The shoot apical meristem of dicotyledonous plants is highly regulated both structurally and functionally, but little is known about the mechanisms involved in this regulation. Here we describe the genetic and phenotypic characterisation of recessive mutations at three loci of Arabidopsis thaliana in which meristem structure and function are disrupted. The loci are Clavata1 (Clv1), Fasciata1 (Fas1) and Fasciata2 (Fas2). Plants mutant at these loci are fasciated having broad, flat stems and disrupted phyllotaxy. In all cases, the fasciations are associated with shoot apical meristem enlargement and altered floral development. While all the mutants share some phenotypic features they can be divided into two classes. The pleiotropic fas1 and fas2 mutants are unable to initiate wild-type organs, show major alterations in meristem structure and have reduced root growth. In contrast, clv1 mutant plants show near wild-type organ phenotypes, more subtle changes in shoot apical meristem structure and wild-type root growth.

The shoot apical meristem of higher plants is laid down during embryogenesis and gives rise to all the aerial parts of the plant. The meristem is usually a dome of cells consisting of a peripheral zone (PZ) of rapidly dividing cells in which leaf initiation occurs, and a central zone (CZ) of more slowly dividing cells which replenish the PZ. Around the base of the peripheral zone, leaves are initiated in a specific pattern (phyllotaxy). In dicotyledonous plants, superimposed on this zonation, there are three generative layers of cells. The layers are maintained by the pattern of cell division in the meristem. The outer layer (L1) divides anticlinally to give rise to the plant epidermis; the middle layer (L2) divides anticlinally in the dome apex, and anticlinally and periclinally in the dome base to give rise to the plant mesoderm; and the inner layer (L3) divides in various planes to give rise to the central tissues of the plant (Satina et al., 1940). Occasional periclinal divisions occur resulting in the insertion of cells derived from one layer into the neighbouring layer, where they adopt a fate appropriate to their new layer. Despite this, the layers are largely ontogenetically separate yet they function in an integrated and coordinated manner to produce the plant body. While it is clear that there is a high degree of structural and functional organisation in the meristem, little is known about how this organisation is achieved or maintained.

We are taking a genetic approach to study the control of meristem structure and function using the model plant Ara - bidopsis thaliana. Arabidopsis provides an excellent model system for both genetic and molecular studies because of its rapid life cycle, small genome size and high fecundity, and because of the availability of extensive genetic maps of both visible (Koornneef et al., 1983) and restriction fragment length polymorphism markers (Chang et al., 1988, Nam et al., 1989). Arabidopsis has been increasingly used as a model system for the study of the genetics and molecular biology of higher plant development (Finkelstein et al., 1988). For example, a broad spectrum of mutants affecting embryonic (Mayer et al., 1991) and floral (Meyerowitz et al., 1991) development have been isolated and characterised.

In Arabidopsis, postembryonic growth starts with the initiation of a variable number of rosette leaves in a spiral phyllotaxy with unextended internodes. The leaf number depends on genotype and growth conditions (Rédei, 1969). At the end of the rosette stage, the bolting stem and a variable number of cauline leaves are initiated. Finally the meristem switches from producing leaf primordia to producing flower primordia. Organ initiation by the apical meristem is in the same spiral phyllotaxy in all three phases of growth, but individual plants show either clockwise or anticlockwise spirals (Smyth et al., 1990).

Previous work has shown that the shoot apical meristem of Arabidopsis is typical of dicots. Miksche and Brown (1965) describe the meristem of 6-day-old Arabidopsis plants (ecotype Estland) as a shallow dome, four cells deep and approximately 40 μ m in diameter. During vegetative growth, the meristem maintains this shape but increases in size. Vaughn (1952) measured an average diameter of 90 μ m for the meristems of 25-day-old Arabidopsis plants kept vegetative in short days. On transition to flowering, the dome becomes more convex (Miksche and Brown, 1965, Vaughn and Jones, 1953, Vaughn, 1955). These workers also describe typical zonation patterns within the Arabidopsis meristem (Brown et al., 1964).

Here we report the genetic analysis and morphological characterisation of three fasciated mutants of Arabidopsis. Fasciation is a term used to describe a variety of developmental abnormalities in the shoot system (Worsdell, 1905). These include distortions in phyllotaxy and broadening, flattening and, in extreme cases, bifurcation of the stem. Fasciations can result from a variety of causes such as wounding (Loiseau, 1959) or pathogen attack (Murai et al., 1980), or they may be genetic (MacArthur, 1926; McKelvie, 1962; Reinholz, 1966; Krickhahn and Napp-Zinn, 1975; Usmanov and Startsev, 1979; Haughn and Somerville, 1988). Fasciation represents a breakdown in the pattern of organogenesis and has been associated with meristematic enlargement (Loiseau, 1959; Krickhahn and Napp-Zinn, 1975). The study of fasciation might reveal how meristem structure and function are established and maintained in normal plants.

Seed of the Enkheim, Landsberg erecta and fas1 genotype were obtained from the Arabidopsis Information Service Seed Bank. Seed of the clv1-1 genotype was obtained from Vivian Irish and seed of the flo-5 genotype in the Columbia genetic background was obtained from George Haughn, and then crossed into a Lands-berg erecta background.

Plants were grown in an autoclaved mix of 80% sand and 20% Fisons Levington multi-purpose compost. The plants were maintained at 25° C in continuous light. Plants were grown axenically on Murashige and Skoog (1962) medium (MS) with 1% sucrose and solidified with 0.6% Difco Bacto Agar. Seedlings were grown in Petri dishes and mature plants were grown in jars or Magenta boxes (Sigma). Seeds were sterilized by soaking for ten minutes in a solution of 10% Clorox and 0.02% Triton X-100 followed by 30 seconds in 70% ethanol. The seeds were then washed in five changes of sterile distilled water and transferred to the MS medium using a sterile toothpick or a Pasteur pipette. The plants were maintained at 25° C in continuous light.

Ethyl methanesuphonate (EMS) mutagenesis

Mutagenesis was carried out in a fume cupboard. Landsberg erecta seed was soaked for five hours in a solution of 0.1 M potassium phosphate buffer at pH 5, 5% dimethyl sulphoxide and an appropriate concentration of EMS (between 12 and 500 mM). The seed was then washed twice for 15 minutes in 100 mM sodium thiosulphate and twice for 15 minutes in water. The seed was allowed to dry on Whatman 3MM paper, diluted with dry sand and planted at an approximate density of one seed per cm2. The mutagenised seed (the M1 generation) was grown up and their progeny (the M2 generation) were harvested from each M1 plant separately to give a collection of M2 families. Approximately 10 members from each of 1200 families were planted in soil and screened for fasciated individuals.

Shoot apical meristem sections

Shoots from axenically grown seedlings were fixed for at least three days in 2.5% glutaraldehyde, 50 mM PIPES pH 7.2 and 0.1% caffeine. The tissue was then washed in 50 mM PIPES pH 7.2, stained with OsO4 and dehydrated through an ethanol series of five 3-hour steps ending in absolute ethanol. The ethanol was gradually replaced with epoxy resin to allow infiltration. The resin was polymerised by baking overnight at 67° C. The embedded tissue was sectioned using a pyramitome with glass knives. The sections were stretched in a drop of 100% ethanol and cleared and mounted with Gurr’s neutral mounting medium supplied by BDH.

Mutant isolation and genetic analysis of the fasciated mutants

1200 Landsberg erecta EMS mutagenised M2 families were generated and screened for fasciated plants. Ten fasciated lines were isolated but only two were studied further since the remaining eight showed variable penetrance and expressivity. A literature survey revealed a fasciated line isolated by Reinholz (1966) following X-ray mutagenesis of seed of the Enkheim genetic background. This line was named fasciata (Fas1). We obtained seed from this line from the Arabidopsis Information Service seed bank.

Genetic analysis of the mutant lines showed that in each case, the fasciations resulted from a single recessive Mendelian mutation (data not shown). Pairwise crosses between the three mutants gave all wild-type plants in the F1, suggesting that each mutation is in a different gene. The genes were named Fas1 (Reinholz, 1966), Fas2, and Fas3. The Fas loci were mapped with respect to various of the genetic markers described by Koornneef et al. (1983). Fas1 was found to map between Gl2 and Ch1 on Chromosome1 (Table 1), and Fas2 was found to map between Yi and Ttg on chromosome 5 (Table 2). Fas3 was found to map to chromosome 1 and to be allelic to both clv1 (Koornneef et al., 1983) and flo5 (Haughn and Somerville, 1988) (data not shown). Koornneef et al. (1983) mapped Clv1 to chromosome 1, 8.6 map units proximal of Gl2 and Medford had previously shown flo5 and clv1 mutants to be allelic (Medford personal communication). In pairwise crosses, fas3, flo5, and clv1, all failed to complement each other in the F1 and no wild-type plants segregated in the F2. Therefore the fas3, clv1 and flo5 mutations are allelic and so flo5 was renamed clv1-2 and fas3 was renamed clv1-3.

Table 1.

The F2 segregation of fas1 ch1, and gl2 in crosses between fas1, gl2 double homozygotes and wild-type plants; and fas1, ch1 double homozygotes and wild-type plants

The F2 segregation of fas1 ch1, and gl2 in crosses between fas1, gl2 double homozygotes and wild-type plants; and fas1, ch1 double homozygotes and wild-type plants
The F2 segregation of fas1 ch1, and gl2 in crosses between fas1, gl2 double homozygotes and wild-type plants; and fas1, ch1 double homozygotes and wild-type plants
Table 2.

The F 2 segregation of fas2, ttg, and yi, in crosses between fas2 plants and ttg, yi double homozygotes

The F 2 segregation of fas2, ttg, and yi, in crosses between fas2 plants and ttg, yi double homozygotes
The F 2 segregation of fas2, ttg, and yi, in crosses between fas2 plants and ttg, yi double homozygotes

The fas1 mutation was isolated in the Enkheim background (Reinholz, 1966) which is wild-type at the erecta locus. When backcrossed into the Landsberg erecta genetic background the fas1 mutant plants show reduced fertility in plants homozygous for the erecta mutation. Because of this, the fas1 mutation was characterised in the Enkheim (ER+) background with Enkheim controls.

Phenotypic analysis of the fasciated mutants

Fasciation is characterised by altered phyllotaxy and stem broadening. These aspects of the mutant phenotypes were analysed and the results are presented below. Interestingly, both these aspects of the mutant phenotypes became pro-gressively worse as the plants developed. If plant growth was slowed, for example by low temperatures, then more extreme fasciations developed. Apart from fasciation, the mutants also showed a variety of additional phenotypes including altered floral phyllotaxy, altered floral organ structure and number, unusual leaf shapes and inhibition of root elongation. The analysis of each of these phenotypes is described below.

Disturbed leaf phyllotaxy

For both the wild-type strains used (Landsberg erecta and Enkheim), phyllotaxy was found to be spiral with new leaves initiating at an angle averaging 138° from the previous leaf. The direction of the spiral was random and the phyllotaxy did not change on transition to floral growth. All the mutant lines studied show deviations from this phyllotaxy. A precise analysis of leaf divergence angles is not practical for large numbers of plants but major alterations in the relative positions of leaves can be seen in intact plants (Fig. 1). The phyllotaxy of the first six leaves of plants of each genotype were analysed in this way and the percentage of plants showing deviation from the normal phyllotaxy was recorded (Fig. 2). Some individuals in all the mutant lines showed disturbed phyllotaxy within the first six leaves. An allelic series was found within the Clv1 alleles with 60% of clv1-2 plants, 42% of clv1-3 plants and 30% of clv1-1 plants showing phyllotactic distortions within the first six leaves. This figure is 40% for fas2 plants and 23% for fas1 plants. These data show that in all five mutant lines examined, some proportion of seedlings initiate leaves in unusual positions with respect to the other early leaves.

Fig. 1.

The photographs and interpretive drawings show the phyllotaxy of (A) Landsberg erecta wild-type and (B) two clv1 plants. In the drawings leaves are outlined with solid lines and cotyledons are outlined with dotted lines. The leaves are numbered in order of initiation and the displaced leaves of clv1 plants are marked with an asterisk.

Fig. 1.

The photographs and interpretive drawings show the phyllotaxy of (A) Landsberg erecta wild-type and (B) two clv1 plants. In the drawings leaves are outlined with solid lines and cotyledons are outlined with dotted lines. The leaves are numbered in order of initiation and the displaced leaves of clv1 plants are marked with an asterisk.

Fig. 2.

The graphs show the percentage of clv1, fas1 and fas2 plants that show deviations from wild-type phyllotaxy within the first six rosette leaves. For each genotype 50 plants were scored except for fas2 where only 25 plants were scored. No phyllotactic abnormalities were observed in a samples of 50 Landsberg erecta plants and 50 Enkheim plants.

Fig. 2.

The graphs show the percentage of clv1, fas1 and fas2 plants that show deviations from wild-type phyllotaxy within the first six rosette leaves. For each genotype 50 plants were scored except for fas2 where only 25 plants were scored. No phyllotactic abnormalities were observed in a samples of 50 Landsberg erecta plants and 50 Enkheim plants.

Fasciation in the inflorescence

The inflorescence of Arabidopsis consists of a bolting stem, cauline leaves and flowers, initiated in the same spiral phyllotaxy as the leaves (Smyth et al., 1990). In all the mutants, the fasciations observed in the rosette persisted into the inflorescence and tended to worsen with time (Fig. 3 A-F). The severity of inflorescence fasciations varied from plants in which only a single flower was not in the normal phyllotaxy, to plants in which the initiation of flowers failed entirely leaving the meristem exposed. Such exposed meristems occurred frequently in the fas2 line, more rarely in the fas1 line and never in the clv1 mutants.

Other less extreme phenotypes observed include overall stem enlargement, stem flattening (line or ribbon fasciation) (Fig. 3 E-F), and stem flattening accompanied by stem bifurcation. Bifurcation appears to normalise the phyllotaxy somewhat resulting in the establishment of two or more new spiral phyllotaxies. Another feature of the mutant inflorescences is the failure to maintain stem elongation so that the mature structure appears crowded with flowers and siliques.

Changes in the number and structure of floral organs

In addition to fasciations, all the mutant lines show abnormalities in floral phenotype (Fig. 3 A-B). In clv1 mutants, additional floral organs may arise in all or any of the floral whorls (Fig. 4). The most consistently affected whorl is the carpel whorl. Here the wild-type number of two carpels is seen rarely, and only in clv1-1 plants. In all three alleles the modal carpel number is four. The organs of clv1 flowers appear to be structurally wild-type although some double organs arise.

Fig. 3.

Photographs A-D show the typical inflorescence morphology of (A) a wild-type Landsberg erecta plant, (B) a clv1-2 plant, (C) a fas1 plant and (D) a fas2 plant. The oldest clv1-2 flower has been opened to show the additional floral organs. The inflorescences of fas1 and fas2 plants are similar in morphology being generally disorganised. The narrow sepals and petals of the fas mutant flowers allow anthers to protrude as shown. Photographs E and F show the bolting stem of (E) a wild-type Landsberg erecta plant and (F) a fas1 plant. The wild-type stem is of even width throughout. Toward the apex, the fas1 stem shows a typical ribbon fasciation, being broad and flat. Photographs G-J show typical rosette leaves from (G) a wild-type Landsberg erecta plant, (H) a clv1-2 plant, (I) a fas1 plant and (J) a fas2 plant. The leaves of clv1 plants are consistently somewhat rounder than wild-type. In the fas mutants the leaves are variable in shape. Typically rosette leaves of fas1 plants are dentate as shown, while the rosette leaves of fas2 plants are more nearly wild-type but may be asymmetrical. Particularly in fas2 mutant plants, more extreme leaf shape abnormalities are observed later in development (see Fig. 6).

Fig. 3.

Photographs A-D show the typical inflorescence morphology of (A) a wild-type Landsberg erecta plant, (B) a clv1-2 plant, (C) a fas1 plant and (D) a fas2 plant. The oldest clv1-2 flower has been opened to show the additional floral organs. The inflorescences of fas1 and fas2 plants are similar in morphology being generally disorganised. The narrow sepals and petals of the fas mutant flowers allow anthers to protrude as shown. Photographs E and F show the bolting stem of (E) a wild-type Landsberg erecta plant and (F) a fas1 plant. The wild-type stem is of even width throughout. Toward the apex, the fas1 stem shows a typical ribbon fasciation, being broad and flat. Photographs G-J show typical rosette leaves from (G) a wild-type Landsberg erecta plant, (H) a clv1-2 plant, (I) a fas1 plant and (J) a fas2 plant. The leaves of clv1 plants are consistently somewhat rounder than wild-type. In the fas mutants the leaves are variable in shape. Typically rosette leaves of fas1 plants are dentate as shown, while the rosette leaves of fas2 plants are more nearly wild-type but may be asymmetrical. Particularly in fas2 mutant plants, more extreme leaf shape abnormalities are observed later in development (see Fig. 6).

Fig. 4.

The graphs show the mean deviation from wild-type floral organ number in each of the fasciated mutant lines. In both Landsberg erecta and Enkheim flowers there are invariantly four sepals, four petals and two carpels. Enkheim plants always have six stamens but in Landsberg erecta flowers one, or both of the abaxial stamens may fail to develop such that the mean number of stamens in these flowers is 5.7±0.02. The error bars represent the standard error of the means.

Fig. 4.

The graphs show the mean deviation from wild-type floral organ number in each of the fasciated mutant lines. In both Landsberg erecta and Enkheim flowers there are invariantly four sepals, four petals and two carpels. Enkheim plants always have six stamens but in Landsberg erecta flowers one, or both of the abaxial stamens may fail to develop such that the mean number of stamens in these flowers is 5.7±0.02. The error bars represent the standard error of the means.

In contrast, both fas1 and fas2 mutants have fewer floral organs in their petal and stamen whorls and on average more sepals, while the carpel number is unaffected (Fig. 4). The organs in all whorls are not wild-type. The sepals and petals are narrow and both mutant lines show reduced fertility. In fas2 mutants, floral initiation can break down entirely or fused partial flowers may arise.

Leaf morphology and root growth

The leaves of clv1 mutants are slightly rounder than wild-type (Fig. 3 G-H) and occasional double leaves are observed. All three clv1 mutants have near wild-type root growth (Fig. 5). The fas1 and fas2 mutants show variable leaf shapes (Fig. 3 I-J) and reduced root elongation (Fig. 5). The leaves of fas1 plants are dentate and generally narrower than wild-type. Fas2 plants show a variety of leaf shapes but the leaves are also often narrower than wild-type and occasionally lanceolate, particularly the cauline leaves (Fig. 6). The roots of fas1 plants are 60% wild-type length after 2 weeks of growth and those of fas2 are 38% wild-type length.

Fig. 5.

The graph shows the mean root length of wild-type (Landsberg erecta and Enkheim) and mutant plants after two weeks growth on vertically positioned Petri dishes. The error bars represent the standard error of the means.

Fig. 5.

The graph shows the mean root length of wild-type (Landsberg erecta and Enkheim) and mutant plants after two weeks growth on vertically positioned Petri dishes. The error bars represent the standard error of the means.

Fig. 6.

(A) The photograph shows the shoot apical meristem of a fas2 plant. Organ initiation in this plant has been affected to such an extent that the meristem is not hidden by young leaves or buds as it would be in a wild-type plant. The meristem has degenerated into a large callus-like structure (M). Note also the narrow leaf (L) on the bolting stem. (B) The photomicrographs show 6 μ m median, longitudinal sections through the shoot apical meristem of a 5-day-old wild-type Enhkheim plant on the left, and a 5-day-old fas1 plant on the right. (C) The photomicrographs show 6 μ m median, longitudinal sections through the shoot apical meristems of a 5-day-old wild-type Landsberg erecta plant on the left, and a 5-day-old clv1-3 plant on the right. In B and C, the plants were stained with OsO4 during fixation. Meristematic cells are identifiable because their cytoplasm stains more densely than that of nearby, more vacuolate cells. The meristems are marked M and leaf initials are marked I. Xylem vessels are marked X. Bar, 20 μ m.

Fig. 6.

(A) The photograph shows the shoot apical meristem of a fas2 plant. Organ initiation in this plant has been affected to such an extent that the meristem is not hidden by young leaves or buds as it would be in a wild-type plant. The meristem has degenerated into a large callus-like structure (M). Note also the narrow leaf (L) on the bolting stem. (B) The photomicrographs show 6 μ m median, longitudinal sections through the shoot apical meristem of a 5-day-old wild-type Enhkheim plant on the left, and a 5-day-old fas1 plant on the right. (C) The photomicrographs show 6 μ m median, longitudinal sections through the shoot apical meristems of a 5-day-old wild-type Landsberg erecta plant on the left, and a 5-day-old clv1-3 plant on the right. In B and C, the plants were stained with OsO4 during fixation. Meristematic cells are identifiable because their cytoplasm stains more densely than that of nearby, more vacuolate cells. The meristems are marked M and leaf initials are marked I. Xylem vessels are marked X. Bar, 20 μ m.

The shoot apical meristems of wild-type and mutant plants

Fasciation represents a breakdown in the control of shoot apical meristem function and so may involve or result from breakdown in shoot apical meristem structure. The apical meristem of fas2 plants can frequently be observed directly in individuals in which organ initiation has failed to such an extent that the meristem is not hidden by recently initiated leaves. These plants show that the meristem has enlarged greatly, mainly in one plane (Fig. 6). In fas1 and clv1 plants, direct observation of the meristem is not possible so longitudinal sections were taken though the meristems of 5-day-old fas1, clv1-3, Enkhiem and Landsberg erecta plants (Fig. 6). At least 4 meristems of each genotype were examined. The sections show that at this stage the meristems of both the wild-type lines used consist of a shallow dome of avacuolate cells. In Landsberg erecta plants the meristem measures between 50 μ m and 62 μ m across the base and between 20 μ m and 25 μ m in height. At its highest point, the meristem has 4 layers of avacuolate cells. The meristem of clv1-3 plants is on average both broader and taller than wild-type measuring between 60 μ m and 90 μ m across the base, and between 25 μ m and 30 μ m in height. At its highest point, the clv1-3 meristem has between 5 and 7 layers of avacuolate cells. Although the dome structure is maintained, clv1-3 meristems have more gently sloping sides so that they are bell shaped.

In wild-type Enkheim plants the meristem measures between 50 μ m and 65 μ m across the base and between 25 μ m and 30 μ m in height. In fas1 plants, the meristem is usually broader than wild-type being between 50 μ m and 80 μ m across the base but the dome shape of the wild-type is not observed. The fas1 meristem is nearly flat, measuring between 15 μ m and 23 μ m in height, and having only 1 or 2 layers of avacuolate cells.

Double mutant combinations

In order to examine possible interactions between the mutations, pairwise crosses were performed between clv1-2 and the fas1 and fas2 mutants. The resultant F2 populations were screened for possible double mutant homozygotes. Plants were observed which showed the leaf morphology of the fas mutants and the distinctive silique phenotype of the clv1 mutant. In the F3, the progeny of these plants were all of similar phenotype with no segregation of distinct phenotypic classes suggesting that these F2 plants were indeed doubly homozygous for clv1-2 and each of the fas mutations. The phenotypes of the fas1, clv1-2 and fas2, clv1-2 double mutants were morphologically similar (as fas1 and fas2 plants are similar), and appear to represent the super-imposition of the two phenotypes with no obvious interaction. Like fas mutants, the double mutants are slow growing with narrow and variable leaves, sepals and petals but like the clv1 mutants they have club-like siliques. Such plants show stem and inflorescence fasciation and the only novelty is that flower initiation tends to become localized on one side of the apex. This gives a “brushlike” appearance with siliques and flowers clustered on one side of the stem (Fig. 7).

Fig. 7.

The photograph shows the inflorescence of a clv1-2, fas1 double mutant plant. The bolting stem has curled round and floral initiation has become limited to one side of the stem.

Fig. 7.

The photograph shows the inflorescence of a clv1-2, fas1 double mutant plant. The bolting stem has curled round and floral initiation has become limited to one side of the stem.

Similar crosses were performed between the two fas mutants but no obvious double mutant class could be identified in the F2. In the F2 plants, 99 wild-type and 73 fasciated types were found. This is in good agreement with the segregation expected of two unlinked recessive mutants of similar phenotype; 9:7 (χ2 = 0.11). However, the data are also compatible with the double mutant class having a lethal phenotype; 9:6 (χ2 = 0.43). In an attempt to resolve this, seven fasciated F2 plants were harvested and the F3 sown in the expectation that segregation of the other fas mutation might be observed. Two of the seven families segregated a seedling lethal phenotype. The affected individuals die with two to four true leaves. There is no direct evidence that the lethal phenotype is related to the interaction of fas1 and fas2. So the phenotype of the double mutant class may either be the seedling-lethal or indistinguishable from the single fas homozygotes. Further genetic analysis is needed to resolve this.

In order to learn about the control of meristem structure and function in plants, we have characterised three shoot apical meristem mutants of Arabidopsis thaliana:-fas1, fas2 and clv1. The fas1 and fas2 loci were mapped. The clv1 locus has previously been mapped by Koornneef et al. (1983) and three independent clv1 alleles were characterised in this study. The mutants can be grouped into two classes with the clv1 mutants making up one class, and fas1 and fas2 making up the other. The main features that separate these classes are that fas1 and fas2 mutants have darker green, abnormally shaped leaves, short roots, abnormal floral organs and a tendency for breakdown in the ability of the meristem to initiate unique and distinct organs. Conversely, clv1 mutants have nearly wild-type leaves, wild-type roots and additional, but structurally normal, floral organs and, except for rare double organs, the plants retain the ability to initiate distinct and unique organs.

There are also features common to all the mutant phenotypes. The mutants were selected for fasciation and all show characteristic broad stems and distortions in phyllotaxy in both the vegetative and floral apex. In all cases, the fasciations are associated with enlargement and shape change in the meristem. The fasciations become progressively worse with time and can eventually lead to stem bifurcations. Following bifurcation, more normal phyllotaxy is temporarily restored.

Implications for the field theory of phyllotaxy

The association that we observe between phyllotactic distortions with meristematic enlargement is consistent with the results of other workers studying both genetic (Krickhahn and Napp-Zinn, 1975) and wound induced (Loiseau, 1959) fasciations. The observation that a more normal phyllotaxy is restored following stem bifurcation further strengthens the correlation between meristem enlargement and fasciation. Bifurcation results from the splitting of the apical meristem which presumably reduces its size. This correlation can be viewed as support for the field theory of phyllotaxy (Wardlaw, 1949). The field theory proposes that the position of new leaves is determined by the interaction of fields of substances inhibitory to leaf initiation which are produced by recently initiated leaves. If it is assumed that the size and strength of these fields is not affected by the mutations, then the observed enlargement of the meristems would lead to alterations in the interactions of the inhibitory fields, provoking alterations in the position of new leaf initiation, and hence alterations in phyllotaxy.

It is possible that the size and/or strength of the fields are affected by the mutations. This would lead to phyllotactic distortion even in the absence of meristematic enlargement. Even so, it seems unlikely that any alteration in these fields is the primary affect of the mutations described here since this would not explain why the meristems of all the mutations are altered in size and shape, and why in all cases floral development is also affected. It therefore seems more likely that the mutations primarily affect meristem development and their phenotype should be interpreted in this light.

Implications for the independence of PZ and CZ function

The functions of the shoot apical meristem can be considered as firstly the initiation of distinct organs in a particular pattern, which is largely achieved by the peripheral zone (PZ), and secondly the maintenance of meristem size and shape which is largely achieved by the stem-cell-like activities of the central zone (CZ). In fas1 and fas2 plants both these functions are simultaneously disrupted such that distinct and wild-type organs are not always initiated and the size and shape of the meristem is not maintained.

In clv1 plants, more specific alterations are observed. The mutant plants retain the ability to initiate wild-type leaves but fail to maintain apical meristem size and shape. The phenotype of clv1 mutants clearly demonstrates the partial independence of these two meristem functions although the phenotypes of fas1 and fas2 indicate that this separation is not total. An attractive hypothesis to explain the clv1 phenotype is that only a subset of cells in the meristem are affected by the mutations. The curious bell shape of some clv1 meristems, coupled with the observation that the phenotype of the leaves of clv1 plants is nearly wild-type, suggests that only the CZ and not the PZ is affected by the mutations. If the CZ enlarges, the PZ would be forced out-ward and, if the field theory of phyllotaxy is correct, this would lead to phyllotactic distortions as discussed above. Despite this enlargement, the PZ of clv1 plants retains the ability to direct the differentiation of nearly wild-type leaves. The rare double leaves may not represent a break-down in the normal mechanisms directing organ initiation, instead they might represent the normal functioning of these mechanisms without compensating for the increased size of the meristem. This will lead, on occasions, to two leaves initiating very close together, which could result in the observed double leaves. This contrasts with the frequent failure in distinct organ initiation observed in fas1 and particularly in fas2 plants. If this failure occurs at all in any one plant, it usually continues until the meristem fails altogether. This phenotype may be the result of a more general breakdown in meristem function.

One possible way to test the relative effects of the mutations described on the PZ and the CZ would be to use a PZ-specific reporter gene. Medford et al. (1991) have isolated a promoter from cauliflower that expresses specifically in the PZ of the meristem. This promoter has been fused to the β-galacturonidase reporter gene (Jefferson et al., 1987) for which a number of chromogenic substrates are available. The transfer of these constructs into plants allows histochemical localisation of gene expression. If such constructs were introduced into wild-type and mutant plants, then the relative sizes of the PZ and CZ could be studied in the mutants and compared to those of wild-type plants. In this way, it might be possible to test the theory that there are CZ-specific alterations in clv1 meristems but less specific alterations in fas1 and fas2 meristems.

Implications for the relationship between the floral primordium and shoot apical meristem

Mutations at all the loci studied give rise to non-wild-type floral development. The nature of the floral phenotype in each mutant line closely parallels the phenotype of the shoot system. In fas1 and fas2 plants, the floral organs, like the leaves, are narrow and variable resulting in reduced fertility. In clv1 plants, while additional organs are observed, these organs are wild-type, although occasional double floral organs are formed as were occasional double leaves. The carpels are consistently duplicated in clv1 plants. These organs are derived from the centre of the flower primordium. The clv1 mutation results in enlargement of the centre of the apical meristem and the carpel phenotype may involve a similar enlargement of the centre of the flower primordium. The parallel phenotype of the shoot apical meristem and the floral primordia suggests that there are common factors involved in regulating both these structures. This is consistent with the view that the floral primordium is evolutionarily derived from the shoot apical meristem. This idea is supported by genetic evidence which suggests floral organs are modified leaves. Arabidopsis plants mutant in genes involved in floral development may fail to produce floral organs and produce leaves in the positions where floral organs would normally develop (Bowman et al., 1989).

Attempts were made to construct all three possible double mutant combinations of fas1, fas2 and clv1-2. The fas2, fas1 combination may either resemble the fas mutants or it may be lethal. The fas1, clv1-2 and fas2, clv1-2 plants are similar in phenotype, suggesting that the affects of these mutation are additive.

The phenotypes of the mutants described here probably result from meristem enlargement and shape change. This in turn alters the pattern of organ initiation and to a lesser extent the structure of the organs formed. In this respect, these mutations are distinct from other groups of developmental mutants in Arabidopsis which affect organ identity (Meyerowitz et al., 1991), developmental timing (Rédei, 1962) and presence or absence of pattern elements (Mayer et al., 1991). It is probably not meaningful to speculate on the specific roles of the wild-type genes defined by these mutations since fasciations can be provoked by rather non-specific stimuli (Loiseau, 1959; Murai et al. 1980). However, particularly in the case of clv1 mutants where such precise changes are seen, it is possible that these genes are involved in regulating the proliferation and differentiation of meristem cells.

We would like to thank Patrick Echlin and Brian Chapman for help with the microscopy, Joanne Pumfrey for help with double mutant construction, Mark Estelle and Stephen Day for critical reading of the manuscript and Joanne Griffiths for typing.

This work was supported by the Agriculture and Food Research Council. O.L. was supported by a Science and Engineering Research Council studentship.

Bowman
,
J. L.
,
Smyth
,
D. R.
and
Meyerowitz
,
E. M.
(
1989
).
Genes directing flower development in Arabidopsis
.
The Plant Cell
1
,
37
52
.
Brown
,
J. A. M.
,
Miksche
,
J. P.
and
Smith
,
H. H.
(
1964
).
An analysis of 3H-thymidine distribution throughout the vegetative meristem of Arabidopsis thaliana L. Heynh
.
Radiation Botany
4
,
107
113
.
Chang
,
C.
,
Bowman
,
J. L.
,
Dejohn
,
A. W.
,
Lander
E. S.
and
Meyerowitz
,
E. M.
(
1988
).
Restriction fragment length polymorphism linkage map of Arabidopsis thaliana
.
Proc. Natn. Acad. Sci. U.S.A
.
85
,
6856
6860
.
Finkelstein
,
R.
,
Estelle
,
M.
,
Martinez-Zapater
,
J.
and
Somerville
,
C.
(
1988
).
Arabidopsis as a tool for the identification of genes involved in plant development
.
In Temporal and Spatial Regulation of Plant Genes
. (eds.
D. P. S.
Verma
and
R. B.
Goldberg
) pp.
1
25
.
Wien
:
Springer Verlag
.
Haughn
,
G. W.
and
Somerville
,
C. R.
(
1988
).
Genetic control of morphogenesis in Arabidopsis
.
Developmental Genetics
9
,
73
89
.
Jefferson
,
R. A.
,
Kavanagh
,
T. A.
and
Bevan
,
M. W.
(
1987
).
Gus fusions. b-glucuronidase as a sensitive and versatile gene fusion marker in higher plants
.
EMBO J
.
6
,
3901
3907
.
Koornneef
,
M.
,
van Eden
,
J.
,
Hanhart
,
C. J.
,
Stam
P.
,
Braaksma
,
F. J.
and
Feenstra
,
W. J.
(
1983
).
Linkage map of Arabidopsis thaliana
.
J. Heredity
74
,
265
272
.
Krickhahn
,
D.
and
Napp-Zinn
,
K.
(
1975
).
Fasciation studies in Arabidopsis thaliana (L.) Heynh
.
Arabid. Inf. Serv
.
12
,
9
.
Loiseau
,
J.-E.
(
1959
).
Observations et experimentation sur la phyllotaxie et le fonctionnement du sommet vegetatif chez quelques Balsaminacees
.
Ann. Sci. Nat. Bot. Ser
. 11
20
,
1
214
.
MacArthur
,
J. W.
(
1926
).
Linkage studies with the tomato
.
Genetics
11
,
387
405
.
Mayer
,
U.
,
Tores Ruiz
,
R. A.
,
Berleth
,
T.
,
Misera
,
S.
and
Jurgens
,
G.
(
1991
).
Mutations affecting body organisation in the Arabidopsis embryo
.
Nature
532
,
402
407
.
McKelvie
,
A. D.
(
1962
).
A list of mutant genes in Arabidopsis thaliana (L.) Heynh
.
Radiation Botany
1
,
233
241
.
Medford
,
J. I.
,
Scott Elmer
,
J.
and
Klee
,
H. J.
(
1991
).
Molecular cloning and characterisation of genes expressed in shoot apical meristems
.
The Plant Cell
3
,
359
370
.
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
.
Development 1991 Supplement
1
,
157
168
.
Miksche
,
J. P.
and
Brown
,
J. A. M.
(
1965
).
Development of vegetative and floral meristems of Arabidopsis thaliana
.
Amer. J. Bot
.
52
,
533
537
.
Murai
,
N.
,
Skoog
,
F.
,
Doyle
,
M. E.
and
Hanson
,
R. S.
(
1980
).
Relationships between cytokinin production, presence of plasmids, and fasciation caused by strains of Corynebacterium fascians
.
Proc. Natn. Acad. Sci. U.S.A
.
77
,
619
623
.
Murashige
,
T.
and
Skoog
,
F.
(
1962
).
A revised medium for growth and bioassays with tobacco cell cultures
.
Physiol. Plant
.
15
,
473
497
.
Nam
,
H-G.
,
Giraudat
,
J.
,
den Boer
,
B.
,
Moonan
,
F.
,
Loos
,
W. D. B.
,
Hauge
,
B. M.
and
Goodman
,
H. M.
(
1989
).
Restriction fragment length polymorphism linkage map of Arabidopsis thaliana
.
The Plant Cell
1
,
699
705
.
Rédei
,
G.
(
1962
).
Supervital mutants of Arabidopsis
.
Genetics
47
,
443
460
.
Rédei
,
G.
(
1969
).
Arabidopsis thaliana (L) Heynh, a review of the genetics and biology
.
Bibliographica Genetica
21
,
1
151
.
Reinholz
,
E.
(
1966
).
Radiation induced mutants showing changed inflorescence characteristics
.
Arabid. Inf. Serv
.
3
,
19
20
.
Satina
,
S.
,
Blakeslee
,
A. F.
and
Avery
,
A. G.
(
1940
).
Demonstration of the three germ layers in the shoot apex of Datura by means of induced polyploidy periclinal chimeras
.
Amer. J. Bot
.
27
,
895
905
.
Smyth
,
D. R.
,
Bowman
,
J. L.
and
Meyerowitz
,
E. M.
(
1990
).
Early flower development in Arabidopsis
.
The Plant Cell
2
,
755
767
.
Usmanov
,
P. D.
and
Startsev
,
G. A.
(
1979
).
The appearance of stem fasciation in Arabidopsis thaliana F2 by crossing the mutant 90 with LuCo
.
Arabid. Inf. Serv
.
16
,
99
102
.
Vaughn
,
J. G.
(
1952
).
Structure of the angiosperm apex
.
Nature
169
,
458
459
.
Vaughn
,
J. G.
and
Jones
F. R.
(
1953
).
Structure of the angiosperm inflorescence apex
.
Nature
171
,
751
752
.
Vaughn
,
J. G.
(
1955
).
The morphology and growth of the vegetative and reproductive apices of Arabidopsis thaliana (L.) Heynh., Capsella bursa-pastoris (L.) Medic., and Anagallis arvensis
.
J. Linn. Soc. Bot
.
55
,
279
301
.
Wardlaw
,
C. W.
(
1949
).
Experiments on organogenesis in ferns
.
Growth (Suppl.)
13
,
93
131
.
Worsdell
,
W. C.
(
1905
).
Fasciation: Its meaning and origin
.
New Phytol
.
4
,
55
74
.