Plant Cell, Vol. 11, 1911-1924, October 1999, Copyright © 1999, American Society of Plant Physiologists
A Defect in ß-Oxidation Causes Abnormal Inflorescence Development in Arabidopsis
Todd A. Richmond1,a and
Anthony B. Bleeckera
a Department of Botany, University of Wisconsin–Madison, Madison, Wisconsin 53706
Correspondence to:
Anthony B. Bleecker, bleecker{at}facstaff.wisc.edu (E-mail), 608-262-7509 (fax)
 |
ABSTRACT |
|---|
The abnormal inflorescence meristem1 (aim1) mutation affects inflorescence and floral development in Arabidopsis. After the transition to reproductive growth, the aim1 inflorescence meristem becomes disorganized, producing abnormal floral meristems and resulting in plants with severely reduced fertility. The derived amino acid sequence of AIM1 shows extensive similarity to the cucumber multifunctional protein involved in ß-oxidation of fatty acids, which possesses L-3-hydroxyacyl-CoA hydrolyase, L-3-hydroxyacyl-dehydrogenase, D-3-hydroxyacyl-CoA epimerase, and 
3, 2-enoyl-CoA isomerase activities. A defect in ß-oxidation has been confirmed by demonstrating the resistance of the aim1 mutant to 2,4-diphenoxybutyric acid, which is converted to the herbicide 2,4-D by the ß-oxidation pathway. In addition, the loss of AIM1 alters the fatty acid composition of the mature adult plant.
|
|
INTRODUCTION |
|---|
The ß-oxidation pathway includes the only enzymes that completely degrade fatty acids by the sequential removal of two carbon units, resulting in the formation of acetyl–coenzyme A (CoA). The process of ß-oxidation is common to all eukaryotic and prokaryotic organisms. There are two major systems of ß-oxidation, mitochondrial and peroxisomal. Both systems are present in animals, whereas ß-oxidation in fungi and plants appears to take place almost entirely in glyoxysomes and peroxisomes. Plants are able to degrade fatty acids completely within their peroxisomes, whereas animals require an additional mitochondrial ß-oxidation system because the peroxisomal ß-oxidation system seems to function only as a chain-shortening system (Tolbert 1981 
; Mannaerts and Debeer 1982 ; Gerhardt 1986 ). The ß-oxidation systems in animals have been well characterized both molecularly and biochemically (reviewed in Kunau et al. 1995 ; Eaton et al. 1996 ; Hashimoto 1996 ; Mannaerts and van Veldhoven 1996 ).
The discovery of a peroxisomal system of ß-oxidation was first made in plants (Cooper and Beevers 1969 ), and for many years, ß-oxidation research in plants has focused primarily on fatty tissues, including the mobilization of storage reserves from fatty tissues such as castor bean endosperm and the mesophyll of fat-storing cotyledons (Cooper and Beevers 1969 ; ap Rees 1980 ). Although the enzymes of the plant peroxisomal ß-oxidation systems have been well characterized biochemically (see Gerhardt 1992 ), it is not known whether the ß-oxidation system plays as important a role in plants as it does in animals. Loss of a single bifunctional protein acting in the ß-oxidation pathway in humans can cause multiple disease symptoms and is eventually fatal (Watkins et al. 1989 ).
We do not know what role ß-oxidation enzymes play in the normal development of plants. The abnormal inflorescence meristem1 (aim1) mutation is a novel mutation in a plant ß-oxidation enzyme that affects the mature plant. We present here the cloning and characterization of the AIM1 gene. We discuss the consequences of the absence of one of the key enzymes in the ß-oxidation pathway and the possible role of AIM1 in Arabidopsis development.
|
|
RESULTS |
|---|
Isolation of the aim1 Mutant
A set of transgenic lines constructed by Agrobacterium-mediated seed transformation of Arabidopsis thaliana ecotype Wassilewskija (Ws) (Feldmann 1991 ) was screened for inflorescence meristem mutants. The aim1 mutant was found segregating in a single F2 family. The inheritance of the aim1 mutation is recessive. For reasons described below, the aim1 mutant has severely reduced fertility and is primarily maintained as a heterozygote. The mutant was backcrossed more than five times before detailed characterization was initiated.
Phenotypic Characterization of the aim1 Mutant
The vegetative phase of development in the aim1 mutant is similar to that of the isogenic Ws ecotype when plants are grown under long-day conditions (>16 hr of daylight). The leaves of the vegetative rosette are often slightly smaller and darker green in aim1 plants, but the differences in phenotype at this stage are insufficient to score aim1 plants unambiguously in segregating populations. The difference in vegetative development between wild-type Ws (Figure 1A) and aim1 (Figure 1B) plants can be enhanced by growing them under short-day conditions (8 to 12 hr of daylight). Rosette diameter is then reduced two- to threefold, and the leaves often have a twisted appearance in aim1 homozygotes (Figure 1B). All leaves appear to be equally affected, regardless of age.
View larger version (96K):
[in this window]
[in a new window]
|
Figure 1.
Phenotypic Characterization of the aim1 Mutant.
(A) Wild-type Col rosette grown under short-day conditions (8 hr of light). Bar = 1 in.
(B) aim1 rosette in the Col background grown under short-day conditions (8 hr of light). Bar = 1 in.
(C) Close-up view of young developing aim1 inflorescence grown under long-day conditions (>16 hr of light).
(D) Wild-type Ws inflorescence.
(E) Mature inflorescence showing the most severe aim1-conferred phenotype.
(F) Close-up of the axis of a secondary branch from an AIM1 antisense plant.
(G) Wild-type Ws plant.
(H) aim1 plant.
(I) aim1 plant from same F2 population as the plant shown in (H), demonstrating the range of phenotypes seen in the mutant.
|
|
The aim1 mutant undergoes the transition to reproductive development at the same time as the wild-type plant. Shoot apical meristems at the transition from vegetative to reproductive development were harvested from plants that were grown under short days (8 hr of light) initially and then transferred to long days (16 hr of light) to induce flowering. Shoot apices of aim1 plants harvested before the transition to flowering as well as 4 days after the transition were indistinguishable from wild-type vegetative meristems when examined by scanning electron microscopy (data not shown).
It is after this point that the mutant phenotype becomes most pronounced. The primary inflorescence meristem of the wild-type Ws plant produces two or three aerial secondary inflorescence branches subtended by cauline leaves, with each separated by elongated internodes, and then produces 30 to 40 flowers. Each of the secondary meristems produces two or three vegetative nodes before converting to a flower-producing meristem. Secondary inflorescence branches appearing from the rosette, called basal secondary branches, develop in the same way. The result is a highly branched set of flower-bearing stems that is typical of Arabidopsis (Figure 1G).
The aim1 plant initiates the same basic pattern of inflorescence development as the wild type (Figure 1D). However, the conversion from leaf-producing meristems to flower-producing meristems is severely affected in the aim1 mutant (Figure 1C). Shoot apices were harvested from long-day plants when the primary inflorescence shoot was 3 to 5 cm above the basal rosette. At this time, wild-type apices displayed a typical spiral array of developing flower buds at various stages of development (Figure 2A). By contrast, a wide range of abnormalities was observed in inflorescence meristems from aim1 plants (Figure 2B to 2H). Inflorescence branches show a range of phenotypes. In the most severe cases, the inflorescence meristem does not produce any recognizable floral structures but rather produces only a small mass of undifferentiated tissue that subsequently ceases to develop all together. More often, the inflorescence meristems will produce a few to several floral structures in what appears to be a normal spiral pattern before terminating (Figure 1E). These inflorescence meristems terminate in either flowerlike structures or undifferentiated masses of tissue. Inflorescence meristems harvested from aim1 plants grown under short-day conditions were more similar to wild type (Figure 2I), although the flower buds produced by these meristems were clearly abnormal (Figure 2J).
View larger version (194K):
[in this window]
[in a new window]
|
Figure 2.
Examination of Later-Stage Inflorescence Meristems under Long-Day and Short-Day Conditions.
(A) Wild-type inflorescence meristem with floral buds ranging from stages 1 to 6 (stages following Smyth et al. 1990 ).
(B) Long-day aim1 inflorescence. The inflorescence terminates in structures that are unlike any of the normal floral stages.
(C) Long-day aim1 inflorescence. Although several floral buds have been initiated, the meristem is abnormal, unorganized, and more rounded than that of the wild type.
(D) Four floral buds produced by an aim1 inflorescence.
(E) Terminally differentiated aim1 inflorescence. The flower in the foreground shows an abnormal, lobed sepal. The flower in the rear has a filamentous structure commonly seen in aim1 flowers.
(F) Long-day aim1 inflorescence.
(G) Close-up of a wild-type inflorescence meristem grown under long-day conditions.
(H) Close-up of a long-day aim1 inflorescence meristem (M). The 2 denotes a stage 2 floral bud.
(I) Wild-type inflorescence meristem grown under short-day conditions.
(J) Short-day aim1 inflorescence meristem. Stamen primordia (white arrowheads) are indicative of stage 5 flowers, yet the sepals (black arrows) have not yet covered the floral meristem, a stage 4 characteristic.
(K) Wild-type flower.
(L) aim1 flower.
(M) aim1 flower. The gynoecium is flattened and is V-shaped at the top (arrow).
(N) Young aim1 flower showing delayed or arrested development.
(O) Gynoecium from a long-day aim1 flower.
(P) Terminal aim1 flower. There are filaments tipped with stigmatic tissue (white arrowheads), a deformed carpal, and stigmatic papillae visible on the gynoecium (black arrow).
Bars in (A) to (P) = 100 µm.
|
|
The basal secondary branches of the aim1 mutant are much the same as those of the primary ones, although many more secondary branches are initiated in aim1 mutants than in the wild type. A single aim1 plant may have 10 to 200 higher order inflorescences in the Ws (Figure 1H to 1I) and Columbia (Col) backgrounds but fewer in the Landsberg erecta background. Multiple third-, fourth-, or fifth-order branches arise nested in the axils of the cauline leaves on these secondary branches, often with little or no internode elongation (Figure 1F). It could not be determined whether the final structures in this reiterated pattern are inflorescence meristems that have deteriorated to the point at which they can no longer produce additional inflorescences or are determinate floral meristems that have failed to develop any normal floral organs. Increased branching, such as that observed with the aim1 mutant, is often associated with reduced fertility or sterility (Hensel et al. 1994 ).
Defects in Flower Development Caused by the aim1 Mutation
The aim1 mutant does not produce any normal, fertile flowers like those of the wild type (Figure 2K). Figure 2 shows the wide variation in appearance of the flowers and flowerlike structures in the aim1 mutant. Under long-day conditions, there are often many floral organs missing from each flower, and there are often homeotic conversions (Figure 2O and Figure 2P). In general, flowers from aim1 plants grown under short-day conditions are much more normal in appearance, although the plant itself is still basically sterile. If allowed to grow for extended periods of time (12 to 16 weeks), a small amount of seed (50 to 100 seeds) can be collected from aim1 plants. Although this seed is darker in color than is seed of the wild type and is variable in size and shape, it is viable. The production of small amounts of seed from aim1 plants shows that there is not an absolute block in either male or female fertility.
Two broad classes of flower types can be distinguished. In the first broad class, which represents ~60% of the floral structures when the plants are grown under long-day conditions, sepal growth appears arrested, leaving the interior of the flower exposed (Figure 2J). In the other broad class, sepal growth is mostly normal, but the interior of the flower shows arrested or delayed development (Figure 2N). Table 1 shows the percentage of each type of organ found in aim1 flowers of this class. There are often multiple stamens missing (Figure 2M), or stamens are arrested at an early stage of development (Figure 2J and Figure 2N). Sepals and the gynoecium are the least affected in the mutant, whereas petals and stamens are often missing or abnormal (Figure 2L), in agreement with the floral development model that points to carpels and sepals as the ground state (Haughn and Somerville 1988 ; Meyerowitz et al. 1991 ). Approximately 10% of this type of aim1 flower has the normal complement of floral organs yet is still sterile. The primary defect appears to be in pollen development, because few mature pollen grains were seen during flower dissection.
Molecular Cloning of the AIM1 Gene
To ensure that the phenotype of the aim1 mutant was the result of the T-DNA inserting in a gene and not from another mutation caused by the transformation process, cosegregation analysis was performed. Individual mutants from an F2 population were examined by DNA gel blot analysis, using the right border of the T-DNA as a probe. Twenty-four of 24 individual mutants examined contained the T-DNA (data not shown), giving a <1% chance that the T-DNA insertion assorts independently of the mutation. The gene adjacent to the T-DNA insert contains 18 exons with an average size of 120 bp and 17 introns with an average size of 152 bp. The T-DNA insertion is located 194 bp 5' of the predicted start codon. This gene encodes a protein of 721 amino acids, with a predicted size of 77.9 kD.
BLAST (Altschul et al. 1990 ) searches using sequences from genomic subclones showed an identical match of ~500 bp to the 3' end of two overlapping Arabidopsis expressed sequence tags (ESTs) placed in the dEST database (Hofte et al. 1993 ).
Verification That the Phenotype of the aim1 Mutant Is Due to the Disruption of the Gene Associated with the T-DNA Insertion
Three experiments were performed to confirm that loss of expression of the gene associated with the T-DNA insert was responsible for the phenotype of the aim1 mutant. A fragment adjacent to the tag was used to probe total RNA isolated from wild-type and aim1 inflorescences. A 2.5-kb message could be found in wild-type plants but not in the aim1 mutant (Figure 3A). Second, wild-type genomic sequences encompassing the gene were cloned into a vector containing hygromycin resistance and used to transform a line that was heterozygous for aim1. A T3 population was found that was homozygous for both hygromycin and kanamycin resistance and failed to segregate for the aim1 mutation, indicating that the wild-type genomic sequence can complement the aim1 lesion. Finally, wild-type plants were transformed with an antisense construct composed of the coding sequence of the gene of interest fused in reverse orientation to the cauliflower mosaic virus 35S promoter. Twelve of 24 transgenic lines generated with this construct displayed phenotypes similar to the original phenotype of the aim1 mutant (Figure 1F). Taken together, these results provide compelling evidence that the phenotype of the aim1 mutant is due to the loss of expression of the gene associated with the T-DNA insert. Consequently, we refer to this gene as the AIM1 gene.
View larger version (57K):
[in this window]
[in a new window]
|
Figure 3.
RNA Gel Blot Analysis of the AIM1 Gene.
(A) RNA gel blot of total RNA (5 µg) from wild-type (Ws) and aim1 inflorescences, probed with DNA flanking the T-DNA insert.
(B) Time course of AIM1 expression in developing seedlings. rRNA represents the 45S rRNA precursor used as a control for equal loading. Lanes are labeled by age (in days) of tissue. Cont., continuous; L, leaf.
(C) RNA gel blot analysis of AIM1 RNA levels, using total RNA (5 µg). R, root; L, rosette leaf; S, stem; C, cauline leaf; Si, silique; F, flower.
(D) Comparison of AIM1 and AtMFP2 RNA levels in various mature tissues. Lanes are labeled as given above.
|
|
Mapping of AIM1 and the Multifunctional Protein Gene AtMFP2
The AIM1 gene was mapped to chromosome 4, position 72.4, using recombinant inbred lines (Lister and Dean 1993 ). The AtMFP2 gene was mapped to the top of chromosome 3, 4.9 centimorgans from marker g4119, using a segregating population of F5 families (Chang et al. 1993 ).
The Predicted AIM1 Protein Is Similar to MFPs Involved in ß-Oxidation
A comparison of the predicted AIM1 amino acid sequence with sequences in the current protein databases (NCBI; BLAST network server) shows that it exhibits significant similarity to the cucumber multifunctional protein (CuMFP) (accession number CAA55630) acting in glyoxysomal fatty acid ß-oxidation. CuMFP is a 79.0-kD plant glyoxysomal protein that has been shown to possess L-3-hydroxyacyl-CoA hydrolyase (enoyl-CoA hydratase; EC 4.2.1.17), L-3-hydroxyacyl-dehydrogenase (HDH; EC 1.1.1.211), D-3-hydroxyacyl-CoA epimerase, and 3, 2-enoyl-CoA isomerase (EC 5.3.38) activities (Preisig-Muller et al. 1994 ). The predicted AIM1 gene product is 56% identical to CuMFP across the entire length of the protein. In addition, searches of the EST database using the predicted protein sequence revealed a second MFP gene in Arabidopsis. Full-length cDNAs were isolated and sequenced for this gene, which we have designated as AtMFP2. AtMFP2 is more closely related to CuMFP than to AIM1, showing 76% identity. AIM1 is as similar to AtMFP2 to as it is to CuMFP (57% identity). A number of other proteins in the database also show similarity to AIM1. These proteins also function in the ß-oxidation pathway in a wide variety of organisms, including humans, rats, fungi, and bacteria.
Based on enzymatic evidence from purified CuMFP (Preisig-Muller et al. 1994 ), the AIM1 gene product is expected to have an isomerase domain in the first 125 amino acids of the protein, a hydratase domain between amino acids 100 and 200, and a dehydrogenase domain between amino acids 390 and 590 (Figure 4A). An epimerase activity has been localized to the N-terminal end of CuMFP, but it has not been delimited explicitly.
View larger version (46K):
[in this window]
[in a new window]
|
Figure 4.
Amino Acid Sequence Comparison of AIM1 with Other ß-Oxidation Enzymes.
For alignments, the PROSITE consensus pattern is underlined, and the asterisks indicate residues where AIM1 does not match the PROSITE consensus. The residues shaded in black are those that are identical in all proteins.
(A) Domain structure of various multifunctional ß-oxidation proteins (from Preisig-Muller et al. 1994 ). Boxes represent known domains. Protein sequences are as follows, with accession numbers given in parentheses: CuMFP, cucumber multifunctional protein (CAA55630); EcFADB, E. coli FADB protein (CAB40809); HuPBE, human peroxisomal bifunctional enzyme (Q08426).
(B) Alignment of the PROSITE consensus region (PDOC00150) of the isomerase/hydratase domain of the AIM1 protein. The symbol x is used for a position at which any amino acid is accepted. Ambiguities are indicated by listing the acceptable amino acids for a given position between square brackets. Repetition of an element within the pattern is indicated by following that element with a numerical value within parentheses. Proteins are as given above, with the addition of AtMFP2, an Arabidopsis AIM1 homolog.
(C) Alignment of the PROSITE consensus region (PDOC00065) of the dehydrogenase domain of the AIM1 protein.
|
|
Figure 4B shows an alignment of the region surrounding the isomerase/hydratase PROSITE (Bucher and Bairoch 1994 ) consensus sequence of the AIM1 gene product with proteins known to have hydratase activity, including CuMFP, the human peroxisomal bifunctional enzyme, and the large 
subunit of the Escherichia coli fatty acid oxidation complex (FADB). The AIM1 protein differs from the PROSITE consensus sequence for the isomerase/hydratase superfamily at only one amino acid residue. However, the nonconserved alanine in AIM1 (A123) indicated in Figure 4B is shared by all of the putative plant MFPs (data not shown), suggesting that plant MFPs may have a different isomerase/hydratase consensus than the PROSITE pattern.
Figure 4C provides an alignment of the region surrounding the dehydrogenase PROSITE consensus sequence. The AIM1 gene product differs from the PROSITE consensus sequence for HDH in three places, but at each residue, CuMFP has the same nonconserved change. Because CuMFP has been shown to have HDH activity, it is likely that AIM1 has HDH activity as well, although this has not been demonstrated.
Enoyl-CoA Hydratase Activity of the AIM1 Gene Product
To verify that AIM1 and AtMFP2 have the enzymatic activities suggested by their amino acid sequences, both were expressed as fusions to glutathione S-transferase in a bacterial overexpression system (rAIM1 and rAtMFP2). Enoyl-CoA hydratase activity was found in both proteins. The Km values for both rAIM1 (115 µM) and rAtMFP2 (240 µM) are similar to those reported for the CuMFPs. It appears that AIM1 has a higher affinity for short-chain acyl-CoAs than does AtMFP2, although definitive proof would require purification of the enzymes from plant tissue. These data confirm that both AIM1 and AtMFP2 have an enzymatic activity associated with ß-oxidation of fatty acids.
Fatty Acid Composition
Lack of a functioning ß-oxidation system may result in an accumulation of long-chain fatty acids (LCFAs). Table 2 shows the fatty acid composition of total lipids from mature wild-type and aim1 rosette leaves. As expected, there are changes in the fatty acid composition. In particular, there are changes in unsaturated 18-carbon fatty acids, including significantly elevated levels of C18:2 and C18:1.
Germination Rate in the aim1 Mutant
ß-Oxidation is known to play an important role in mobilizing seed lipid reserves in germinating seedlings. To test whether the aim1 mutation affects seed germination in any way, seedlings were grown on half-strength Murashige and Skoog (MS) plates (Murashige and Skoog 1962 ) with no additional carbon source. As a recessive mutation, one-quarter of the seed from the heterozygous parent would be expected to fail to germinate or show poor seedling growth. Seed from heterozygous plants was plated with wild-type controls on half-strength MS plates with and without 2% sucrose, in both the light and the dark. No difference was seen in seed germination or seedling growth under any conditions tested.
However, when the rare seeds from aim1 homozygous plants were tested, a different result was found. When plated on half-strength MS media without sucrose, none of the seeds germinated (<0.1%). However when plated on half-strength MS media plus 2% sucrose, 12% of the seed germinated. Because aim1 homozygotes from an aim1 heterozygous parent show no dependence on sucrose for germination, a maternal effect of the aim1 mutation on seed germination is indicated.
The aim1 Mutant Is Resistant to the Herbicide 2,4-Diphenoxybutyric Acid
A diminished capacity for ß-oxidation in aim1 seedlings was demonstrated using the herbicide 2,4-diphenoxybutyric acid (2,4-DB). In the presence of a functioning ß-oxidation system, nontoxic 2,4-DB is converted to toxic 2,4-D (Synerholm and Zimmerman 1947 ; Garraway 1970 ; Hayashi et al. 1998 ). Seeds from an aim1 heterozygote were grown vertically on plates using a range of concentrations from 0.1 to 100 µM 2,4-DB. aim1 seedlings are resistant to 2,4-DB, which is consistent with a defect in ß-oxidation. Table 3 shows that there is a small window of concentrations at which aim1 exhibited resistance to 2,4-DB. At concentrations >4 µM, root growth was completely inhibited in both wild-type seedlings and the aim1 segregating population. At 2 µM 2,4-DB, however, one-quarter of the population segregating for aim1 showed elongated roots, and when transplanted to soil, all of the seedlings with the elongated roots had the aim1 phenotype. This result demonstrates that ß-oxidation in vegetative tissues is diminished in the aim1 mutant.
Expression Pattern of the AIM1 Gene
The expression of the AIM1 gene in mature plants was examined by both RNA gel blot analysis and by using an AIM1–ß-glucuronidase (GUS) reporter construct. RNA was isolated from various tissues of mature Arabidopsis plants and probed using the full-length AIM1 cDNA. The AIM1 transcript proved to be expressed at approximately equal levels in all tissues examined (Figure 3C). Only a single transcript was detected, in accordance with DNA gel blot data indicating that AIM1 is a single-copy gene (data not shown). Expression of AtMFP2 was examined in roots, rosette leaves, flowers, and siliques and compared with that of AIM1. In each tissue examined, AIM1 and AtMFP2 were expressed at approximately equal levels (Figure 3D). Expression of AIM1 was examined in developing seedlings grown in both the light and the dark. Only faint expression was seen in etiolated seedlings during days 3 through 6. Strong expression was detected only in older seedlings (>8 days) grown in continuous light (Figure 3B). However, full-length cDNAs were isolated from a cDNA library made from 3-day-old dark-grown seedlings, indicating that both AIM1 and AtMFP2 are expressed at this stage of development, although expression may be low.
To examine the pattern of AIM1 expression in vegetative, inflorescence, and floral meristems, the putative AIM1 promoter region was placed upstream of the GUS reporter gene in the Agrobacterium binary vector pBI101.2 (Jefferson 1987 ) to create a chimeric AIM1–GUS gene. Multiple transformants were obtained and three lines were chosen for further characterization because of the range of staining that they exhibited. All show the same pattern of expression but differ in the time it takes for GUS staining to develop fully. AIM1–GUS seedlings show GUS staining within the first 24 hr after germination. There is high expression in the cotyledons and at the hypocotyl–root junction (Figure 5A and Figure 5B). The lack of staining in the midsection of the hypocotyl may be due to problems with penetration of the substrate as opposed to lack of expression, because other seedlings show darker staining in the hypocotyl. There is faint staining in the roots. The expression pattern was evaluated in both light- and dark-grown seedlings and was shown to be similar (data not shown). A rosette leaf (Figure 5E), stem, cauline leaf, inflorescence meristem, and flowers (Figure 5F) from a mature plant all stain deeply, indicating strong expression in these tissues as well. Roots from mature plants stain faintly, consistent with a lower level of expression detected by RNA gel blots.
View larger version (75K):
[in this window]
[in a new window]
|
Figure 5.
Histochemical Localization of GUS Expression in Plants Transformed with AIM1–GUS.
(A) Seven-day-old seedling grown in continuous light.
(B) Two-micron section from a 3-day-old seedling grown in continuous light. Staining, represented by the red crystals, is evident in all cell layers.
(C) Developing silique from a 4-week-old plant.
(D) Close-up of a 2-µm section from a cross-section of a mature silique.
(E) Rosette leaf from a 4-week-old plant grown under 16-hr days.
(F) Inflorescence from a 4-week-old plant grown under 16-hr days.
(G) Close-up of a 2-µm section from a developing inflorescence.
(H) In situ hybridization on mature inflorescence from a 4-week-old wild-type plant. There was no expression seen with a sense control probe (data not shown).
|
|
Two-micron sections of GUS-stained tissue were examined using dark-field optics, with GUS expression evident as pink crystals. Examination of 3-day-old seedlings shows that the AIM1 promoter directs GUS protein expression in all cell types of the developing seedling (Figure 5B). Developing siliques and embryos also show expression (Figure 5C and Figure 5D). There is staining in the interior of the silique near the developing embryos, although the mature embryos show no evidence of GUS activity.
GUS staining is observed in the inflorescence meristem and in the earliest floral buds as well as in the mature flowers (Figure 5F and Figure 5G). Stained inflorescences were sectioned and expression was seen in all cell types in the meristem and flowers, with particularly intense staining in the developing anthers (Figure 5G). To confirm the GUS reporter gene data, in situ hybridization was used to examine AIM1 gene expression in the inflorescence meristem. Figure 5H shows expression of AIM1 throughout the inflorescence meristem and in the developing flower buds, consistent with the GUS data. No expression was detected when a sense control probe was used.
|
|
DISCUSSION |
|---|
Developmental Context of the Phenotype of the aim1 Mutant
The AIM1 gene encodes an MFP involved in the ß-oxidation of fatty acids. AIM1 appears to be expressed in most tissues, and phenotypic effects of loss of function of this gene are detectable in seedlings and in the vegetative rosette. However, these effects on vegetative development are relatively minor. It is after the transition to reproductive development that the lack of AIM1 function has debilitating effects on growth and development.
We found a wide range of morphological phenotypes associated with reproductive development in the aim1 mutant. Many of these phenotypes overlap with or are reminiscent of phenotypes produced by other known mutations that affect reproductive development in Arabidopsis. Mutations in floral organ identity genes may result in lack of development of organs within specific whorls of the flower and in homeotic conversions of organ identity in specific whorls (Weigel and Meyerowitz 1994 ; Weigel and Clark 1996 ). However, individual genes in this class affect primarily two adjacent whorls in the flower, whereas the mutation in aim1 may affect any of the four whorls. Homeotic conversions were observed in aim1, but these were generally restricted to structures arising directly from the inflorescence meristem. When organs were produced in floral meristems of aim1, the organ identity was appropriate to the whorl. The chimeric structures observed at shoot apices of aim1 may be a pleiotropic effect of floral programs invading the inflorescence meristem. Such structures are commonly observed at the inflorescence apices of sterile plants (Hensel et al. 1994 ).
Biochemical Context of the aim1 Mutation
The AIM1 protein exhibits extensive similarity to enzymes responsible for conducting the process of fatty acid ß-oxidation. We have established that AIM1 has enoyl-CoA hydratase activity, characteristic of an enzyme involved in ß-oxidation. In addition, the inability of the aim1 mutant to properly metabolize 2,4-DB to 2,4-D supports the theory that AIM1 plays a role in the normal fatty acid metabolism of the plant via the ß-oxidation pathway. This theory is supported by the altered fatty acid composition of the aim1 mutant.
A total of four MFPs have been purified and characterized from cucumber (Behrends et al. 1988 ; Guhnemann-Schafer and Kindl 1995a , Guhnemann-Schafer and Kindl 1995b ), although only one has been cloned and sequenced (Preisig-Muller et al. 1994 ). Three of these proteins have been isolated from cotyledons (MFPI to MFPIII), whereas the fourth (MFPIV) was isolated from leaf tissue. The enzymatic activities of these four proteins differ; MFPI and MFPIV have no isomerase activity. In addition, the hydratase activity of these four proteins shows chain-length specificity; MFP II shows no activity toward C18:1 fatty acids, whereas MFPIV shows a higher activity toward shorter chain fatty acids (Guhnemann-Schafer and Kindl 1995a ). Our work supports the presence of at least two MFPs in Arabidopsis; AtMFP2 is more closely related to CuMFP (Preisig-Muller et al. 1994 ) than is AIM1.
In plants, ß-oxidation of fatty acids provides energy and/or substrates for gluconeogenesis. Gluconeogenic ß-oxidation is particularly important during seed germination, in which lipid reserves must be mobilized by conversion to sugars. Indeed, there are reported examples of deficiencies in ß-oxidation that affect seed germination. Hayashi et al. 1998 devised a selection scheme for ß-oxidation mutants using resistance to the herbicide 2,4-DB. Four of the 12 mutants they isolated show sucrose-dependent germination. This emphasizes the importance of ß-oxidation in mobilizing seed lipid reserves in germinating seedlings. The lack of an adult phenotype in any of these mutants suggests that there may be separate systems for mobilizing lipid reserves and for normal housekeeping function and/or senescence-associated remobilization of lipids from leaves.
In contrast, the loss of AIM1 does not result in a sucrose-dependent germination phenotype. This may be due to functional redundancy among the MFPs in the developing seedling. It may be that there are other enzymes primarily responsible for mobilizing seed lipid reserves in the young seedling, whereas AIM1 is responsible for conducting ß-oxidation in nonfatty tissues. However, the activity of AIM1 is not superfluous at this early developmental stage because aim1 seeds show reduced sensitivity to 2,4-DB; it is possible that a higher affinity of AIM1 for short-chain fatty acids as compared with AtMFP2 allows a small window of resistance for aim1 seedlings.
Reconciling the Developmental Phenotype of aim1 with the Biochemical Defect in ß-Oxidation
We can consider three different hypotheses that account for the developmental phenotype of the aim1 mutant. (1) ß-Oxidation provides energy and reducing power from lipid reserves that are specifically needed during reproductive development. (2) ß-Oxidation is used to produce a lipid metabolite(s) that plays a structural or signaling role in reproductive development. (3) The specific block in ß-oxidation in aim1 results in the buildup of a lipid metabolite that disrupts intercellular communication in the reproductive meristems.
The first hypothesis is consistent with the notion that reproductive development has a higher energy demand associated with it than vegetative development. A precedent for this idea is provided by forms of cytoplasmic male sterility in maize (Levings 1993 ). In the cms line, a specific defect in mitochondrial function in all tissues of the plant has a noticeable effect only on pollen development, ostensibly because of the higher energy demand for male gametophyte development (Levings 1993 ). Although AIM1 is expressed in the reproductive meristems, it is unlikely to function cell autonomously in energy generation in meristems because no lipid reserves are present in these tissues. On the other hand, senescence of rosette leaves occurs concomitantly with the transition to reproductive development (Hensel et al. 1994 ); thus, failure to convert lipids stored in leaf tissues during senescence could affect the energy budget of the plant during this phase of development. However, observation of vegetative growth is inconsistent with this hypothesis. Vegetative development, particularly of stem and cauline leaf tissue, is quite vigorous in the mutant. In fact, at 12 weeks after germination, the aim1 meristems produced more total biomass than did their wild-type counterparts (data not shown).
The hypothesis that aim1 is blocked in the production of a lipid signaling molecule is intriguing. However, so little information is available as to how intercellular signaling is used to set up morphogenic fields in plant meristems that the merits of this hypothesis are difficult to evaluate. Most identified genes that affect pattern formation in reproductive meristems are presumptive transcription factors that are likely to act cell autonomously (Weigel and Meyerowitz 1994 ; Weigel and Clark 1996 ). However, the recent discovery that the terminal flower gene is related to lipid binding proteins is consistent with the possibility that lipid signaling may play a role in patterning associated with plant morphogenesis (Bradley et al. 1997 ). In this regard, there are a number of fatty acid–derived signals in plants: jasmonic acid, the traumatin family and related alkenals (Farmer 1994 ), highly oxygenated fatty acid derivatives reminiscent of animal lipoxins (Farmer 1994 ), and lipooligosaccharides (Spaink et al. 1991 ). Fatty acids also have been shown to modulate protein kinases in plants (Scherer 1996 , and references therein), which could provide a connection between aim1 and the phenotypically similar tousled (tsl) mutant that results from loss of function of a protein kinase (Roe et al. 1993 ).
Perhaps the most plausible hypothesis to account for the phenotypic effects of aim1 is that a block in ß-oxidation leads to the accumulation of intermediate or unique lipid metabolites. These might disrupt the process of reproductive development in a variety of ways. In humans, peroxisomal bifunctional enzyme deficiency results in multiple biochemical abnormalities, including elevated levels of very long chain fatty acids (VLCFAs) in both plasma and fibroblasts, impaired ß-oxidation in cultured fibroblasts, and abnormal bile acid metabolism (Watkins et al. 1989 ). Accumulation of VLCFAs can have dramatic effects on plant morphology (Millar et al. 1998 ). Although there is no accumulation of VLCFAs in the aim1 mutant, there are significant changes in LCFAs that warrant further investigation.
Concluding Statement
There is much work to be done with aim1 to determine the link between ß-oxidation and normal plant development. There are at least two MFPs involved in ß-oxidation in Arabidopsis. Loss of function of one of these genes results in abnormal inflorescence development. The aim1 mutation affects normal patterning at the inflorescence meristem and the maintenance of cell growth, as is apparent by the delayed development seen when examining flowers. Although the fatty acid composition of the mutant is altered, there is no clear indication of how this affects plant development. The link between ß-oxidation and development has been established but not elucidated. However, this link holds the promise of establishing a new role for lipid metabolism in regulating plant growth and may lead to the discovery of new lipid signaling molecules.
|
|
METHODS |
|---|
Growth of Plants
Plants (Arabidopsis thaliana) were grown as previously described (Hensel and Bleecker 1992 ; Hensel et al. 1993 ). Seeds were surface sterilized with 30% bleach/0.5% Triton X-100 and stratified 2 to 5 days at 4°C before planting. Kanamycin-resistant seedlings were selected on half-strength Murashige and Skoog (MS) media (Murashige and Skoog 1962 ) containing 100 µg/mL kanamycin sulfate, plus 2% sucrose where noted. Stock solutions of 2,4-diphenoxybutyric acid (2,4-DB; Sigma) were prepared in dimethyl sulfoxide.
Isolation of the AIM1 Gene
DNA was prepared from aim1 heterozygous plants, using a hexadecyltrimethylammonium bromide protocol (Murray and Thompson 1980 ). A genomic library was constructed and screened using standard techniques (Sambrook et al. 1989 ). Wild-type genomic clones were isolated from a Landsberg erecta library (Arabidopsis Biological Resource Center [ABRC], Columbus, OH; Voytas et al. 1990 ). A size-selected cDNA library (ABRC; Kieber et al. 1993 ) was screened for full-length clones.
DNA Sequencing
Sequencing of larger clones was performed by making serial deletions using an ExoIII nuclease–based procedure (Erase-A-Base; Promega, Madison, WI). cDNA clones were sequenced on both strands, whereas single-pass one-strand sequencing was used with genomic clones to identify intron/exon boundaries. The DNA sequences for AIM1 and AtMFP2 have been deposited in Genbank as accession numbers
AF123253 and
AF123254, respectively.
DNA and RNA Gel Blot Analyses
RNA was isolated using the protocol of Chomczynski and Sacchi 1987 . Five micrograms of total RNA was loaded on 0.8% agarose gels with 1.1% formaldehyde. RNA gel blot hybridization was performed at 68°C in modified Church buffer (1 mM EDTA, 0.25 M Na2PO4, 1.0% casein, and 7.0% SDS, pH 7.4). DNA gel blot hybridization was performed using standard techniques (Sambrook et al. 1989 ).
ß-Oxidation Enzyme Assays
Oligonucleotides were designed to construct a full-length AIM1 protein coding sequence suitable for overexpression in a bacterial host system. A BamHI site was engineered 5' of the AIM1 start codon to place it in frame in pGEX4T-1, a glutathione S-transferase fusion vector (Pharmacia). Protein expression was induced with 0.1 mM IPTG, and the fusion protein was purified by batch method using glutathione–agarose beads. A NotI fragment from an AtMFP2 cDNA clone, containing the entire coding sequence, was excised and ligated into pGEX4T-2 and used to produce recombinant protein as described above. The resulting proteins were >95% pure (data not shown). Hydratase activity was measured by following the decrease in absorbance at 263 nm due to the hydration of the double bond of crotonyl-CoA (Binstock and Schulz 1981 ). Kinetic parameters (Vmax and Km) were estimated using nonlinear regression on data collected (n = 2) using crotonyl-CoA concentrations ranging from 5 to 200 µM.
Fatty Acid Methyl Ester Analysis
Fatty acid composition of total lipids was determined as described by Browse et al. 1985 .
Construction of the Cauliflower Mosaic Virus 35S–Antisense AIM1 Gene Construct
The ß-glucuronidase (GUS) cassette from the plant transformation vector pBI121 (Clontech, Palo Alto, CA) was excised using BamHI and SacI and replaced with a full-length AIM1 cDNA clone in the antisense orientation. This construct was transformed into Agrobacterium tumefaciens GV1301 and introduced into Arabidopsis ecotype Wassilewskija (Ws) by using the vacuum infiltration protocol of Bechtold et al. 1993 .
Construction of the AIM1–GUS Chimeric Gene and Histochemical Localization of GUS
The putative AIM1 promoter was amplified using a 5' upstream primer and a 3' primer with an engineered BamHI site. By using a native XhoI site and the engineered BamHI site, a 2-kb fragment containing the first 30 amino acids of the AIM1 protein was ligated into pBI101.2 (Clontech), resulting in an in-frame fusion with the GUS reporter gene. This construct was transformed into Arabidopsis, as described above.
Histochemical localization of GUS was performed as described by Craig 1992 . For light microscopy, the procedure below was followed, and tissue sections were visualized using dark-field optics.
Light and Electron Microscopy
All shoot apices were harvested when the primary shoot was >3 cm in length. For light microscopy, samples were fixed overnight in 4% paraformaldehyde or in FAA (4% paraformaldehyde, 50% ethanol, and 5% acetic acid) and taken through a graded ethanol series to 100%. Samples embedded in London Resin (LR) White (Electron Microscopy Sciences, Fort Washington, PA) were taken through a graded LR White/ethanol series with several changes at 100% LR White. Resin was polymerized according to the manufacturer's recommendations. Samples embedded in wax (Paraplast Plus; Oxford Labware, St. Louis, MO) were prepared as described by Hensel et al. 1994 . Samples for electron microscopy were fixed in FAA and further prepared as described by Hensel et al. 1994 . In situ hybridization was performed as described by Lincoln et al. 1994 .
|
|
FOOTNOTES |
|---|
1 Current address: Department of Plant Biology, Carnegie Institution of Washington, Stanford, CA 94305. 
|
|
ACKNOWLEDGMENTS |
|---|
This work was funded by grants from the National Science Foundation (Grant No. DMB-9005164), National Institute of Aging (Grant No. 5F32AGO5542-02), and the U.S. Department of Energy/National Science Foundation/U.S. Department of Agriculture Collaborative Research in Plant Biology Program (Grant No. BIR92-20331). T.A.R. has been supported by a National Science Foundation Graduate Fellowship, a University of Wisconsin–Madison Genetics Department Training Grant (National Institutes of Health Grant No. GM07133-181), and a Wisconsin Alumni Research Foundation Graduate Fellowship. Special thanks to Sara Patterson, who helped with the sectioning and the microscopy, and to Michelle Nelson, who supplied several of the scanning electron microscope (SEM) photographs. We are grateful to Claudia Lipke for photographic work, Heide Barnhill for help with the SEM work, and Eric Schaller for comments on the manuscript.
Received February 9, 1999; accepted August 9, 1999.
|
|
REFERENCES |
|---|
Altschul, S.F., Gish, W., Miller, W., Myers, E.W., and Lipman, D.J. (1990) Basic local alignment search tool. J. Mol. Biol. 215:403-410[CrossRef][ISI][Medline].
ap Rees, T. (1980) Assessment of the contribution of metabolic pathways to plant respiration. In Davies D.D., ed. The Biochemistry of Plants, Vol 2, Metabolism and Respiration. New York, Academic Press. 1–29.pp.
Bechtold, N., Ellis, J., and Pelletier, G. (1993) In planta Agrobacterium gene transfer by infiltration of adult Arabidopsis thaliana plants. C. R. Acad. Sci. Ser. III Sci. Vie 316:1194-1199.
Behrends, W., Engeland, K., and Kindl, H. (1988) Characterization of two forms of the multifunctional protein acting in fatty acid ß-oxidation. Arch. Biochem. Biophys. 263:161-169[CrossRef][Medline].
Binstock, J.F., and Schulz, H. (1981) Fatty acid oxidation complex from Escherichia coli.. Methods Enzymol. 71:403-410.
Bradley, D., Ratcliffe, O., Vincent, C., Carpenter, R., and Coen, E. (1997) Inflorescence commitment and architecture in Arabidopsis.. Science 275:80-83[Abstract/Free Full Text].
Browse, J., McCourt, P., and Somerville, C.R. (1985) A mutant of Arabidopsis lacking a chloroplast-specific lipid. Science 227:763-765[Abstract/Free Full Text].
Bucher, P., and Bairoch, A. (1994) A generalized profile syntax for biomolecular sequence motifs and its function in automatic sequence interpretation. ISMB 2:53-61.
Chang, C., Kwok, S.F., Bleecker, A.B., and Meyerowitz, E.M. (1993) Arabidopsis ethylene-response gene ETR1: Similarity of product to two-component regulators. Science 262:539-544[Abstract/Free Full Text].
Chomczynski, P., and Sacchi, N. (1987) Single-step method of RNA isolation by acid guanidine thiocyanate–phenol–chloroform extraction. Anal. Biochem. 162:156-159[ISI][Medline].
Cooper, T.G., and Beevers, H.J. (1969) Beta oxidation in glyoxysomes from castor bean endosperm. J. Biol. Chem. 224:3514-3520.
Craig, S. (1992) The GUS reporter gene—Application to light and transmission electron microscopy. In Gallagher S.R., ed. GUS Protocols: Using the GUS Gene as a Reporter of Gene Expression. San Diego, CA, Academic Press. 115–124.pp.
Eaton, S., Bartlett, K., and Pourfarzam, M. (1996) Mammalian mitochondrial beta-oxidation. Biochem. J. 320:345-357.
Farmer, E.E. (1994) Fatty acid signaling in plants and their associated microorganisms. Plant Mol. Biol. 26:1423-1437[CrossRef][ISI][Medline].
Feldmann, K.A. (1991) T-DNA insertion mutagenesis in Arabidopsis: Mutational spectrum. Plant J. 1:71-82.
Garraway, J.L. (1970) ß-Oxidation of certain growth regulators by plant tissues. Pestic. Sci. 1:132-137.
Gerhardt, B. (1986) Basic metabolic function of the higher plant peroxisome. Physiol. Vég. 24:397-410.
Gerhardt, B. (1992) Fatty acid degradation in plants. Prog. Lipid Res. 31:417-446[CrossRef][ISI][Medline].
Gühnemann-Schäfer, K., and Kindl, H. (1995a) The leaf peroxisomal form (MFP IV) of multifunctional protein functioning in fatty acid ß-oxidation. Planta 196:642-646.
Gühnemann-Schäfer, K., and Kindl, H. (1995b) Fatty acid ß-oxidation in glyoxysomes. Characterization of a new tetrafunctional protein (MFP III). Biochim. Biophys. Acta 1256:181-186[Medline].
Hashimoto, T. (1996) Peroxisomal ß-oxidation: Enzymology and molecular biology. Ann. NY Acad. Sci. 804:86-98[Medline].
Haughn, G.W., and Somerville, C.R. (1988) Genetic control of morphogenesis in Arabidopsis.. Dev. Gen. 9:73-89.
Hayashi, M., Toriyama, K., Kondo, M., and Nishimura, M. (1998) 2,4-Dichlorophenoxybutryic acid–resistant mutants of Arabidopsis have defects in glyoxysomal fatty acid ß-oxidation. Plant Cell 10:183-195[Abstract/Free Full Text].
Hensel, L.L., and Bleecker, A.B. (1992). Arabidopsis as a model system for analysis of leaf senescence and inflorescence meristem longevity. In Proceedings of the 21st Steenbock Symposium: Cellular Communication in Plants, R.M. Amasino, ed (New York: Plenum Press), pp. 123–130.
Hensel, L.L., Grbic, V., Baumgarten, D.B., and Bleecker, A.B. (1993) Developmental and age-related processes that influence the longevity and senescence of photosynthetic tissues in Arabidopsis. Plant Cell 5:553-564[Abstract].
Hensel, L.L., Nelson, M.A., Richmond, T.A., and Bleecker, A.B. (1994) The fate of inflorescence meristems is controlled by developing fruits in Arabidopsis.. Plant Physiol. 106:863-876[Abstract].
Höfte, H. et al. (1993) An inventory of 1152 expressed sequence tags obtained by partial sequencing of cDNAs from Arabidopsis thaliana.. Plant J. 4:1051-1061[CrossRef][ISI][Medline].
Jefferson, R.A. (1987) Assaying chimeric genes in plants: The GUS gene fusion system. Plant Mol. Biol. Rep. 5:387-405.
Kieber, J.J., Rothenberg, M., Roman, G., Feldmann, K.A., and Ecker, J.R. (1993) CTR1, a negative regulator of the ethylene response pathway in Arabidopsis, encodes a member of the Raf family of protein kinases. Cell 72:427-441[CrossRef][ISI][Medline].
Kunau, W.-H., Dommes, V., and Schulz, H. (1995) ß-Oxidation of fatty acids in mitochondria, peroxisomes, and bacteria: A century of continued progress. Prog. Lipid Res. 34:267-342[CrossRef][ISI][Medline].
Levings, C.S., III (1993) Thoughts on cytoplasmic male sterility in cms-T maize. Plant Cell 5:1285-1290[Free Full Text].
Lincoln, C., Long, J., Yamaguchi, J., Serikawa, K., and Hake, S. (1994) A knotted1-like homeobox gene in Arabidopsis is expressed in the vegetative meristem and dramatically alters leaf morphology when overexpressed in transgenic plants. Plant Cell 6:1859-1876[Abstract/Free Full Text].
Lister, C., and Dean, C. (1993) Recombinant inbred lines for mapping RFLP and phenotypic markers in Arabidopsis thaliana.. Plant J. 4:745-750[CrossRef][ISI].
Mannaerts, G.P., and Debeer, L.J. (1982) Mitochondrial and peroxisomal ß-oxidation of fatty acids in rat liver. In Kindl H., Lazarow P.B., eds. Peroxisomes and Glyoxysomes. New York, Ann. NY Acad. Sci. 30–38.pp.
Mannaerts, G.P., and van Veldhoven, P.P. (1996) Functions and organization of peroxisomal beta-oxidation. Ann. NY Acad. Sci. 804:99-115[ISI][Medline].
Meyerowitz, E.M., Bowman, J.L, Brockman, L.L, Drews, G.N., Jack, T., Sieburth, L.E., and Weigel, D. (1991) A genetic and molecular model for flower development in Arabidopsis thaliana.. Dev. Suppl. 1:157-167[Free Full Text].
Millar, A.A., Wrischer, M., and Kunst, L. (1998) Accumulation of very-long-chain fatty acids in membrane glycerolipids is associated with dramatic alterations in plant morphology. Plant Cell 10:1889-1902[Abstract/Free Full Text].
Murashige, T., and Skoog, F. (1962) A revised medium for rapid growth and bioassays with tobacco tissue culture. Physiol. Plant. 15:493-497.
Murray, M.G., and Thompson, W.F. (1980) Rapid isolation of high molecular weight plant DNA. Nucleic Acids Res. 8:4321-4325[Abstract/Free Full Text].
Preisig-Müller, R., Gühnemann-Schäfer, K., and Kindl, H. (1994) Domains of the tetrafunctional protein acting in glyoxysomal fatty acid ß-oxidation. J. Biol. Chem. 269:20475-20481[Abstract/Free Full Text].
Roe, J.L., Rivin, C.J., Sessions, R.A., Feldmann, K.A., and Zambryski, P.C. (1993) The Tousled gene in A. thaliana encodes a protein kinase homolog that is required for leaf and flower development. Cell 75:939-950[CrossRef][ISI][Medline].
Sambrook, J., Fritsch, E.F., and Maniatis, T. (1989) Molecular Cloning: A Laboratory Manual. 2nd ed Cold Spring Harbor, NY, Cold Spring Harbor Laboratory Press.
Scherer, G.F.E. (1996) Phospholipid signaling and lipid-derived second messengers in plants. Plant Growth Regul. 18:125-133[CrossRef].
Smyth, D.R., Bowman, J.L., and Meyerowitz, E.M. (1990) Early flower development in Arabidopsis.. Plant Cell 2:755-767[Abstract/Free Full Text].
Spaink, H.P., Sheeley, D.M., van Brussel, A.A.N., Glushka, J., York, W.S., Tak, T., Geiger, O., Kennedy, E.P., Reinhold, V.N., and Lugtenberg, B.J.J. (1991) A novel highly unsaturated fatty acid moiety of lipo-oligosaccharide signals determines host specificity of Rhizobium.. Nature 354:125-130[CrossRef][Medline].
Synerholm, M.E., and Zimmerman, P.W. (1947) Preparation of a series of 
-(2,4-dichloropheoxy)-alphatic acids and some related compounds with a consideration of their biochemical role as plant-growth regulators. Contrib. Boyce Thompson Inst. 14:369-382.
Tolbert, N.E. (1981) Peroxisomes and glyoxysomes. Annu. Rev. Biochem. 50:133-157[CrossRef][ISI][Medline].
Voytas, D.F., Konieczny, A., Cummings, M.P., and Ausubel, F.M. (1990) The structure, distribution and evolution of the Ta1 retrotransposable element family of Arabidopsis thaliana.. Genetics 126:713-721[Abstract].
Watkins, P.A., Chen, W.W., Harris, C.J., Hoefler, G., Hoefler, S., Blake, D.C., Jr., Balfe, A., Kelley, R.I., Moser, A.B., Beard, M.E., and Moser, H.W. (1989) Peroxisomal bifunctional enzyme deficiency. J. Clin. Invest. 83:771-777.
Weigel, D., and Clark, S.E. (1996) Sizing up the floral meristem. Plant Physiol. 112:5-10[Medline].
Weigel, D., and Meyerowitz, E.M. (1994) The ABCs of floral homeotic genes. Cell 78:203-209[CrossRef][ISI][Medline].
ABSTRACT
INTRODUCTION
