarf mutant, dwf5 (Choe et al., 2000). We also included twisted dwarf 1 (twd1), a non-BR dwarf mutant, because the corresponding gene encodes a predicted immunophilin-like protein that is not expected to act in a common pathway with sterol biosynthesis enzymes (Schulz et al., personal communication). All three mutants displayed an increased sensitivity to 0.10 mM 15-azasterol in comparison to wild-type (Figure 5a). The increased sensitivity of dwf1 mutants to 15-azasterol suggests that FK acts independently of DWF1 in the biosynthesis of sterols essential for growth, consistent with our genetic data for the fk dwf1 combination. Moreover, the 15-azasterol studies suggest that FK functions independently of DWF5 and TWD1 to promote post-embryonic growth. To explore the relationship between the upstream sterol biosynthesis branch and BR signalling, we examined genetic interactions between FK and BRI1, which encodes a leucine-rich repeat receptor kinase that is an essential component of a receptor for BRs (He et al., 2000; Wang et al., 2001). Analysis of the seedling progeny of a plant heterozygous for fk and bri1 showed that the fk bri1 double mutant is viable and displays a fk phenotype: 19% (5/27) of fk seedling progeny were also bri1, approximating the Mendelian expectation of 25%. This result suggests that FK acts upstream of BRI1. As fk mutants are de®cient in BRs (Jang et al., 2000), it is conceivable that BRI1 receptors are not functional in the fk background simply due to the absence of BR ligands. This idea is supported by the fact that exogenous application of BRs induces a growth response in fk seedlings (Schrick et al., 2000). On treatment with 0.10 mM 15-azasterol, bri1 seedlings, unlike dwf1, dwf5 and twd1 mutants, did not exhibit a hypersensitive response (Figure 5a,b), consistent with our genetic data. As bri1 mutants do not exhibit an embryonic phenotype, it appears that the BRI1-mediated BR signalling pathway is not essential for cell proliferation during embryogenesis. We conclude that FK and BRI1 act in a common BR signalling pathway to promote postembryonic growth. The embryonic defects in cph, hyd1 and fk mutants suggest that steroid molecules in addition to BRs are important for embryogenesis. Alternatively, other types of BR receptors besides BRI1 may mediate the

BR signalling that is essential for embryogenesis. However, none of the BR-speci®c biosynthesis mutants identi®ed so far exhibit embryonic defects, making this explanation less plausible. In addition, GC±MS analysis of cph tissues showed that levels of campesterol, which is the precursor of the BRs, are not signi®cantly reduced (Table 1), suggesting that it is not the absence of BRs, but the absence of other essential steroid molecules, and/ or the presence of abnormal steroids, that lead to the embryonic defects. Concluding remarks Our results indicate that CPH and HYD1, like FK, are required for proper pattern formation during embryogenesis. However, double mutant analysis shows that these genes do not strictly depend on one another for the production of essential steroids for growth and cell division. Thus the sterol biosynthesis pathway in plants is not a simple linear pathway (Figure 1) but represents a more complex interplay of enzymatic activities (Figure 6a). Some components such as FK and HYD1 appear to act closely with one another to modify sterol substrates, whereas others, such as the C-24 sterol methyl transferase CPH, act independently. Moreover, from genetic analysis and experiments utilizing the sterol C-14 reductase inhibitor 15-azasterol, it appears that enzymes of the downstream pathway leading to BR biosynthesis, such as DWF1 and DWF5, may also act independently of upstream enzymes such as FK. Future investigations of sterol biosynthesis enzymes will help to elucidate the relationships between additional components, and lead to an improved understanding of which products of the sterol biosynthesis pathway play important structural or signalling roles in embryogenesis. Experimental procedures Plant strains and growth conditions The Arabidopsis thaliana ecotypes Landsberg erecta (Ler) and Columbia (Col) were used as wild-type controls. Niederzenz (Nd) was used for genetic mapping of CPH. The cph and hyd1-R216 alleles were induced in Ler by EMS mutagenesis (Mayer et al., 1991). The hyd1-E508 allele was induced in Col by a separate EMS mutagenesis (C. Bellini, unpublished results). cph and hyd1 alleles were propagated as heterozygotes. Plants were grown on soil at 18°C or 25°C under 16-h day conditions (Lumiluxâ Plus Cool White L36W/21-840 (OSRAM, Germany) at about 6,700 at the level of the pot).

Phenotypic analyses To assay seedling phenotypes, seeds from heterozygous plants were germinated on simple agar growth medium or on rich medium as previously described (Schrick et al., 2000). Mutant ã Blackwell Science Ltd, The Plant Journal, (2002), 31, 61±73

Sterol biosynthesis genes in embryonic development seedlings were transferred to rich medium in magenta boxes to allow growth to vegetative and reproductive stages. Preparation of callus cell cultures was as described by Schrick et al. (2000), except that the liquid-rich medium contained 3% sucrose. Wholemount preparation, histological analysis and microscopy of ovules were as described previously (Hamann et al., 1999; Mayer et al., 1993). Digital images of seedlings and plants were captured with a Nikon Coolpix or Sony DSC-S75. Images were processed with Photoshop 5.0 and Illustrator 9.0 software (Adobe, Mountain View, California, USA).

Molecular cloning of CPH and molecular characterization of cph and hyd1 alleles The F2 progeny from a cross between CPH/cph-GXIII (Ler) and wild-type (Nd) were used as the mapping population. By linkage to CAPS markers R89998, F20L, nga151 and LRP1, CPH was mapped to a 1.9 cM region on the top of chromosome 5. This region contained three sequenced P1 clones: MXC9, MSH12 and MAC12 (Kazusa). Only MSH12 contained an open reading frame, MSH12.10, with similarity to a sterol biosynthesis enzyme, namely C-24 sterol methyl transferase from Glycine max. A set of gene-speci®c primers for MSH12.10 was used to amplify and sequence genomic DNA from cph seedlings. Point mutations representing GC to AT transitions in all four alleles con®rmed that CPH is MSH12.10, also reported as SMT1 (Diener et al., 2000). Similarly, four speci®c primers from the gene for sterol 8,7isomerase (Grebenok et al., 1998) were used to amplify the genomic DNA from hyd1 seedlings. GC to AT transitions in both hyd1 alleles identi®ed lesions in the sterol 8,7-isomerase gene. Information regarding the gene-speci®c primers used for sequencing the cph and hyd1 alleles are available upon request.

CAPS markers The F20L marker is derived from the centromeric and the telomeric ends of BACs F18G11 and F20K17, respectively (S. Wallisch and G. JuÈrgens, unpublished results). Primers 5¢-GGACGTTCCTCGAATTAGAG-3¢ and 5¢-ACTCACTCTCCTTTCAGTTATCAG-3¢ (annealing temperature of 56°C) ampli®ed a 0.67 kb fragment. TaqI digestion resulted in two major fragments in Ler (0.30 and 0.22 kb) and three major fragments in Nd (0.25, 0.22 and 0.05 bp). The LRP1 marker is derived from the LATERAL ROOT PRIMORDIA 1 (LRP1) gene (Smith and Fedoroff, 1995). Primers 5¢AGTAAGCAATCCAACAGTGTCGGAAGGATG-3¢ and 5¢-ATAAACAACATTAGGCGGGTCGAGAATAGG-3¢ ampli®ed a 2.21 kb fragment (annealing temperature of 58°C). HpaII or MspI digestion resulted in four fragments in Ler (0.83, 0.76, 0.33 and 0.29 kb) and ®ve fragments in Nd (0.76, 0.47, 0.36, 0.33 and 0.29 kb). The markers nga151 and R89998 are described in Bell and Ecker (1994) and at TAIR (http://www.Arabidopsis.org/servlets/TairObject? accession=marker:1945641), respectively.

71

Allele-speci®c polymorphisms The fk-T293 and fk-X224 alleles were described previously (Schrick et al., 2000). fk-5D8, an EMS-induced allele in the Col ecotype, encodes a splice site mutation (AG to AA) prior to exon 12 at amino acid 292. The bri1-2 and dwf1-6 alleles were used in double mutant combinations with the fk-X224 and fk-T293 alleles, respectively. bri1-2 contains an 8 bp imperfect repeat (ATGTCATA) between amino acids 483 and 484 (S. Clouse, personal communication). dwf1-6 contains a 6 bp perfect repeat (TTGTAT) at amino acid 520 (this study). Primer sequences and restriction polymorphisms to identify the alleles are listed in Table S1.

CLV1±GUS reporter An enhancer-trap screen utilizing the Ac/Ds transposon system identi®ed tissue-speci®c markers for embryogenesis (G. Martin and G. JuÈrgens, unpublished results). The CLV1±GUS reporter line 1054, which marks the shoot apical meristem and vasculature in seedlings, and contains the GUS gene inserted in the promoter of the CLV1 gene at position 16891 of BAC T4012, was crossed to cph-GX111 and fk-X224. GUS staining was performed as described by Sundaresan et al. (1995).

Sterol extraction and GC±MS conditions The plant material (0.9±1.2 g) was dried overnight at 55°C and extracted in acetone for 72 h. After evaporation in vacuo, the residue was partitioned between CHCl3 and water. The CHCl3soluble fraction was evaporated in vacuo, resuspended in nhexane, puri®ed with an IsoluteTM. SPE column (500 mg silica) and eluted twice with 7/3 n-hexane/ethyl acetate. The eluate was evaporated in vacuo, resuspended in 98/2 MeOH/water, puri®ed with a RP18 cartridge (Merck, Darmstadt, Germany) and eluted three times with absolute MeOH. Extracts were assayed by TLC (running buffer 95/5 CHCl3/MeOH). Acetylation of 0.2±1.3 mg of the puri®ed fraction was performed in 1 ml of 1/1 pyridine/acetic acid anhydride overnight, followed by washes with CHCl3 and evaporation in vacuo. Analysis of the residue was performed under the following conditions: GC±MS (Voyager/Trace GC 2000, Thermo Quest CE Instruments): 70 eV EI, source temperature 200°C; column DB-5MS (J & W, 15 m 3 0.25 mm, 0.25 mm ®lm thickness); injection temperature 250°C, interface temperature 300°C; carrier gas He, ¯ow rate 1.2 ml min±1, constant pressure mode; splitless injection, column temperature program: 170°C for 1 min, then raised to 270°C at a rate of 25°C min±1 and then to 290°C at a rate of 2°C min±1.

RT±PCR

15-Azasterol treatments

RNA was isolated from 150 mg of seedling material using the RNAeasy Mini Plant Kit (Qiagen, Hilden, Germany). cDNA synthesis and ampli®cation was performed using the OneStep RT± PCR Kit (Qiagen). Gene-speci®c primers were 5¢-CCCCACCACTCGAATTCATCTTTATCCTC-3¢ and 5¢-TGGTGAATAAGAAACTCTGCAACAAATCCG-3¢ for CPH, and 5¢-TCTAGTAGTTCCACAAACCACATTGG-3¢ and 5¢-AGACTCATGTTCATTTGCTATAACAGC-3¢ for HYD1.

Seeds from heterozygous CPH/cph, HYD1/hyd1 and FK/fk plants, and homozygous bri1-2, dwf1-2, dwf1-6, dwf5 and twd1 plants were germinated on 15-azasterol-containing half-concentrated MS medium (Murashige and Skoog, 1962) with 0.8% agar. 15Azasterol has a formula weight of approximately 500 g (Schmitt et al., 1980). A 1 mg ml±1 stock solution made in absolute ethanol served as 20 000 3, 2000 3 and 200 3 for 0.10, 1.0 and 10 mM concentrations, respectively.


72


Acknowledgements We thank Martin Souter, Keith Lindsey, Burkhard Schulz and Steve Clouse for communicating results prior to publication. The hyd1-1 and twd1 seeds were gifts from Martin Souter and Burkhard Schulz, respectively. The bri1-2, dwf1-6 and dwf5 seeds were from the Nottingham Arabidopsis Stock Centre. The dwf1-2 seeds were from the Arabidopsis Biological Resource Center. 15-Azasterol (A25822B azasterol) was provided by Lilly Research Laboratories (Green®eld, Indiana, USA). We thank Ariane Alvarez, Marion Bauch, Mikhail Fokin, Christina Heinlein, Alexei Kostigov, Gerd Schrick and Sophia von der Hardt for technical assistance, and Eva Benkova, Jiri Friml, Burkhard Schulz and Claus Schwechheimer for critical reading of the manuscript. This work was supported by the VolkswagenStiftung and a Leibniz Prize to G.J.

Supplementary Material The following material is available from http://www. blackwell-science.com/products/journals/suppmat/TPJ/ TPJ1333/TPJ1333sm.htm Table S1. PCR primers and conditions used to identify allele-speci®c polymorphisms.

References Altmann, T. (1999) Molecular physiology of brassinosteroids revealed by the analysis of mutants. Planta, 208, 1±11. Athenstaedt, K., Zweytick, D., Jandrositz, A., Kohlwein, S.D. and Daum, G. (1999) Identi®cation and characterization of major lipid particle proteins of the yeast. Saccharomyces cerevisiae. J. Bacteriol. 181, 6441±6448. Bach, T.J. and Benveniste, P. (1997) Cloning of cDNAs or genes encoding enzymes of sterol biosynthesis from plants and other eukaryotes: heterologous expression and complementation analysis of mutations for functional characterization. Prog. Lipid. Res. 36, 197±226. Bell, C.J. and Ecker, J.R. (1994) Assignment of 30 microsatellite loci to the linkage map of Arabidopsis. Genomics, 19, 137±144. Benveniste, P. (1986) Sterol biosynthesis. Annu. Rev. Plant Physiol. 37, 275±308. Bishop, G.J. and Yokota, T. (2001) Plant steroid hormones, brassinosteroids: current highlights of molecular aspects on their synthesis/metabolism, transport, perception and response. Plant Cell Physiol. 42, 114±120. Bloch, K.E. (1983) Sterol structure and membrane function. Crit. Rev. Biochem. 14, 47±91. Choe, S., Noguchi, T., Fujioka, S., Takatsuto, S., Tissier, C.P., Gregory, B.D., Ross, A.S., Tanaka, A., Yoshida, S. and Tax, F.E. (1999a) The Arabidopsis dwf7/ste1 mutant is defective in the D7 sterol C-5 desaturation step leading to brassinosteroid biosynthesis. Plant Cell, 11, 207±221. Choe, S., Dilkes, B.P., Gregory, B.D., Ross, A.S., Yuan, H., Noguchi, T., Fujioka, S., Takasuko, S., Tanaka, A., Yoshida, S., Tax, F.E. and Feldmann, K.A. (1999b) The Arabidopsis dwarf1 mutant is defective in the conversion of 24-methylenecholesterol to campesterol in brassinosteroid biosynthesis. Plant Physiol. 119, 897±907. Choe, S., Tanaka, A., Noguchi, T., Fujioka, S., Takatsuto, S., Ross, A.S., Tax, F.E., . Yoshida, S. and Feldmann, K.A. (2000) Lesions

in the sterol D7 reductase gene of Arabidopsis cause dwar®sm due to a block in brassinosteroid biosynthesis. Plant. J. 21, 431± 443. Clouse, S.D. (2000) Plant development: a role for sterols in embryogenesis. Curr. Biol. 10, R601±R604. Clouse, S.D. and Sasse, J.M. (1998) Brassinosteroids: essential regulators of plant growth and development. Annu. Rev. Plant Physiol. Plant Mol. Biol. 49, 427±451. Diener, A.C., Li, H., Zhou, W., Whoriskey, W.J., Nes, W.D. and Fink, G.R. (2000) STEROL METHYLTRANSFERASE 1 controls the level of cholesterol in plants. Plant Cell, 12, 853±870. Edwards, P.A. and Ericsson, J. (1999) Sterols and isoprenoids: signaling molecules derived from the cholesterol biosynthetic pathway. Annu. Rev. Biochem. 68, 157±185. Gachotte, D., Meens, R. and Benveniste, P. (1995) An Arabidopsis mutant de®cient in sterol biosynthesis: heterologous complementation by ERG3 encoding a D7-sterol-C-5desaturase from yeast. Plant J. 8, 407±416. Grebenok, R.J., Ohnmeiss, T.E., Yamamoto, A., Huntley, E.D., Galbraith, D.W. and Della Penna, D. (1998) Isolation and characterization of an Arabidopsis thaliana C-8,7 sterol isomerase: functional and structural similarities to mammalian C-8,7 sterol isomerase/emopamil-binding protein. Plant Mol. Biol. 38, 807±815. Hamann, T., Mayer, U. and JuÈrgens, G. (1999) The auxininsensitive bodenlos mutation affects primary root formation and apical-basal patterning in the Arabidopsis embryo. Development, 126, 1387±1395. Hartmann, M.-A. (1998) Plant sterols and the membrane environment. Trends Plant Sci. 3, 170±175. He, Z., Wang, Z.Y., Li, J., Zhu, Q., Lamb, C., Ronald, P. and Chory, J. (2000) Perception of brassinosteroids by the extracellular domain of the receptor kinase BRI1. Science, 30, 2360±2363. Husselstein, T., Schaller, H., Gachotte, D. and Benveniste, P. (1999) D7-sterol-C5-desaturase: molecular characterization and functional expression of wild-type and mutant alleles. Plant Mol. Biol. 39, 891±906. Jang, J.C., Fujioka, S., Tasaka, M., Seto, H., Takatsuto, S., Ishii, A., Aida, M., Yoshida, S. and Sheen, J. (2000) A critical role of sterols in embryonic patterning and meristem programming revealed by the fackel mutants of Arabidopsis thaliana. Genes Dev. 14, 1485±1497. Klahre, U., Noguchi, T., Fujioka, S., Takatsuto, S., Yokota, T., Nomura, T., Yoshida, S. and Chua, N.H. (1998) The Arabidopsis DIMINUTO/DWARF1 gene encodes a protein involved in steroid synthesis. Plant Cell, 10, 1677±1690. Marshall, J.A. and Nes, W.D. (1999) Isolation and characterization of an active-site peptide from a sterol methyl transferase with a mechanism-based inhibitor. Bioorg. Med. Chem. Lett. 9, 1533± 1536. Mayer, U., Torres-Ruiz, R.A., Berleth, T., Misera, S. and JuÈrgens, G. (1991) Mutations affecting the body organization in the Arabidopsis embryo. Nature, 353, 402±407. Mayer, U., BuÈttner, G. and JuÈrgens, G. (1993) Apical±basal pattern formation in the Arabidopsis embryo: studies on the role of the gnom gene. Development, 117, 149±162. Murashige, T. and Skoog, F. (1962) A revised medium for rapid growth and bioassays with tobacco tissue cultures. Physiol. Plant. 15, 473±497. Ourisson, G. (1994) Peculiarities of sterol biosynthesis in plants. J. Plant Physiol. 143, 434±439. Patterson, G.W., Hugly, S. and Harrison, D. (1993) Sterols and phytyl esters of. Arabidopsis thaliana under normal and chilling temperatures. Phytochemistry, 33, 1381±1383. ã Blackwell Science Ltd, The Plant Journal, (2002), 31, 61±73

Sterol biosynthesis genes in embryonic development Sakurai, A. and Fujioka, S. (1997) Studies on biosynthesis of brassinosteroids. Biosci. Biotechnol. Biochem. 61, 757±762. Schaeffer, A., Bronner, R., Benveniste, P. and Schaller, H. (2001) The ratio of campesterol to sitosterol that modulates growth in Arabidopsis is controlled by STEROL METHYLTRANSFERASE 2;1. Plant J. 25, 605±615. Schmitt, P., Scheid, F. and Benveniste, P. (1980) Accumulation of D8,14-sterols in suspension cultures of bramble cells cultured with an azasterol antimycotic agent (A25822B). Phytochemistry, 19, 525±530. Schrick, K. and Laux, T. (2001) Zygotic embryogenesis: the formation of an embryo from a fertilized egg. In: Current Trends in the Embryology of Angiosperms (Bhojwani, S.S. and Soh, W.Y., eds). Dordrecht, The Netherlands: Kluwer Academic Publishers, pp. 249±277. Schrick, K., Mayer, U., Horrichs, A., Kuhnt, C., Bellini, C., Dangl, J., Schmidt, J. and JuÈrgens, G. (2000) FACKEL is a sterol C-14 reductase required for organized cell division and expansion in Arabidopsis embryogenesis. Genes Dev. 14, 1471±1484. Schuler, I., Milon, A., Nakatani, Y., Ourisson, G., Albrecht, A.M., Benveniste, P. and Hartmann, M.A. (1991) Differential effects of

GenBank accession number AF494289 ® CPH mRNA, cds


73

plant sterols on water permeability and on acyl chain ordering of soybean phosphatidylcholine bilayers. Proc. Natl Acad. Sci. USA, 88, 6926±6930. Smith, D.L. and Fedoroff, N.V. (1995) LRP1, a gene expressed in lateral and adventitious root primordia of Arabidopsis. Plant Cell, 7, 735±745. Souter, M., Topping, J., Pullen, M., Friml, J., Palme, K., Hackett, R., Grierson, D. and and. Lindsey, K. (2002) The hydra1 mutants of Arabidopsis are defective in sterol pro®les and auxin and ethylene signaling. Plant Cell, 14, 1017±1031. Sundaresan, V., Springer, P., Volpe, T., Haward, S., Jones, J.D., Dean, C., Ma, H. and Martienssen, R. (1995) Patterns of gene action in plant development revealed by enhancer trap and gene trap transposable elements. Genes Dev. 9, 1797± 1810. Topping, J.F., May, V.J., Muskett, P.R. and Lindsey, K. (1997) Mutations in the HYDRA1 gene of Arabidopsis perturb cell shape and disrupt embryonic and seedling morphogenesis. Development, 124, 4415±4424. Wang, Z.Y., Seto, H., Fujioka, S., Yoshida, S. and Chory, J. (2001) BRI1 is a critical component of a plasma-membrane receptor for plant steroids. Nature, 410, 380±383.