The Plant Cell, Vol. 13, 31–46, January 2001, www.plantcell.org © 2001 American Society of Plant Physiologists

RESEARCH ARTICLE

Stamina pistilloida

, the Pea Ortholog of

Fim

and

UFO

, Is Required for Normal Development of Flowers, Inflorescences, and Leaves

Scott Taylor,

a,1,2

Julie Hofer,

b

and Ian Murfet

a

a

School of Plant Science, University of Tasmania, Hobart, Tasmania, 7001, Australia

b

John Innes Centre, Colney Lane, Norwich, NR4 7UH, United Kingdom

Isolation and characterization of two severe alleles at the

Stamina

pistilloida

(

Stp

) locus reveals that

Stp

is involved in awide range of developmental processes in the garden pea. The most severe allele,

stp-4

, results in flowers consistingalmost entirely of sepals and carpels. Production of ectopic secondary flowers in

stp-4

plants suggests that

Stp

is in-volved in specifying floral meristem identity in pea. The

stp

mutations also reduce the complexity of the compound pealeaf, and primary inflorescences often terminate prematurely in an aberrant sepaloid flower. In addition,

stp

mutantswere shorter than their wild-type siblings due to a reduction in cell number in their internodes. Fewer cells were alsofound in the epidermis of the leaf rachis of

stp

mutants. Examination of the effects of

stp-4

in double mutant combina-tions with

af

,

tl

,

det

, and

veg2-2

—mutations known to influence leaf, inflorescence, and flower development in pea—suggests that

Stp

function is independent of these genes. A synergistic interaction between weak mutant alleles at

Stp

and

Uni

indicated that these two genes act together, possibly to regulate primordial growth. Molecular analysis re-vealed that

Stp

is the pea homolog of the Antirrhinum gene

Fimbriata

(

Fim

) and of

UNUSUAL FLORAL ORGANS

(

UFO

)from Arabidopsis. Differences between

Fim

/

UFO

and

Stp

mutant phenotypes and expression patterns suggest that ex-pansion of

Stp

activity into the leaf was an important step during evolution of the compound leaf in the garden pea.

INTRODUCTION

The complex process of floral meristem identity and flowerdevelopment has been partially dissected by studies of mu-tants in Antirrhinum

and Arabidopsis. These studies haverevealed the presence of an evolutionarily conserved hierar-chy of genes regulating flower development. Floral meristemidentity genes

Floricaula

(

Flo

) and

Squamosa

(

Squa

) fromAntirrhinum, and their Arabidopsis orthologs

LEAFY

(

LFY

)and

APETALA1

(

AP1

), are required during the transition toflowering (Coen et al., 1990; Huijser et al., 1992; Mandel etal., 1992; Weigel et al., 1992; Schultz and Haughn, 1993).Differentiation of the four floral organ types (sepals, petals,stamens, and carpels) requires overlapping activity of threeclasses of floral homeotic genes: A, B, and C (Schwarz-Sommer et al., 1990; Bowman et al., 1991; Coen andMeyerowitz, 1991). A activity alone produces sepals, A ac-

tivity in combination with B activity produces petals, B activ-ity in combination with C activity produces stamens, and Cactivity alone produces carpels (reviewed in Coen andMeyerowitz, 1991; Weigel and Meyerowitz, 1994; Yanofsky,1995).

Identification of

Fimbriata

(

Fim

) from Antirrhinum and

UN-USUAL FLORAL ORGANS

(

UFO

) from Arabidopsis sug-gested a mechanism by which

Flo

and

LFY

may regulateactivity of B- and C-class floral homeotic genes (Simon etal., 1994; Levin and Meyerowitz, 1995; Wilkinson andHaughn, 1995; Ingram et al., 1997; Lee et al., 1997). Muta-tions in

Fim

and

UFO

disrupt floral meristem identity andnormal expression of B- and C-class genes. Overexpressionof the class B gene

APETALA3

(

AP3

)

in a

ufo

mutant back-ground is able to rescue floral organ defects in

ufo

mutants,demonstrating that

AP3

acts downstream of

UFO

(Krizekand Meyerowitz, 1996). In Antirrhinum,

Fim

expression isdependent on early

Flo

expression (Simon et al., 1994), andregulation of class B activity by

Flo

may occur through

Fim

.However, in Arabidopsis,

UFO

does not act downstream of

LFY

because overexpression of

UFO

is unable to rescue flo-ral defects in

lfy

mutants (Lee et al., 1997). Thus,

LFY

and

1

Current address: Department of Applied Genetics, John InnesCentre, Colney Lane, Norwich, NR4 7UH, UK.

2

To whom correspondence should be addressed. E-mail scott.taylor@bbsrc.ac.uk; fax 44-1603-450027.

32 The Plant Cell

UFO

act as co-regulators specifying floral meristem identity.

UFO

may also possess functions independent of

LFY

be-cause expression patterns of

LFY

and

UFO

do not overlap(Lee et al., 1997). However,

UFO

appears to partner with

LFY

to control floral meristem identity and class B gene ac-tivity in Arabidopsis.

The greatest departure from the paradigm of

Flo

ortho-logs playing a role solely in floral meristem identity occurs inthe garden pea. The

unifoliata

(

uni

) mutant of pea is charac-terized by simplification of the compound leaf, and mutantplants bear flowers that consist entirely of sepals and car-pels (Marx, 1987; Hofer et al., 1997). Molecular analysis hasshown that

Uni

is the ortholog of

Flo

and

LFY

(Hofer et al.,1997), suggesting that activity of

Uni

has been recruited tocompound leaf development during the evolution of the gar-den pea. Alternatively, common regulation of leaf and flowerdevelopment by

Uni

may reflect an ancestral, pre-angio-sperm function in leaves and leaf-like sporophylls of seedferns (Hofer et al., 1997). This alternative was supported inpart by identification of the

Flo

ortholog in tomato,

Falsiflora

(

Fa

), because the

fa

mutation reduces the number of smallleaflets present on tomato compound leaves (Molinero-Rosales et al., 1999). The petunia ortholog of

Flo

,

aberrantleaf and flower

(

alf

), does not affect development of thesimple leaves of petunia (Souer et al., 1998); however, itdoes have a similar effect on floral meristem identity as

uni

,

flo

,

lfy

, and

fa

mutations in pea, Antirrhinum, Arabi-dopsis, and tomato, respectively. Analysis of floral mer-istem identity genes in other species has been hamperedby the lack of mutants.

Characterization of

Stamina pistilloida

(

stp-1

) mutant inthe garden pea by Monti and Devreux (1969) suggests that

Stp

is required for normal development of petals and sta-mens. Mutant flowers contain petals with green sepaloidstreaks, and two stamens show partial conversion to car-pels (Monti and Devreux, 1969). This phenotype is charac-teristic of mutations in class B floral homeotic genes.Consistent with this, a more complete transformation of pet-als to sepals and stamens to carpels was found in a moresevere mutant,

stp-2

(Ferrandiz et al., 1999). However,

stp-2

mutant flowers are also characterized by the production ofectopic flowers, a phenotype inconsistent with a strictly ho-meotic role, and it was suggested that

Stp

may play an im-portant role in controlling growth of the common petal/stamen primordia in pea flowers (Ferrandiz et al., 1999). Thisarticle describes two additional alleles at

Stp

. Phenotypicanalysis of these two mutants indicated that

Stp

, in additionto its role in the flower, is involved in leaf and inflorescencedevelopment. Many of the effects of

stp

on flower develop-ment in pea are also seen in

fim

mutants from Antirrhinumand

ufo

mutants from Arabidopsis (see Simon et al., 1994;Ingram et al., 1995, 1997; Levin and Meyerowitz, 1995;Wilkinson and Haughn, 1995). Molecular analysis revealsthat

Stp

is the pea ortholog of

Fim

/

UFO

. Together,

Stp

and

Uni

define a developmental pathway that may regulate thegrowth pattern of all shoot-derived meristems in pea.

Figure 1. Schematic Illustration of Heteroblastic Leaf Developmentand Inflorescence Morphology of Wild-Type and stp Mutant Plants.

(A) Wild-type plants always produce two scale-leaves before thefirst compound leaf, which bears two leaflets, at node 3 (n3). Thepea leaf increases in complexity during ontogeny by the addition oftendrils and leaflets. After floral induction, the main apex is con-verted to an indeterminate primary inflorescence that bears second-ary inflorescences in the leaf axils. Secondary inflorescencestypically bear one or two flowers before terminating in a hairy stub.(B) stp plants produce three scale-leaves and the first compoundleaf at node 4 (n4). Heteroblastic development of the compound leafis delayed, and leaves bearing more than two leaflets may not beproduced. The primary inflorescence of stp plants often terminatesin an aberrant flower, and the number of flowers borne on secondaryinflorescences is reduced.The nodes of leaflet changes and floral initiation are given as nx,where x denotes the node number. Typical values for wild-type andstp plants are given (see Table 1), and only significant changes inleaf form are shown. Each tick on the main axis represents a com-pound leaf–bearing node.

Stp and Pea Development 33

RESULTS

Morphology of Wild-Type Garden Pea

The wild-type pea is a caulescent model plant that bears zy-gomorphic flowers typical of the Papilionoideae. A sche-matic representation of a wild-type pea plant and secondaryinflorescence is illustrated in Figure 1. Pea flowers are pen-tamerous, with five sepals fused at the base forming a cup.As illustrated in Figure 2A, the five petals are colored anddifferentiate into three forms. The standard petal is largest(25 to 30 mm across in domestic cultivars) and adaxial (up-permost); there are two wing petals laterally and two petalsthat fuse to form the keel (Figure 2A). Enclosed within thekeel are 10 stamens—nine fused and one free—that sur-round a single central carpel (Makasheva, 1983; Tucker,1989; Reid et al., 1996). The five petals and five of the 10stamens are derived from four common primordia that arisesoon after the adaxial sepal primordium appears. For a de-tailed description of floral ontogeny in pea, see Tucker(1989) and Ferrandiz et al. (1999).

Flowers are borne on leafless secondary inflorescences thatarise in leaf axils on the primary inflorescence (Figure 1); in pea,flowering and the production of secondary inflorescences aresynonymous. Primary and secondary inflorescences are mor-phologically distinct. Secondary inflorescences terminate in ahairy stub after producing one or two flowers (Tucker, 1989;Reid et al., 1996). The primary inflorescence bears leaves and,except for secondary inflorescence production, appears iden-tical to the vegetative shoot from which it is derived. Pealeaves are compound and show regional differentiation; proxi-mal pinnae are leaflets, distal ones tendrils; and two leafystipules flank the petiole base (Makasheva, 1983; Marx, 1987).The terminal tendril appears to be a continuation of the leaf ra-chis but is indistinguishable from lateral tendrils. There is aconsistent heteroblastic progression in the complexity of pealeaves; the first two nodes bear rudimentary scale leaves, andthe first compound leaf occurs at node 3 (Figure 1). Leaf de-velopment occurs through the sequential addition of tendrilsand leaflets at a steady rate until the most complex (adult) leafform, typically with six leaflets, is produced (Makasheva, 1983;Wiltshire et al., 1994).

Origins and Characterization of the stp-3 andstp-4 Mutants

Two mutations with pronounced effects on floral morphol-ogy were isolated from an ethyl methanesulfonate mutagen-esis program on cv Torsdag (HL107). Both mutants werecharacterized by partial or complete homeotic transforma-tions of petals into sepals and the presence of additionalcarpels at the expense of stamens. A further distinguishingfeature of the more severe mutant was the production of ec-topic secondary floral meristems on elongated pedicels

Figure 2. stp Allelic Series.

(A) Wild-type flower illustrating normal morphology of the standard(st) and wing (w) petals. The two fused keel petals (k) are just visiblebetween the two wings. Within the keel are 10 stamens and a singlecentral carpel.(B) stp-1 flower with petals and stamens removed to show twofused carpels.(C) stp-1 flower with green streaks through the standard petal (whitearrow) and unfused keel petals (black arrow).(D) Flowers from stp-3 plants showing a green streak through thestandard petal (arrow).(E) stp-3 flower illustrating homeotic transformation of petals to se-pals (arrows) and stamens to carpels (c).(F) stp-4 flower lacking petals and stamens, with the ectopic floweralways present in stp-4 mutant flowers (x).(G) stp-4 flower with the ectopic flower always present (x) and addi-tional ectopic flowers on elongated pedicels (arrows) surroundingthe central carpel (c).

34 The Plant Cell

but differed in severity among populations grown at differenttimes (e.g., cf. Figures 2D and 2E). This may reflect an envi-ronmental effect on expression, as noted for stp-1 (Montiand Devreux, 1969). Severity of petal and stamen transfor-mation was related such that plants with more normal petaldevelopment also possessed more normal stamen develop-ment. Lateral shoots produced late on stp-3 mutant plantsbore flowers with a more severe phenotype that occasion-ally produced an ectopic flower on an elongated pedicel(not shown). These ectopic flowers were primarily com-posed of sepals and sepaloid carpels.

Primary flowers (those borne directly on secondary inflo-rescences) of the most severe mutant, stp-4, always con-tained five sepals in the first whorl and a central carpel. Thesecond whorl was composed of four to eight sepals. Withinthese, and surrounding the central carpel, were a variablenumber of additional carpels, carpel-sepal chimeric organs,or sepals. There was always one ectopic (secondary) floweron an elongated pedicel in primary flowers of stp-4 mutants(Figures 2F and 2G). This was found above the central carpelof the primary flower and was contiguous to the carpel base(Figure 2G). Where additional ectopic flowers were found,they surrounded the central carpel. Ectopic flowers had oneto eight sepals outermost, which were usually in a whorled orpartially spiraled arrangement (Figure 2G). Within this whorl,the flower was disorganized. Although occasionally followingthe pattern of primary flowers, ectopic flowers often con-sisted of irregularly placed sepals, carpelloid sepals, andcarpels. Petals, sepaloid petals, staminoid petals, and sta-mens (single or in pairs) were also found in secondary andtertiary ectopic flowers in varying proportions but were neverseen in primary flowers (data not shown). If stamens and pet-als were found, they usually occurred together, although rarestamens and/or petals were found in ectopic flowers thatconsisted mostly of sepaloid and carpeloid organs. Second-

within the primary flower. Backcrosses to their initial lineHL107 confirmed that both mutants showed single-gene re-cessive inheritance. Allelism tests between these mutantsand the type line for stp (JI2163) revealed that these twonew mutants represented more severe lesions at the previ-ously identified Stp locus in pea (Monti and Devreux, 1969).Therefore, the new alleles were named stp-3 and stp-4, fol-lowing the numbering system of Ferrandiz et al. (1999). Thestp-2 allele described by Ferrandiz et al. (1999) was unavail-able for this study.

A reinvestigation of the stp-1 mutant phenotype from thepure breeding type line JI2163 confirmed the analysis per-formed by Monti and Devreux (1969). The primary effect ofstp-1 on flower development is a partial conversion of sta-mens on either side of the free adaxial stamen to carpels(Figure 2B). This homeosis ranged from slightly carpelloidstamens to fully formed carpels and was seen in all flowersexamined. These carpelloid stamens were usually attachedto the central carpel. In addition, wing, keel, and, morerarely, standard petals contained green sepal-like tissue instreaks through the center of the organ. Fusion of the twokeel petals was often disrupted (Figure 2C).

The stp-3 mutant has a more severe phenotype than doesstp-1. Green streaks were always found through the centerof the petals (Figure 2D), keel petals failed to fuse, and in themost severe cases, all petals were converted to sepal-likeorgans (Figure 2E). Stamen-to-carpel transformations werealso observed: the two adaxial stamens were converted tocarpels or sepal/carpel chimeric organs. The free adaxialstamen was often completely converted to a carpel. Abaxialstamens formed groups of two or three connected by tubetissue that normally joins the nine fused stamens in wild-type flowers. The central carpel was unaffected and able toproduce seed after pollination. The floral phenotype of stp-3mutants was consistent within populations grown together

Table 1. Characteristics of stp-3 and stp-4 Mutant Plants and Their Wild-Type Siblingsa

Characteristic Stpb stp-3 tc Stpb stp-4 t

Stem length (cm) 49.44 6 0.65 43.41 6 0.98 5.13 50.44 6 0.39 45.20 6 0.59 7.41Node of first leaf with three or more leaflets 11.27 6 0.11 15.75 6 0.38 11.27 11.48 6 0.11 20.60 6 0.51d 17.48Node of flower initiation 15.95 6 0.06 18.73 6 0.20 13.31 16.29 6 0.08 20.09 6 0.21 16.84Total nodes 20.24 6 0.14 31.18 6 0.62 17.21 20.56 6 0.14 31.73 6 0.68 16.09Flowerse NDf ND — 1.69 6 0.06 1.28 6 0.09 3.82

a Data shown are mean 6SE. Student’s t test was calculated between wild-type and mutant siblings (P , 0.001 in all cases).b Stp values were derived from wild-type siblings from populations segregating for either stp-3 or stp-4: wild type, 41 individuals; stp-3 or stp-4,11 individuals.c t, Student’s t value.d Only five stp-4 plants produced leaves bearing three or more leaflets; this value is from these five plants.e Number of flowers present on the first secondary inflorescence scored from wild-type and stp-4 segregants from two separate populations:wild type, 55 individuals; stp-4, 29 individuals.f ND, not determined.

Stp and Pea Development 35

ary flowers consisting entirely of spirally arranged sepalswere also occasionally observed.

Secondary flowers also often possessed their own ec-topic (tertiary) flowers (not shown). By this stage, thewhorled arrangement of organs was completely abandoned.Outer organs of tertiary flowers were generally sepaloid; in-ner organs were variable and could be sepal-, carpel-, petal-,or stamen-like in appearance, or chimeras of these. Sepal-oidy was most prevalent.

Pleiotropic Effects of stp-3 and stp-4

In addition to their profound effects on flower development,stp-3 and stp-4 mutations delayed the node of flower initia-tion (the node bearing the first secondary inflorescence; Ta-ble 1 and Figure 1). This delay was more pronounced undershort-day (non-inductive) conditions (not shown). In aHL107 background, wild-type plants typically produce twoflowers on the first secondary inflorescence. In contrast,secondary inflorescences of stp-4 plants regularly bore onlya single flower such that, on average, there were 0.4 fewer“flowers” on stp-4 inflorescences (Table 1), suggesting arole for Stp in secondary inflorescence development. Inde-terminacy of primary inflorescences was also affected bystp mutations (Table 1). Growth of the primary inflores-cences of stp-3 and stp-4 plants was occasionally arrestedwith an apparently terminal flower (Figure 3A). Under an 18-hr photoperiod, these terminal flowers were produced afteras few as five (although usually more) reproductive nodeshad formed and were often associated with a simplificationof the leaf from compound to unifoliate in stp-4 mutantplants (Figure 3A). Wild-type plants typically produced onlyfour or five reproductive nodes before undergoing mono-carpic senescence and apical arrest, whereas stp-3 and stp-4mutants not producing a terminal flower often produced.12 reproductive nodes and retained potential for furthergrowth (Table 1). The stp-4 mutation was also associatedwith a release from dormancy of the two buds present inleaf axils on the primary inflorescence (Figure 3B). Thesebuds are normally suppressed in the primary inflorescence

Figure 3. Characterization of the stp Mutant Phenotype.

(A) Primary inflorescence of an stp-4 plant apparently terminating ina sepaloid flower (arrowhead). The final leaf produced on this pri-mary inflorescence is unifoliate.(B) Primary inflorescence from an stp-4 plant that has released addi-tional axillary shoots from suppression. Three axillary shoots are vis-

ible at each node. Arrowhead 1 indicates the secondary infloresencebearing a flower. Arrowhead 2 shows a lateral shoot bearing a leafand a terminal sepaloid flower. Arrowhead 3 indicates the develop-ing third axillary bud. A similar pattern is seen in all three nodes illus-trated.(C) Seedling phenotype of stp-4, wild type (WT), and uni-224. Ascale-like leaf in stp-4 mutants (arrowheads) replaces the first com-pound leaf produced at node 3 on wild-type plants. In contrast, uni-224 seedlings possess unifoliate leaves at all early nodes (arrow).

36 The Plant Cell

Figure 4. Interactions between stp-4 and Mutations Affecting Leaf and Inflorescence Development in Pea.

Stp and Pea Development 37

of wild-type peas so that only secondary inflorescences de-velop.

The stp-3 and stp-4 mutants also showed pleiotropic effectson leaf development. Normal heteroblastic development ofpea leaves occurs through an increase in the number ofpinnae up to a typical maximum of six tendrils and six leaf-lets (Wiltshire et al., 1994). Transition from two to four leaf-lets per leaf and four to six leaflets per leaf appears to begenetically controlled and occurs at specific nodes (Wiltshireet al., 1994). stp-3 and stp-4 mutations delayed productionof the first foliage leaf from node 3 to node 4; wild-typeplants normally possess two scale leaves, with the firstcompound leaf at node 3 (Figures 1 and 3C). Infrequently,the first compound leaf on stp-4 mutant plants occurred atnode 5, and four scale leaves were produced. In addition,both mutants significantly delayed the transition from two tofour leaflets per leaf (Table 1 and Figure 1), and some stp-4plants never produced leaves bearing more than two leaf-lets. Neither stp-4 nor stp-3 single mutant plants ever pro-duced leaves bearing five or six leaflets. An example of theeffect of stp-4 on wild-type leaf development can be seen inFigure 4A.

Measurements of internode lengths revealed that stp-3and stp-4 plants were marginally but significantly (P ,

0.001) shorter than their wild-type siblings (Table 1). In pea,shorter internodes may result from fewer cells per internode,from smaller cells, or from a combination of the two (Murfet,1990). Measurements of epidermal cell length from inter-nodes of stp-3, stp-4, and wild-type siblings suggested thatthe reduced internode length of stp-3 and stp-4 plants re-sulted from a reduction in cell number at each internode andnot from reduced cell length (Table 2). A similar reduction incell number was found in leaf petioles of stp-3 and stp-4plants in comparison with their wild-type siblings (Table 2),indicating that the reduction in cell number is not specific tothe main axis. These results suggest that Stp may playsome role in regulating cell proliferation.

Observed pleiotropic phenotypes were found in both stp-3and stp-4 plants, and were typically correlated with severityof floral phenotype. In addition, these pleiotropic traits coseg-regated with the floral phenotype in all crosses examined.Thus, it was concluded that leaf simplification, reduction ininternode length, and effects on inflorescence developmentwere pleiotropic effects of the stp mutant alleles and did notresult from a second, unrelated mutation. The relationshipbetween the observed phenotypic effects of stp-3 and stp-4and their effects on cell number in leaf rachis and internodeis unknown.

Interaction between Leaf Homeotic Mutants and stp

To clarify the effect of stp mutants on leaf development, weexamined the leaf phenotype of stp-4 in three genotypicbackgrounds: tendril-less (tl), in which pinnae are all leaflets;afila (af), in which pinnae are all tendrils; and the double mu-tant pleiofila leaf form (af tl double mutant). The stp-4 tl dou-ble mutant plants possessed fewer pinnae than did theirwild-type siblings (Figure 4B). Counts of leaflet production instp-4 tl and Stp tl plants showed that stp-4 reduced themaximum number of leaflets borne on a tl leaf (data notshown). af stp-4 plants had fewer primary and secondaryramifications typical of the af single mutant such that doublemutant leaves typically consisted of simple tendrils arisingfrom the central rachis (Figure 4C). In contrast, afila (af Stp)leaves bear compound proximal tendrils and simple tendrilsdistally (Lu et al., 1996). However, effects of stp-4 on leafdevelopment were most pronounced in the double mutantaf tl (pleiofila) leaf form (Figure 4E). Complex pinnae normallyseen in pleiofila leaves were severely reduced in the triplemutant, which produced leaves consisting of three similarunits bearing leaflets smaller than those on wild-type plantsbut larger than those normally present on pleiofila leaves(Figure 4E). The additive phenotype of stp-4, tl, and af

Figure 4. (continued).

In each case, the stp-4 mutant is at left and the corresponding Stp segregant is at right.(A) Primary inflorescences from an stp-4 single mutant and its wild-type sibling illustrating the effect of stp-4 on leaf development at the onset offlowering. The leaf from the mutant possesses only two leaflets and four tendrils, whereas five leaflets and five tendrils are present on the corre-sponding wild-type leaf.(B) The most complex leaves born on tl stp-4 and tl Stp siblings. The tl mutant converts tendrils to leaflets. Fewer leaflets were produced on thestp-4 tl leaves than on the Stp tl leaves.(C) and (D) af stp-4 and af Stp (JI1195) adult leaves, respectively, showing the reduction in leaf complexity caused by stp-4 on the af phenotype.(E) Leaves from similar nodes of the triple mutant stp-4 af tl. These also show a simplification of the leaf by the stp-4 mutation when comparedwith the pleiofila, Stp af tl leaf form. In addition to reducing the ramification of the af tl leaf, addition of the stp-4 allele also resulted in an increasein leaflet size.(F) Mature, primary inflorescences from stp-4 det and Stp det siblings. The det mutation results in the conversion of the primary inflorescenceinto a secondary inflorescence bearing flowers, which produce pods. This phenotype is also seen in the stp-4 background. The converted inflo-rescence is indicated by an arrow in each case.(G) Secondary inflorescences from stp-4 veg2-2 and Stp veg2-2 plants. The proliferation of the secondary inflorescence that results from theveg2-2 mutation is apparent in the stp-4 mutant background; however, the secondary inflorescences terminate in a sepaloid flower (arrow).

38 The Plant Cell

Table 2. Epidermal Cell Lengths (mm) and Cell Numbers from Internode 7–8 and Leaf Petiole from Node 8 of stp or uni Mutants and Their Wild-Type Siblings (mean 6SE)

Internode Leaf Petiole

Genotype and t Valuea Cell Length Cell Number Cell Length Cell Number

Stpb 452.5 6 15.0 240.6 6 12.2 376.6 6 8.0 170.4 6 3.8stp-3 467.2 6 9.0 187.3 6 5.8 369.8 6 8.0 139.0 6 2.9t Value 0.84 3.94c 0.60 6.57d

Stpb 468.2 6 9.0 219.6 6 5.0 372.5 6 5.4 171.6 6 5.3stp-4 453.7 6 14.0 194.1 6 8.0 361.3 6 5.8 131.0 6 4.0t Value 0.87 2.69e 1.41 6.10d

Unib 522.5 6 32.7 206.0 6 18.9 NDf NDuni-224 505.3 6 72.1 158.1 6 6.3 ND NDt Value 0.36 2.4e — —

a Student’s t value from the mutant and their wild-type siblings.b Stp and Uni values were derived from wild-type siblings of populations segregating for stp-3, stp-4, or uni-224.c P , 0.01.d P , 0.001.e P , 0.05.f ND, not determined.

mutants suggests that Stp acts independently of genescontrolling the final form of pea leaves.

Interactions between stp and InflorescenceIdentity Genes

The effect of stp-4 on primary and secondary inflorescencedevelopment was examined more closely in two mutant back-grounds affecting inflorescence identity in pea: determinate(det; Marx, 1986a; Singer et al., 1990) and vegetative2-2(veg2-2; Reid et al., 1996). The det mutant promotes a sec-ondary inflorescence identity in the primary inflorescence,often converting the apical meristem into a terminal stub(Singer et al., 1990; Figure 4F). Conversely, veg2-2 mutantsbear indeterminate secondary inflorescences that are pri-mary inflorescence–like and produce compound leaves(Reid et al., 1996; Figure 4G). Thus, Det and Veg2 may beinvolved in the specification of primary and secondary inflo-rescences, respectively. The inflorescence structure charac-teristic of det mutant plants was also seen in stp-4 detdouble mutant plants (Figure 4F), and det appeared unableto suppress the development of ectopic flowers within stp-4primary flowers (Figure 4F). The production of terminalflowers in stp-4 plants was unaffected by veg2-2, and theleaf-bearing proliferating secondary inflorescences of stp-4veg2-2 plants often terminated in an aberrant flower (Fig-ure 4G). The simple additive phenotypes of the stp-4 detand the stp-4 veg2-2 double mutants suggest that Stpfunctions independently of Det and Veg2 and that there-

fore Stp is not involved in primary or secondary inflores-cence identity.

Interactions between stp and uni

The effects of stp-4 on both leaf and flower development arevery similar to those seen in unifoliata (uni) mutants of pea(Marx, 1987; Hofer et al., 1997). Epidermal strips revealedthat the severe uni-224 allele also affects cell number in amanner similar to stp-3 and stp-4 (Table 2). Crosses be-tween plants carrying uni-224 and stp-4 indicated that thetwo genes are not allelic, which raised the possibility thatStp and Uni were involved in the same developmental path-way. Segregation of stp-4 and uni-224 phenotypes in F2 anddihybrid F3 populations was in accordance with a 9:3:4 ratioof wild-type/stp-4/uni-224 (214 wild type, 72 stp-4, and 89uni-224; x2 5 0.33; P . 0.8), indicating that uni-224 is com-pletely epistatic to stp-4. The epistatic relationship of uni-224 to stp-4 encompassed effects of uni on leaf, flower, andinflorescence structure, including production of a small leaf-let at node 2 (Figure 3C), which was present on all uni-224segregants. The epistatic interaction and similar phenotypeof the severe uni-224 and stp-4 mutants suggest that Uniand Stp act in the same developmental pathway. This possi-bility was further supported by the synergistic interactionbetween known weak alleles at Uni and Stp. The weakestmutant allele at Stp, stp-1, does not appear to affect leaf de-velopment, and its mutant phenotype is floral specific (Montiand Devreux, 1969; Figures 2B, 2C, and 5A). In contrast, theweakest described allele at Uni, uni-tac, primarily affects

Stp and Pea Development 39

leaf development, although it can also affect flower develop-ment (Marx, 1986b; Figures 5A and 5B). An F2 populationderived from the cross between the type lines carrying stp-1and uni-tac segregated into four distinct classes: 65 wildtype, 16 stp-1, 10 uni-tac, and 5 stp-1 uni-tac, correspond-ing to a 9:3:3:1 ratio (x2 5 6.11; P . 0.1). Leaves of the dou-ble mutant plants were simplified and rarely possessed thesubterminal tendrils characteristic of uni-tac (Figure 5A). Thedouble mutant plants also had flowers consisting entirely ofsepals and carpels (Figure 5C). Sepal/carpel mosaic organswere noted, and ectopic flowers were also seen. Leaf andflower phenotypes of the putative stp-1 uni-tac double mu-tant were confirmed in the F3 progeny from stp-1 segre-gants. The stp-1 and uni-tac single mutants segregating inthe cross produced typical flower and leaf phenotypes (Fig-ure 5B). The phenotype of stp-1 uni-tac double mutant flow-ers was reminiscent of that produced by severe stp and unimutants, and the leaf phenotype was similar to those pro-

duced by uni mutants, suggesting that Stp and Uni act to-gether to regulate plant development.

Molecular Characterization of Stp

The floral phenotypes of stp mutants and the genetic inter-actions between stp and uni mirror those described for fimin Antirrhinum and ufo in Arabidopsis (Simon et al., 1994; Levinand Meyerowitz, 1995; Wilkinson and Haughn, 1995; Ingramet al., 1997). This raised the possibility that Stp representedthe pea ortholog of Fim and UFO. Homology between Uniand Flo/LFY (Hofer et al., 1997) and the role of Uni in leaf de-velopment also support homology between Stp and Fim/UFO. Furthermore, observed effects of stp-3 and stp-4 mu-tations on cell division (Table 2) are consistent with the pro-posed role of Fim and UFO in cell proliferation (Wilkinsonand Haughn, 1995; Ingram et al., 1997; Meyerowitz, 1997).Therefore, the relationship between Stp and Fim/UFO wasexamined at a molecular level.

A cDNA clone containing a complete open reading frameand flanked by putative untranslated 59 and 39 regions and apoly(A) tail was isolated from a pea flowering-apex cDNA li-brary probed with a polymerase chain reaction (PCR)–de-rived fragment from the conserved 39 coding region of UFO.The complete nucleotide and deduced amino acid se-quences of this clone are displayed in Figure 6. Databasesearches revealed significant similarity between the isolatedcDNA and Antirrhinum and Arabidopsis Fim and UFO se-quences, and to Imp-FIM from impatiens (Pouteau et al.,1997). Cross-hybridizing bands were not detected on DNAgel blots probed with peafim and washed at low stringency,indicating that this clone was present in the pea genome asa single copy. This cDNA clone was named peafim and islikely to represent the pea ortholog of Fim/UFO. Alignmentof amino acid sequences, presented in Figure 7, revealedlarge areas of conservation between pea, Antirrhinum, andArabidopsis sequences and high overall amino acid identitybetween peafim and Fim (63.7%), peafim and UFO (61.1%),and across the 179–amino acid Imp-FIM fragment (62.1%).

To investigate the relationship between peafim and Stp,we first mapped the peafim clone onto a pea molecular map(see Hall et al., 1997). Peafim showed strong linkage tocharacters on pea linkage group VII, shown in Figure 8A,and fell between Cab1 (for chlorophyll a/b binding protein)and Rrn2 (for rRNA gene cluster2). Linkage analysis hadpreviously placed stp in linkage group VII (Monti, 1970). Thismap location was confirmed using the more severe allelestp-4. Stp showed strong linkage (P , 0.0001) with both wa,a waxless mutant, and the isozyme locus Skdh with a mapsequence of Skdh (10 centimorgans [cM]) stp (4 cM) wa(Figure 8C). Figure 8 shows an alignment of the classicaland molecular maps, which indicates that peafim and Stpreside in the same region of linkage group VII in pea. Coseg-regation analysis confirmed close linkage of Stp and peafim:a restriction fragment length polymorphism detected by the

Figure 5. Interaction between Weak Alleles at stp and uni.

(A) Flowers and adult leaves from stp-1 Uni (stp-1), stp-1 uni-tac,and Stp uni-tac (uni-tac) siblings. stp-1 and uni-tac are known to beweak alleles; however, stp-1 uni-tac plants bore leaves that weresimpler than those of either single mutant.(B) uni-tac flower with unfused keel petals (arrow).(C) stp-1 uni-tac double mutant flower. The arrows indicate petalsconverted to sepals. c, carpel.The leaf and floral phenotype of the stp-1 uni-tac double mutant ishighly reminiscent of the severe uni-224 mutant.

40 The Plant Cell

peafim cosegregated with 10 stp-4 mutant plants segregat-ing in an F2 population of 50 individuals. Thus, the mapposition and cosegregation analysis supported the pro-posed relationship between Stp and peafim.

To resolve the relationship between peafim and stp, weamplified peafim from genomic DNA of isogenic Stp(HL107), stp-3, and stp-4 lines. Sequence analysis of ampli-fication products from stp-3 and stp-4 mutants revealed dif-ferent single-base substitutions compared with Stp fromHL107. The stp-4 mutant plants had a G-to-A nucleotidechange at base 1111 that resulted in a stop codon at posi-tion 252 in the deduced amino acid sequence (Figure 6). Thestp-3 sequence had a G-to-A nucleotide change at base1051, which would result in a nonconservative alanine tothreonine substitution at position 232 in the predicted aminoacid sequence (Figure 6). This residue is conserved in allfour sequences (Figure 7) and is therefore likely to be impor-tant for normal function of peafim protein. The nucleotidechanges in peafim sequence from stp-3 and stp-4 plantsand their predicted effects on the translated protein reflectthe relative severity of the two mutant phenotypes.

Expression of peafim

Because peafim transcripts were undetectable by RNA gelblot analysis, we analyzed the spatial distribution of peafimexpression by using in situ hybridization on serial sections of15-day-old (vegetative) seedling apices and flowering apicesof wild-type pea plants (Figure 9). Expression of peafim withinfloral meristems (Figures 9A and 9B) was comparable to thatdescribed for UFO in Arabidopsis and for Fim in Antirrhinum(Simon et al., 1994; Ingram et al., 1995, 1997; Lee et al., 1997)and occurred early during floral development. This expres-sion preceded the appearance of the common petal/stamenprimordia (defined as stage 5 by Ferrandiz et al. [1999]) andformed a ring around the central dome of floral primordia inthe region of incipient common primordia (Figures 9A and 9B,f1). After initiation of these primordia, Stp expression formeda cup shape delineating presumptive petal primordia (Figures9A and 9B, f2). Late expression of peafim in flowers, like thatof Fim and UFO, was confined to the base of petal primordia.The similar expression patterns of peafim, UFO, and Fim inflowers are consistent with the similar floral phenotypes ofufo, fim, and stp mutants.

Expression of peafim was also found in the axils of leafprimordia (Figure 9C). Serial sections indicated that a semi-circular wedge of expression existed in leaf axils up until atleast plastochron 7 (P7), after which it became barely dis-cernible (not shown). Expression was also observed withinleaf primordia at P2, potentially representing a delineation ofthe first pair of leaflet primordia (Figure 9D). Weak expres-sion was also detected within the apical meristem (Figure9C). The presence of peafim transcript in flowers, apicalmeristem, leaf axils, and leaf primordia is consistent with thepleiotropic effects of the two stp mutations.

Figure 6. Nucleotide and Predicted Amino Acid Sequence from thecDNA Clone peafim.

The positions of the point mutations present in the stp-3 and stp-4mutant alleles are shown in boldface. The G-to-A substitution in stp-4is predicted to result in premature termination of the protein. Thestp-3 mutation is predicted to result in a nonconservative substitu-tion of an alanine with a threonine residue. The asterisk denotes thestop codon.

Stp and Pea Development 41

The map position of peafim and Stp, cosegregation analysis,and sequence analysis of the stp mutations support orthologybetween Stp, UFO, and Fim. Identification of such homologiesprovides an opportunity to compare pathways proposed toregulate development in pea, Arabidopsis, and Antirrhinum.

DISCUSSION

We have shown that Stp influences development of leaves,inflorescences, and flowers of pea. These pleiotropic effectsare summarized in Figure 1. Analysis of a cDNA ortholog ofFim and UFO from the garden pea has provided strong evi-dence for orthology among Stp, Fim, and UFO. Although thefloral phenotypes of stp, fim, and ufo mutants are very simi-lar, the leaf phenotype associated with stp mutations isunique to pea. Expression of Stp within leaf primordia sug-gests that the leaf phenotype of stp-3 and stp-4 is indeed aconsequence of those mutations. Stp acts synergisticallywith Uni but is genetically independent of homeotic genescontrolling leaf and inflorescence development.

Stp and Flower Development

All four stp mutant alleles are characterized by homeoticchanges within the flower (Monti and Devreux, 1969; Ferrandiz

et al., 1999; this work). The phenotype of stp-1 consists en-tirely of floral homeotic changes characteristic of class B floralhomeotic mutants (Ferrandiz et al., 1999; this work). Sepaloidtissue in petals of stp-1 flowers is also seen in weak Fim mu-tants (Simon et al., 1994). The severe floral phenotype of stp-4mutants is also very similar to those of severe fim mutants.Both tend to produce flowers composed primarily of sepals,although petalloid and carpelloid tissues are also produced.The production of normal carpels in stp-4 mutants possibly re-flects the early production of carpel primordia in pea (Tucker,1989) or an inability of Stp to regulate C-class homeotic geneactivity. The severe stp-4 mutant appeared less like severe ufomutants of Arabidopsis because the filamentous structuresand reduced or empty flowers were not produced on stp-4mutant plants (Levin and Meyerowitz, 1995; Wilkinson andHaughn, 1995). However, homeotic changes within the flowerare comparable, and termination of ufo mutant inflorescencesby a carpelloid structure was also reflected in the determi-nancy of stp-3 and stp-4 primary inflorescences (Figure 3A).Furthermore, the delay in node of flower initiation caused bystp-3 and stp-4 mutations (Table 1) may be considered equiv-alent to the increase in coflorescence production in ufo mu-tants (Wilkinson and Haughn, 1995), because both represent adelay in production of the first flower. Whether the delay inflower initiation in stp mutants results from a homeotic conver-sion of early secondary inflorescences to vegetative shoots orfrom a delay in attaining reproductive competence is unclear.

The most significant difference among the floral phenotypes

Figure 7. Alignment of the Deduced Amino Acid Sequences Encoded by Stp, Fim, UFO, and Imp-FIM (Impat).

Amino acids present in at least three sequences are shaded; hyphens indicate gaps due to alignment. The F-box motif is underlined.

42 The Plant Cell

of Stp, UFO, and Fim mutants is the production of ectopicflowers within the primary flower (Figure 2). Ectopic flowershave not been described in ufo mutant flowers. Flowers ofsevere Fim mutants produce secondary flowers in the axilsof sepals, and their overall phenotype is that of a proliferousinflorescence (Simon et al., 1994; Ingram et al., 1997). Flow-ers produced on stp-3 or stp-4 plants did not possess typi-cal characteristics of pea inflorescences (e.g., a terminalstub), although the production of ectopic flowers may indi-cate some inflorescence identity. Placement of theseectopic flowers was consistent with their derivation from com-mon petal/stamen primordia, as suggested by Ferrandiz etal. (1999). The position of the last common primordium todifferentiate (e.g., see Tucker, 1989; Ferrandiz et al., 1999)corresponds to the position always occupied by an ectopicflower on stp-4 mutants and the position occasionally occu-pied by an ectopic flower on stp-3 mutants. This primor-dium, as the last to differentiate into individual petal andstamen primordia, appears to be more sensitive to loss ofStp activity. Presumably, the absence of normal Stp activityin stp mutants led to the common primordia reiterating anearlier program—that of the flower—rather than to dividingand differentiating into the petals and stamens. This possi-bility is supported by the presence of Stp transcripts withinthe common primordia (Figures 9A and 9B). The proposedrole Stp plays in the differentiation of these primordia andthe homeotic changes seen in stp mutant flowers is consis-tent with the proposed role Fim and UFO play in regulatingclass B gene activity.

Stp and Leaf Development

That Stp is essential for normal leaf development is seenclearly by a loss of the first foliar leaf at node 3, retarding of

heteroblastic development, and simplification of the lastleaves produced on stp mutants. These effects are mostclearly seen in combination with the leaf homeotic mutantsaf and tl (Figures 4B, 4C, and 4E); however, the additivephenotypes of the double and triple mutants suggest thatStp acts independently of Af and Tl.

Although stp-4 mutants have a floral phenotype very simi-lar to that seen in severe uni mutants, stp-4 plants alwaysproduce pinnate adult leaves. In contrast, severe uni mu-tants bear simple to trifoliate leaves (Hofer et al., 1997). Se-quence analysis suggests that the stp-4 mutant proteinwould comprise only the N-terminal half, suggesting a se-vere disruption of protein function (by comparison, thiswould have a greater effect on the protein than ufo-2 andufo-5 alleles, which have a strong mutant phenotype [Lee etal., 1997]). Therefore, rather than assume that stp-4 is insuf-ficiently severe to simplify the leaves, we suggest that Stp isnot an essential requirement for the difference between sim-ple and compound leaves. Rather, Stp appears to increasecomplexity of an already compound leaf, an effect seenmost clearly in the pleiofila (af tl double mutant) leaf form(Figure 4E). Either Uni is sufficient alone or additional genesmust act with Uni to produce a pinnate leaf.

Expression of Stp in the base and axils of leaf primordiamight explain differences between leaf phenotypes of uniand stp mutants. Stp is not expressed in distal regions ofwild-type leaf primordia older than P1, and stp mutants donot affect the identity of the terminal structure, which retainstendril identity. In contrast, Uni transcripts are detected atthe apex of wild-type leaf primordia until P4 (Hofer et al.,1997), and the terminal tendril is lost in uni mutants. Al-though Uni and Stp interact genetically (e.g., Figure 5A),their expression patterns do not completely overlap, partic-ularly within older leaf primordia (Hofer et al., 1997; Figure9C). It is possible that early overlapping expression of Stpand Uni within leaf primordia at P1 and P2 may be sufficientto coordinate their mutual control of compound leaf proli-feration. Alternately, non-cell autonomy of the Flo protein(Carpenter and Coen, 1995; Hantke et al., 1995) suggeststhat Uni could exert its influence from a distance.

Despite the severe effect uni mutations have on leaf de-velopment in pea, differences between the leaves of Arabi-dopsis and pea appear to depend more on differences ingene expression between UFO and Stp rather than on dif-ferences between LFY and Uni. UFO transcripts are not de-tected in Arabidopsis leaves, and ufo mutations do notaffect leaf development (Levin and Meyerowitz, 1995;Wilkinson and Haughn, 1995; Lee et al., 1997). Overexpres-sion of UFO in Arabidopsis does result in a lobed leaf phe-notype that is dependent on the presence of a wild-typeLFY allele (Lee et al., 1997). However, overexpression ofUFO does not produce a (pea-like) compound leaf in Arabi-dopsis, consistent with the proposal that expression of Stpin leaves is not sufficient to define the difference between acompound and a simple leaf but is only able to increase leafcomplexity. Thus, the basic difference between a com-

Figure 8. Map Location of stp and peafim.

(A) Restriction fragment length polymorphism map position of peafim.(B) Same region from the consensus pea map (Weeden et al., 1996),which has only an approximate position for stp and wa.(C) Classical mapping of stp.Dotted lines connect common markers, and only informative com-parisons are shown. Map distance data for the classical map are ap-proximate only.

Stp and Pea Development 43

pound leaf and a simple leaf depends on alternate gene ac-tivities.

Fim homologs have only been characterized in simple-leaved plants and in pea, which has specialized compoundleaves. In particular, a Fim homolog has not yet been identi-fied in tomato, and it remains possible that mutants in to-mato fim would not have a leaf phenotype, particularlybecause fa mutations have a limited effect on the tomatocompound leaf (Molinero-Rosales et al., 1999). Thus, the

role Fim homologs play in other compound-leafed plants, ifany, is yet to be determined.

Biochemical Activity of the Fim Homologs

The predicted amino acid sequences of Stp, Fim, and UFOshow a high degree of conservation around a motif knownas the F-box (Bai et al., 1996; Figure 7). F-box proteins havebeen demonstrated to be integral components of the SCFcomplex (for Skp1, Cdc53, F-box proteins; reviewed inKrek, 1998; Patton et al., 1998; Galan and Peter, 1999).These act as ubiquitin protein ligases that, in combinationwith a ubiquitin conjugating enzyme, ubiquitylate specificphosphorylated proteins, targeting them for proteolysis. TheF-box protein determines selectivity of SCF complexes,whereas Skp1 and Cdc53 are common components thatassemble into various SCFs containing differing F-box pro-tein partners. A dynamic equilibrium potentially exists be-tween a pool of Skp and Cdc53 components and thevarious F-box proteins (Patton et al., 1998; Galan and Peter,1999). Three fimbriata-associated proteins (FAPs) were iso-lated from Antirrhinum in a yeast two-hybrid screen usingFim as bait. These three FAP proteins share strong homol-ogy with Skp1 from yeast and p19skp1 from humans (Ingramet al., 1997). In a similar screen in Arabidopsis, two ARABI-DOPSIS SKP-like (ASK) proteins interacting with UFO wereisolated (Samach et al., 1999). ASK proteins also share ho-mology to Skp1. These results suggest that Fim and UFOproteins, and by extrapolation the Stp protein, may partici-pate in an SCF complex. Thus, it is possible that these plantF-box genes are involved in the targeting of proteins forubiquitin-mediated proteolysis. F-box proteins were firstidentified through their role in regulating cell division in yeastand humans (Bai et al., 1996; Krek 1998), and the observedeffect of stp mutants on cell division (Table 2) may suggestthat Stp shares this function. However, F-box proteins arenot restricted to forming SCF complexes, nor are they re-stricted to cell-cycle control, and other potential modes ofaction exist (Krek, 1998; Patton et al., 1998). Confirming thebiochemical activity of Fim homologs and determining theirpotential target proteins remain a priority.

Fim and Flo Orthologs

The nature of the interaction between Fim orthologs (Stp,Fim, and UFO) and Flo orthologs (Uni, Flo, and LFY) is un-clear. Flo orthologs potentially act as transcription factors,(Parcy et al., 1998), whereas Fim orthologs may control lev-els of specific proteins by targeting them for ubiquitin-medi-ated proteolysis. Despite this, results from pea, Antirrhinum,and Arabidopsis are consistent in suggesting that Fimorthologs function in concert with Flo orthologs to controlplant development. Within the flower, Fim and Flo orthologsregulate expression of class B floral homeotic genes and

Figure 9. In Situ Hybridization Analysis of Stp Expression in Wild-Type Pea Flowers and Seedlings.

(A) and (B) Serial sections through the same sample separated by16 mm. Stp expression in the youngest flower (f1) occurs in a ringthroughout the region of the incipient common primordia. This pat-tern appears as two discrete patches of expression in these adja-cent sections. This expression becomes restricted to the boundarybetween sepal and common primordia (f2) and delineates each pri-mordium into petal and stamen primordia (arrows). Later in flowerdevelopment (f3), Stp expression becomes restricted to the base ofthe petals.(C) Stp expression was also detected in the apical meristem and leafaxils of wild-type seedlings. Expression of Stp occurred throughoutthe incipient leaf primordium (p0) but became restricted to the leafaxil by p4. p0 to p4 identify leaf primordia initiated at plastochrons 1to 4, respectively.(D) Enlargement of leaf primordium p2 showing strong Stp expres-sion within the leaf (arrow). This expression appears to delineate thedeveloping leaflet primordia.

44 The Plant Cell

influence development of the floral primordium. They mayplay a role in inflorescence development, a function seenmost clearly by the delay in production of flowers, and in ter-minal flower production in mutants in Flo and Fim orthologs.In addition, Fim and Flo orthologs are central to compoundleaf development in pea. Results from pea suggest that Uniand Stp may regulate development of meristems, irrespec-tive of their eventual identity, and therefore play a central rolein determining the final form of a pea plant.

METHODS

Sources of Mutant Lines

Two previously uncharacterized mutations affecting flower develop-ment were isolated from an ethyl methanesulfonate mutagenesisprogram on HL107 (Pisum sativum cv Torsdag) conducted by J.L.Weller (University of Tasmania, Hobart; for details, see Weller et al.,1997). The stp-4 mutant was identified from a single M2 populationgrown under an 8-hr photoperiod, and a single wild-type M2 sibling,found to be heterozygous for stp-4, was the progenitor of the mutantline used in these analyses. The stp-3 allele was originally isolated asa large-scale-leaf mutant from a screen of seedling phenotypes. Thisplant was found to possess floral homeotic defects and was rescuedby crossing into the HL107 background. This mutagenesis programalso provided an additional allele at Uni, uni-224. Allelism with uniwas confirmed by crossing the mutant line with JI1396 (uni-tac); thetwo F1 plants possessed a uni-tac phenotype. Seed from these threepopulations were supplied by J.L. Weller. As mostly sterile mutants,stp-3, stp-4, and uni-224 alleles were maintained as heterozygouslines. Type lines for stp-1 (JI2163) and uni-tac (JI1396) were suppliedby Mike Ambrose (John Innes Centre). All other lines used were fromthe pea germplasm collection at Hobart. Type lines segregating forstp-3 and stp-4 were deposited into the Hobart germplasm collec-tion as lines HL289 and HL288, respectively.

Characterization and Genetic Interactions

Characterization of the two new stp mutant phenotypes was per-formed using mutant plants segregating from first and second back-crosses to the initial line pea cultivar Torsdag (HL107). Flowernumber per inflorescence (Table 1) was scored from two separatepopulations used in analyses of interaction between stp-4 and pim,and stp-4 and veg2-2. veg2-2 and pim segregants were not includedin this analysis.

Crosses between homozygous stp-4 and heterozygous Uni/uni-244 plants were made to determine the relationship between uni andstp. Interaction between veg2-2 and stp-4 was examined in the crossbetween line Wt16123 (veg2-2) and homozygous stp-4 segregantsfrom the original mutant line. Interactions among stp-4, af, and tlwere examined in a single cross between HL117 (af tl) and an stp-4plant from the first backcross. Interaction between stp-4 and det wasexamined in a cross between HL245 (r det veg1) and an stp-4 plantfrom the original mutant line. The det and r loci are tightly linked(Marx, 1986a), and only wrinkled (r) F2 seed were sown. The stp seg-regants in these populations were identified at the seedling stage bythe loss of compound leaf at node 3 (e.g., Figure 3C), and af, tl, det,

and veg2-2 segregants were clearly identifiable by their characteris-tic mutant phenotypes.

Linkage between stp and wa was examined in a cross betweenHL6 (Stp wa Sn) and a homozygous stp-4 plant (stp-4 Wa Sn) fromthe original population. Linkage was confirmed in a cross between anstp-4 wa Sn Aat3-F Skdh-F F3 recombinant and HL59 (Stp Wa snAat3-S Skdh-S). Recombination fractions were calculated using theproduct ratio method (Stephens, 1939), and joint segregation chi-squares were calculated using a 2 3 2 contingency table.

In Hobart, plants were grown one or two per 14-cm slimline pot ina 1:1 mix of vermiculite and dolerite chips topped with sand/peatpotting mix. All plants were grown in a greenhouse under an 18-hrphotoperiod consisting of natural daylight extended before dawnand after dusk, with a mixture of fluorescent and incandescent bulbsproviding 25 mmols m22 sec21 at pot top. Plants used in the coseg-regation analysis were grown at the John Innes Centre as describedby Hofer et al. (1997).

Epidermal Strips

Epidermal strips (Table 2) were taken from internode 7–8; the leafpetiole (between the stem and first leaflet pair) was taken from node8 from six stp-3 and stp-4 plants, and six of their wild-type siblingssegregating in the second backcross to HL107. Samples were alsotaken from internode 7–8 from six uni-224 segregants and their wild-type siblings. Cell numbers were calculated for each plant by dividinginternode or petiole length by the average of 10 cell lengths that weremeasured at random from that structure.

Isolation and Characterization of the peafim cDNA

Plaques (200,000) of a l Zap cDNA library (Stratagene), preparedfrom RNA isolated from flowering shoot apices of JI813 (HL51y),were screened with a 580-bp polymerase chain reaction (PCR) frag-ment spanning nucleotides 244 to 824 of the Arabidopsis UFO gene(Ingram et al., 1995). This fragment was produced from plasmidPJAM 180, which contains a 3.4-kb genomic DNA insert containingthe entire UFO coding region (kindly supplied by G. Ingram, JohnInnes Centre), using two primers specific to the Arabidopsis se-quence: 59 Fim oligo, TTCTCCAACACCTTCCTCGA; and 39 Fim oligo,ACGCTAAAAGGGCTATAGTTCAT.

Plasmid and PCR-derived fragments were sequenced using a Per-kin Elmer 9600 thermal cycler and dye terminator technology (Ap-plied Biosystems International, Foster City, CA), according to themanufacturer’s instructions. Sequences obtained were examinedusing Sequence Navigator (version 1.0; Applied Biosystems), SeqVu(version 1.0.1; Garvan Institute, Sydney, Australia), and Clustalw (ver-sion 1.5). The alignment in Figure 7 was produced using GCG10 lo-calpileup and prettybox functions. Nucleotide and amino acid se-quence data of the peafim clone have a GenBank accession numberof AF004843.

For cosegregation analyses between stp-4 and peafim, total ge-nomic DNA was isolated from HL107, HL111, and 50 F2 plants froma cross between HL111 and the original mutant line, following themethod of Ellis (1994). For cDNA synthesis, total RNA was isolatedfrom HL107 (Stp), JI2163 (stp-1), and stp-3 and stp-4 segregantsfrom HL 289 and HL288, following the method outlined by Michael etal. (1996). Poly(A)1 RNA was purified with an mRNA isolation kit(Boehringer Mannheim) using biotin-labeled oligo(dT)20 and strepta-

Stp and Pea Development 45

vidin magnetic particles, as described in the manufacturer’s proto-col. cDNA was synthesized from this poly(A)1 RNA, using Gibco BRLSuperscript preamplification system (Life Technologies) for first-strand cDNA synthesis, following the manufacturer’s instructions,except that 0.5 mM oligonucleotide SKTTT (59-CGCTCTAGAACT-AGTGGATCCTTTTTTTTTTTTTTTTT-39) was used as a starting pointfor cDNA synthesis. Uni (XM 71751) and uni (XM 71752) cDNA wassynthesized from total RNA using a T17 primer and M-MLV reversetranscriptase (Gibco BRL).

Oligonucleotides designed specifically to match regions of the 59

and 39 untranslated region of peafim sequence were used in PCRanalysis of total genomic DNA and cDNA from mutant and wild-typelines: for peafimV.1, 59-CTGAAAATGAGGAACAGAATCTACC-39 (98to 122); and peafimIII.1, 59-CATATTAATCACTACACATACCATACC-39

(1754 to 1780). The numbers in parentheses refer to nucleotide posi-tions shown in Figure 6; the sequence of peafimIII.1 is the reversecomplement of the sequence given in Figure 6. PCR products pro-duced from genomic DNA from Stp, stp-3, and stp-4 plants were li-gated into pGEM-T vector (Promega) and sequenced as describedabove. Wild-type and mutant sequences were confirmed in a secondand third PCR product that were purified for direct sequencing usinga QIAquick Gel extraction kit (Qiagen).

In Situ Hybridization

In situ hybridization was performed on sections of flowering and veg-etative pea apices from HL107, as described previously (Hofer et al.,1997). The digoxygenin-labeled sense and antisense probes weremade using a cloned wild-type genomic PCR product without thepoly(A) tail present in the cDNA clone. Tissue was fixed in 4% form-aldehyde overnight, dehydrated in an ethanol series, cleared in His-toclear, and embedded in Paramounat extra wax. Serial sections (8mm) were attached to poly L-lysine–coated slides and probed with ei-ther sense or antisense digoxygenin-labeled peafim probes. Slideswere developed using alkaline phosphatase–conjugated anti-digoxy-genin antibodies, 5-bromo-4-chloro-3-indolyl phosphate, and nitro-blue tetrazolium and left for 12 to 18 hr. No signal was detected usingthe sense probe (not shown).

ACKNOWLEDGMENTS

The authors thank Jim Weller for supplying the original mutant seed;Diane Lester and Noel Ellis for assistance with the molecular analysisof Stp; and Ian Cummings, Tracey Jackson, Jodie van de Kamp,Sarah Scott, and Nadine Donovan for technical assistance. We alsothank Noel Ellis, Trine Juul, Guy Wheeler, and Caroline Middlebrookfor comments on the manuscript.

Received May 22, 2000; accepted October 18, 2000.

REFERENCES

Bai, C., Sen, P., Hofmann, K., Ma, L., Goebl, M., Harper, J.W.,and Elledge, S.J. (1996). SKP1 connects cell cycle regulators to

the ubiqitin proteolysis machinary through a novel motif, the F-box.Cell 86, 263–274.

Bowman, J.L., Drews, G.N., and Meyerowitz, E.M. (1991). Geneticinteractions among floral homeotic genes of Arabidopsis. Devel-opment 112, 1–20.

Carpenter, R., and Coen, E.S. (1995). Transposon induced chime-ras show that floricaula, a meristem identity gene, acts non-auton-omously between cell layers. Development 121, 19–26.

Coen, E.S., and Meyerowitz, E.M. (1991). The war of the whorls:Genetic interactions controlling flower development. Nature 353,31–36.

Coen, E.S., Romero, J.M., Doyle, S., Elliot, R., Murphy, G., andCarpenter, R. (1990). floricaula: A homeotic gene required forflower development in Antirrhinum majus. Cell 63, 1311–1322.

Ellis, T.H.N. (1994). Approaches to the genetic mapping of pea. InModern Methods of Plant Analysis, H.-F. Linskens and J.F.Jackson, eds (Berlin: Springer-Verlag), pp. 117–160.

Ferrandiz, C., Navarro, C., Gomez, M.D., Canas, L.A., and Beltran,J.P. (1999). Flower development in Pisum sativum: From the warof the whorls to the battle of the common primordia. Dev. Genet.25, 280–290.

Galan, J.M., and Peter, M. (1999). Ubiquitin dependent deregula-tion of multiple F-box proteins by an autocatalytic mechanism.Cell Biol. 96, 9124–9129.

Hall, K.J., Parker, J.S., Ellis, T.H.N., Turner, L., Knox, M.R., Hofer,J.M.I., Lu, J., Ferrandiz, C., Hunter, P.J., Taylor, J.D., andBaird, K. (1997). The relationship between genetic and cytoge-netic maps of pea. II. Physical maps of linkage mapping popula-tions. Genome 40, 755–769.

Hantke, S.S., Carpenter, R., and Coen, E.S. (1995). Expression offloricaula in single cell layers of periclinal chimeras activatesdownstream homeotic genes in all layers of floral meristems.Development 121, 27–35.

Hofer, J., Turner, L., Hellens, R., Ambrose, M., Matthews, P.,Michael, A., and Ellis, N. (1997). Unifoliata regulates leaf andflower morphogenesis in pea. Curr. Biol. 7, 581–587.

Huijser, P., Klein, J., Lönnig, W.E., Meijer, H., Saedler, H., andSommer, H. (1992). Bracteomania, an inflorescence anomaly, iscaused by the loss of function of the MADS-box gene Squamosain Antirrhinum majus. EMBO J. 11, 1239–1249.

Ingram, G.C., Goodrich, J., Wilkinson, M.D., Simon, R., Haughn,G.W., and Coen, E.S. (1995). Parallels between UNUSUAL FLO-RAL ORGANS and Fimbriata, genes controlling flower develop-ment in Arabidopsis and Antirrhinum. Plant Cell 7, 1501–1510.

Ingram, G.C., Doyle, S., Carpenter, R., Schultz, E.A., Simon, R.,and Coen, E.S. (1997). Dual role for fimbriata in regulating floralhomeotic genes and cell division in Antirrhinum. EMBO J. 16,6521–6534.

Krek, W. (1998). Proteolysis and the G1-S transition: The SCF con-nection. Curr. Opin. Genet. Dev. 8, 36–42.

Krizek, B.A., and Meyerowitz, E.M. (1996). The Arabidopsishomeotic genes APETALA3 and PISTILLATA are sufficient to pro-vide the B class organ identity function. Development 122, 11–22.

Lee, I., Wolfe, D.S., Nilsson, O., and Weigel, D. (1997). A LEAFYco-regulator encoded by UNUSUAL FLORAL ORGANS. Curr.Biol. 7, 95–104.

46 The Plant Cell

Levin, J.Z., and Meyerowitz, E.M. (1995). UFO: An Arabidopsisgene involved in both floral meristem and floral organ develop-ment. Plant Cell 7, 529–548.

Lu, B., Villani, P.J., Watson, J.C., DeMason, D.A., and Cooke, T.J.(1996). The control of pinnae morphology in wildtype and mutantleaves of the garden pea (Pisum sativum L.). Int. J. Plant Sci. 157,659–673.

Makasheva, R.K. (1983). The Pea. (New Delhi: Oxonian Press Pvt. Ltd.)

Mandel, M.J., Gustafson-Brown, C., Savidge, B., and Yanofsky,M.F. (1992). Molecular characterisation of the Arabidopsis floralhomeotic gene APETALA1. Nature 377, 522–524.

Marx, G.A. (1986a). Linkage relationship of Curl, Orc and ‘Det’ withmarkers on chromosome 7. Pisum Newslett. 18, 45–48.

Marx, G.A. (1986b). Tendrilled acacia (tac) an allele at the uni locus.Pisum Newslett. 18, 49–52.

Marx, G.A. (1987). A suite of mutants that modify pattern formationin pea leaves. Plant Mol. Biol. Rep. 5, 311–355.

Meyerowitz, E.M. (1997). Genetic control of cell division patterns indeveloping plants. Cell 88, 299–308.

Michael, A.J., Hofer, J.M.I., and Ellis, T.H.N. (1996). Isolation byPCR of a cDNA clone from pea petals with similarity to petuniaand wheat zinc finger proteins. Plant Mol. Biol. 30, 1051–1058.

Molinero-Rosales, N., Jamilena, M., Zurita, S., Gómez, P., Capel,J., and Lozano, R. (1999). FALSIFLORA, the tomato orthologueof FLORICAULA and LEAFY, controls flowering time and floralmeristem identity. Plant J. 20, 685–693.

Monti, L.M. (1970). Linkage studies on four induced mutants of pea.Pisum Newslett. 2, 21–22.

Monti, L.M., and Devreux, M. (1969). Stamina pistilloida: A newmutation induced in pea. Theor. Appl. Genet. 39, 17–20.

Murfet, I.C. (1990). Internode length and anatomical changes inPisum genotypes crys and cryc in response to extended daylengthand applied gibberellin A1. Physiol. Plant. 79, 497–505.

Parcy, F., Nilsson, O., Busch, M., and Weigel, D. (1998). A geneticframework for floral patterning. Nature 395, 561–566.

Patton, E.E., Willems, A.R., and Tyers, M. (1998). Combinatorialcontrol in ubiquitin dependent proteolysis: Don’t Skp the F-boxhypothesis. Trends Genet. 14, 236–243.

Pouteau, S., Nicolls, D., Tooke, F., Coen, E., and Battey, N.(1997). The induction and maintenance of flowering in Impatiens.Development 127, 3343–3351.

Reid, J.B., Murfet, I.C., Singer, S.R., Weller, J.L., and Taylor, S.A(1996). Physiological-genetics of flowering in Pisum. Semin. CellDev. Biol. 7, 455–463.

Samach, A., Klenz, J.E., Kohalmi, S.E., Risseeuw, E., Haughn,

G.W., and Crosby, W.L. (1999). The UNUSUAL FLORALORGANS gene of Arabidopsis thaliana is an F-box proteinrequired for normal patterning and growth in the floral meristem.Plant J. 20, 433–445.

Schultz, E.A., and Haughn, G.W. (1993). Genetic analysis of the floralinitiation process (FLIP) in Arabidopsis. Development 119, 745–765.

Schwarz-Sommer, Z., Huijser, P., Nacken, W., Saedler, H., andSommer, H. (1990). Genetic control of flower development byhomeotic genes in Antirrhinum majus. Science 250, 931–936.

Simon, R., Carpenter, R., Doyle, S., and Coen, E. (1994). Fimbriatacontrols flower development by mediating between meristem andorgan identity genes. Cell 78, 99–107.

Singer, S.R., Hsiung, L.P., and Huber, S.C. (1990). Determinate(det) mutant of Pisum sativum (Leguminosae: Papilionoideae)exhibits an indeterminate growth pattern. Am. J. Bot. 77, 1330–1335.

Souer, E., van der Krol, A., Kloos, D., Spelt, C., Bliek, M., Mol, J.,and Koes, R. (1998). Genetic control of branching pattern and flo-ral identity during Petunia inflorescence development. Develop-ment 125, 733–742.

Stephens, W.L. (1939). Tables of the recombination fraction esti-mated from the product ratio. J. Genet. 39, 171–180.

Tucker, S.C. (1989). Overlapping organ initiation and common pri-mordia in flowers of Pisum sativum (Leguminosae: Papilion-oideae). Am. J. Bot. 76, 714–729.

Weeden, N.F., Swiecicki, W.K., Timmerman-Vaughan, G.M.,Ellis, T.H.N., and Ambrose, M. (1996). The current pea linkagemap. Pisum Genet. 28, 1–4.

Weigel, D., and Meyerowitz, E.M. (1994). The ABCs of floralhomeotic genes. Cell 78, 203–209.

Weigel, D., Alvarez, J., Smyth, D.R., Yanofsky, M.F., andMeyerowitz, E.M. (1992). LEAFY controls floral meristem identityin Arabidopsis. Cell 69, 843–859.

Weller, J.L., Murfet, I.C., and Reid, J.B. (1997). Pea mutants withreduced sensitivity to far-red light define an important role forphytochrome A in daylength detection. Plant Physiol. 114, 1225–1236.

Wilkinson, M.D., and Haughn, G.W. (1995). UNUSUAL FLORALORGANS controls meristem identity and organ primordia fate inArabidopsis. Plant Cell 7, 1485–1499.

Wiltshire, R.J.E., Murfet, I.C., and Reid, J.B. (1994). The geneticcontrol of heterochrony: Evidence from developmental mutants ofPisum sativum L. J. Evol. Biol. 7, 447–465.

Yanofsky, M.F. (1995). Floral meristems to floral organs: Genescontrolling early events in Arabidopsis flower development. Annu.Rev. Plant Physiol. Plant Mol. Biol. 46, 167–188.

DOI 10.1105/tpc.13.1.31 2001;13;31-46Plant Cell

Scott Taylor, Julie Hofer and Ian MurfetFlowers, Inflorescences, and Leaves

, Is Required for Normal Development ofUFO and Fim, the Pea Ortholog of Stamina pistilloida

 This information is current as of June 16, 2020

 

Permissions https://www.copyright.com/ccc/openurl.do?sid=pd_hw1532298X&issn=1532298X&WT.mc_id=pd_hw1532298X

eTOCs http://www.plantcell.org/cgi/alerts/ctmain

Sign up for eTOCs at:

CiteTrack Alerts http://www.plantcell.org/cgi/alerts/ctmain

Sign up for CiteTrack Alerts at:

Subscription Information http://www.aspb.org/publications/subscriptions.cfm

is available at:Plant Physiology and The Plant CellSubscription Information for

ADVANCING THE SCIENCE OF PLANT BIOLOGY © American Society of Plant Biologists