| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Development and Morphogenesis |
Department of Biological Sciences, The University of Iowa, Iowa City, Iowa 52246 USA
Received for publication August 29, 2002. Accepted for publication November 12, 2002.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Key Words: caryopsis evolution • flower polarity • grass florets • ovule number • spikelet development • zygomorphy
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
). A kernel, or caryopsis, is a fruit containing a single, large seed. In maize, every kernel in every row develops with the embryo on the side closest to the tip of the ear, i.e., on the acropetal side of the kernel. This regularity of kernel position and orientation is so striking that it was noted in one of the earliest English descriptions of maize: "...The grayne or seede which groweth in the eares, is about the quantitie or bignesse of a pease ..., growing orderly around the eares, in nine or tenne ranges or rowes" (Dodoens, 1578
).
This "orderly" array of fruits on the ear is preceded by a number of branching events during the development of the maize inflorescences. Unlike Arabidopsis or snapdragon, in which floral meristems arise directly from the inflorescence meristem, the floral meristems in maize are formed on second-order branches of the inflorescences where initiation events are characterized by a succession of 90° reorientations (Bonnett, 1948
; Irish, 1997
; McSteen et al., 2000
). In normal ear (or tassel) development in maize, the inflorescence meristem initiates acropetally a series of spikelet pair meristems, which are arranged in vertical files around the developing inflorescence (Fig. 1). Each spikelet pair meristem initiates two spikelet meristems, first the sessile spikelet meristem, then the pedicellate, and is consumed in the process of initiating the latter (Bonnett, 1948
; Irish, 1997
; McSteen and Hake, 2001
). The two spikelet meristems arise on a plane perpendicular to that of the inflorescence axis. Each spikelet meristem initiates a pair of glumes (bracts), then two floral meristems, and, as with the spikelet pair meristem, is consumed in the process (Irish, 1997
; Chuck et al., 1998
). This stereotypic pattern of meristem initiation results in quadrangular arrays of four floral meristems, in which the two florets of each spikelet are aligned parallel to the main axis of the inflorescence. The two florets of a spikelet have opposite orientations, facing each other, but the lower floret of each spikelet on ears aborts (Bonnett, 1940
; Cheng et al., 1983
). Consequently, all functional florets (the upper florets) are oriented the same way. The fixed orientations of the ovule in each floret and of the embryo sac in each ovule limit embryos to the acropetal side of all kernels.
|
; Jones, 1934
; Irish and Nelson, 1993
). Both florets of each spikelet also develop in the sweet corn variety Country Gentleman (Bonnett, 1940
), although no specific mutation has been identified that is responsible for that phenotype in this line. In reversed germ orientation (rgo) mutants, the lower rather than the upper floret of each spikelet develops, resulting in all basipetal embryos (Sachan and Sarkar, 1978
). Rgo2 has been found to be an allele of indeterminate spikelet (N. Kaplinski, University of California, Berkeley, California, USA, personal communication), the maize homologue of AP2 (Chuck et al., 1998
). In ts4 and ts6 mutants, ears can have kernels with other than acropetal embryos, but this phenotype is the result of proliferations of spikelet or floret meristems and subsequent crowding of developing kernels (Irish, 1997
), resulting in disorganized spikelets in which specification of certain florets to be suppressed may not be able to occur. This paper describes a new maize mutant, wandering carpel (wcr), in which embryos can be found on almost any face of the kernel. Whereas normal maize florets are zygomorphic, the aberrant orientation of wcr kernels results from a reversion from zygomorphy to actinomorphy (radial symmetry) in some cases and in other cases, from altered polarity of the spikelet. Other defects seen in wcr mutants include failure to suppress the lower floret, development of extra carpels, and, more rarely, development of two seeds in a single kernel. The wcr mutation is inherited as a maternal-sporophyte-effect, semidominant mutation that is located on chromosome 2L. We propose that WCR+ is required for normal maize spikelet polarity and that defects in establishing polarity lead to altered spikelet orientation, failure to suppress lower florets, and radial rather than bilateral symmetry. Because wcr mutants can form kernels with more than the normal single seed, this mutation may also provide clues about the evolution of the caryopsis.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
To score an ear for wcr expression quantitatively, the orientation of each kernel was assessed by the position of the embryo relative to the tip of the ear. Because in most cases, kernel crowding obscured the embryo from sight, each kernel was removed from the ear, one at a time, and its orientation scored.
Scanning electron microscopy
Developing ears were examined by Scanning electron microscopy (SEM) as described previously (Irish, 1997
). The ears were harvested from plants derived from kernels on ears with a strong wcr phenotype that had been self-pollinated and thus were expected to have had a strong wcr kernel phenotype if they had been allowed to develop and be pollinated.
Histology
Developing florets were examined histologically using standard protocols. Ears were fixed in formalin-alcohol-acetic acid (FAA), embedded in wax, and sectioned. Slides were stained with 0.05% toluidine blue.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
), and grasses lack genuine petals, most grass flowers are zygomorphic (Cocucci and Anton, 1988
). In maize, zygomorphy is manifested in all floral whorls. As is typical of monocots, maize flowers are fundamentally trimerous (Cheng et al., 1983
). The single pistil or gynoecium is formed from three congenitally fused carpels (Bonnett, 1953
). These carpels differ from each other (Fig. 2). The two abaxial carpels fuse to form the silk, which grows indeterminately to form the longest stigma of any species (Clifford, 1988
). The third carpel is much more limited in its growth, elongating only enough to cover the developing ovule and, later, the embryo plus a portion of the endosperm. This determinate carpel is the one that forms the single ovule (Randolph, 1926
), whereas the two indeterminate carpels are sterile. Thus, the carpel whorl shows zygomorphy, both in determinacy and fertility.
|
). The small organs external to and alternate with the stamens, the lodicules, have long been considered by many to be highly reduced petals, based on their position (Stebbins, 1982
). Recent molecular evidence from studies of the silky mutation of maize strongly supports this interpretation (Ambrose et al., 2000
). The lodicule whorl also shows zygomorphy: only two and not three lodicules are present in maize (Fig. 2a) and in most other grasses (Clifford and Watson, 1977
). Thus, all three whorls of the maize floret show zygomorphy.
The midpoint of the extra-wide gap between the lateral stamens is on the same radius as the determinate carpel and the "missing" lodicule, presenting an "unadorned" adaxial (relative to the spikelet rachilla) side of the floret (Fig. 2a, c). The unadorned appearance of the adaxial side of the floret provides an easily observed feature for scoring the orientation of maize florets on developing inflorescences. On normal maize ears, all functional florets (the upper floret of each spikelet) are oriented so that their unadorned side is basipetal, relative to the inflorescence. The campylotropous ovule develops within the adaxial carpel (Randolph, 1926
), bending away from it (Bonnett, 1953
). The embryo sac is oriented within the ovule such that the egg is closer to the micropyle than is the central cell; i.e., the egg is abaxial. After fertilization, the single embryo develops from that location and thus is found on the acropetal side of each kernel, perpetuating the zygomorphy of the flower.
wcr mutant identification
Wandering carpel arose spontaneously in a stock of the inbred line W23 that had been converged to full color aleurone. The first-seen phenotype of wcr mutants was aberrant embryo location on many kernels of an ear (Fig. 3). Rather than all embryos developing on the acropetal side of the kernel (Fig. 3b), some embryos were lateral or even on the basipetal side (Fig. 3a, d). The wcr phenotype was very weak initially with only a few kernels per ear with aberrant orientation and was not noted when it first arose. Further self-pollinations in subsequent generations yielded some ears with stronger expression, in which a notable number of kernels had embryos on the "wrong" side. Once the strong phenotype had been observed, careful examination of progenitor ears revealed the presence of the weaker phenotype. Self-pollinations of plants from strong phenotype ears has not resulted in further enhancement of the mutant phenotype, so that, at most, half of the kernels have an aberrant orientation, and the remaining kernels have a normal orientation (Fig. 3a, d).
|
Inheritance
At least half of the kernels on affected ears had a normal orientation, even in ears with a strong wcr phenotype. This gave the appearance of segregation of zygotic genotypes on an ear, which would indicate that the embryo and/or the endosperm regulates the wcr phenotype. Alternatively, the genotype of the ear might be controlling the wcr phenotype, but with partial expressivity. To distinguish between these two possibilities, reciprocal crosses were performed between plants from a strong wcr ear and normal inbreds. If wcr is a maternal-sporophyte-effect mutation, then when mutant plants are used as females in crosses with normal plants, the resulting ears should show the wcr phenotype. Conversely, when wcr mutants are used as males in reciprocal crosses, the resulting ears should be normal. Table 1 shows that this was the case. When wcr was used as a female parent, approximately 40% of kernels on resulting ears had aberrant orientation (a.o.k.), in comparison to only about 3% a.o.k. when wcr was used as the male parent.3 These results show that the wcr phenotype is controlled by the genotype of the ear and its florets but has partial expressivity.
|
|
The F2 populations showed a greater range of wcr phenotypes than did the F1 populations, consistent with wcr being a semidominant mutation. As in the F1, F2 ears from the W23 cross showed stronger expression of wcr than did those from the W22 cross (Fig. 4c, b). The strongest expression of wcr in the W23 F2 population was 57.7% a.o.k. (Fig. 4c), whereas the strongest expression in the W22 F2 population was only about one-sixth of that, at 9.8% a.o.k. (Fig. 4b). Both populations included ears with at most one aberrantly oriented kernel. The median percentage of a.o.k. in the W23 population was about 8.4%, whereas in the W22 population it was about 3.5%. Thus, modifier genes that affect the expression of wcr must exist and differ between W22 and W23.
The most striking aspect of the segregation of the wcr phenotype in the F2 populations was the lack of distinct classes of ears having a certain range of a.o.k. values. If wcr is a semidominant mutation that acts alone to affect the orientation of kernels, then the progeny from a self-pollinated heterozygote would be expected to consist of discrete classes in which one-quarter of the ears had a strong wcr phenotype (say, all approximately 30% a.o.k.), one-half of the ears had a weak wcr phenotype (say, all 8% a.o.k.), and one-quarter had normal ears (all 1–2% a.o.k., which would include the few obliquely oriented kernels typically found on most ears). What was observed instead was an even distribution of percent a.o.k. values, such that when graphed in order of increasing values, the first three-quarters of ears in the F2 population (presumably the homozygous wild-type ears and the heterozygotes) gave a linear distribution of points (Fig. 4c). The last quartile of ears (the presumed homozygotes) showed the greatest range of values as well as a break in the line of points from the first three-quarters. A similarly even distribution of a.o.k. values among W22 F2 ears was seen (Fig. 4b). These observations support the conclusions that wcr is a semidominant mutation and that additional modifier genes affect its expression. These modifiers vary not only between W22 and W23, but also have segregated in the process of generating the F2 populations.
wcr is located on the long arm of chromosome 2
The TB translocation stocks were used to map wcr to a chromosome arm (Beckett, 1993
). As wcr is semidominant, it was possible that wcr/– hemizygotes would have a wcr phenotype. The wcr/wcr plants were crossed by each TB stock for a particular chromosome arm. Ten seeds from each successful cross were planted and the resulting ears were crossed by W23 pollen. After seed set, these F1 ears were examined for the wcr phenotype. Crosses with the TB stock for the chromosome arm on which wcr is located were expected to give a mixture of heterozygous and hemizygous ears, as TB plants produce both euploid and deficient pollen (Carlson, 1988
). In contrast, crosses with any of the other TB stocks would be expected to give only heterozygous +/wcr, and thus weak phenotype, ears. Crosses with TB1Sb–2L, which uncovers both chromosome arms 1S and 2L, yielded three F1 ears, one of which had a very strong wcr phenotype, with 36.4% aberrantly oriented kernels. This hemizygous ear had a phenotype as strong as that of strongly expressing homozygous ears, suggesting that wcr is a loss of function mutation. Crosses with TB1Sb, which uncovers only the short arm of chromosome 1, yielded only weak wcr F1 ears. Thus, wcr is likely located on chromosome 2L. More precise mapping studies are underway.
Additional data quantifying the range of expression of wcr in heterozygous ears was provided by crosses with TB stocks that did not uncover wcr. A range of values of a.o.k. percentage values was found, as had been found in crosses with W23 and W22. The values ranged from 0.7 to 5.4% a.o.k., with a median of 2.5% a.o.k. (Fig. 4d).
Developmental basis of the wcr phenotype
The wcr mutant was discovered from its mature kernel phenotype, in which the position of the embryo was not restricted to the acropetal face of the kernel. There were no detectable abnormalities in the mature cob of wcr/wcr ears, and tassels of mutant plants also appeared normal (Fig. 3c). Thus, the wcr phenotype is likely the result of abnormal floral development. Possible explanations for the wcr phenotype include rotation of the floral or spikelet meristems or aberrant positioning of the ovule or embryo sac within an otherwise normal flower. Aberrant pistil development might result in supernumerary ovule formation. The completely reversed orientation could be the result of failure to suppress the lower floret in some spikelets. To determine which of these possible developmental changes underlies the wcr phenotype, development of wcr/wcr ears was examined by SEM and by histology.
Thirty-six wcr/wcr ears, representing a range of developmental stages, from 3 to 42 mm, were examined by SEM. (As the length of an inflorescence axis is a reliable indicator for developmental stage of its most developmentally advanced flowers ([Irish and Nelson, 1991
, 1993
]), which are found at the base of the inflorescence, length will be used to refer to an ear of a particular developmental stage; however, as long as the inflorescence meristem is still active, the earliest stages of reproductive meristems are present at the tips of the ears throughout a range of ear lengths.) The wcr mutants did not deviate from the normal pattern of acropetal initiation of numerous spikelet pair meristems, followed by their initiation of two spikelet meristems, each of which in turn initiates two floral meristems (Fig. 5a), consistent with the ear phenotype in mature plants.
|
|
Examination of the patterning of aberrant spikelets among the normal spikelets revealed that their distribution was not random. Chi-square analysis of aberrant spikelet distribution showed that aberrant spikelets were very likely to be paired with a normal spikelet (P < 10–8). Because pairs of spikelets arise from a common primordium, the spikelet pair meristem, this association of normal and abnormal spikelets suggests that the defect that results in aberrant polarity occurs very early in ear morphogenesis, perhaps at the time of spikelet meristem initiation. On ears, pedicellate spikelets can be distinguished from sessile spikelets only briefly, and well before any defects associated with wcr become visible. Thus, it was impossible to determine whether one or the other type of spikelet was preferentially affected by this mutation.
Histological examination by serial longitudinal sectioning of 10 wcr ears, ranging from 6 mm long to 24 mm long (not shown) confirmed the conclusions from SEM analyses: the aberrant embryo location in mature ears is the consequence of altered orientation of spikelets, altered symmetry of some flowers, and, in the case of some kernels with basipetal embryos, the development of the lower floret of some spikelets. In addition, several pistils with two ovules were found, establishing the basis for the two-seeded kernel phenotype (Fig. 6a). The multiple-carpel silk phenotype persisted as the ears developed to anthesis. Many pistils were found with multiple silks, some of which were formed from more than two carpels, whereas others formed from single carpels (Fig. 6b).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
). Among the first morphogenetic events in plant embryogenesis is the organization of the root and shoot apical meristems at opposite ends of the embryo. This root/shoot polarity is maintained throughout the life of the plant. Polarity is also manifested in lateral organs, such as leaves, which typically have distinct patterns of differentiation on upper (adaxial) vs. lower (abaxial) sides. Flowers, which are short shoots, show polarity along their basal/terminal axis: the differentiation of the various organ types in nonmutant flowers follows the stereotypical pattern of sepals at the base, followed by petals, stamens, and the pistil(s). Whereas many plants develop flowers with radial symmetry (actinomorphy), in which all organs within a floral whorl develop identically and thus the flowers lack a second axis of polarity, other species form zygomorphic flowers. In zygomorphy, each organ within a whorl is distinct, yet the whorl as a whole shows reflective symmetry. Polarity in this dimension of the flower is expressed in the differentiation of upper, lower, and lateral organs (Endress, 1999
). Such transectional symmetry is typically most obvious in petals, as seen in snapdragons and orchids. Zygomorphy is considered to be the derived condition (Tucker, 1999
).
How different members of a floral whorl differentiate in zygomorphic flowers has been elucidated by the genetic and molecular analysis of peloric mutants (Almeida et al., 1997
; Theissen, 2000
), such as cycloidea in Antirrhinum (Luo et al., 1996
), in which there is a reversion to radial symmetry. CYC+ encodes a member of the TCP family of transcription factors (Cubas et al., 1999b
) and shows homology to teosinte branched of maize (Doebley et al., 1997
), a gene that when mutated fails to suppress the growth of axillary meristems. CYC+ expression is localized to the upper portion of the floral meristem (Luo et al., 1996
), where it down-regulates the expression of D-cyclin in that region, thereby locally inhibiting growth in the upper portion of the flower (Gaudin et al., 2000
). Cycloidea mutants develop flowers in which entire whorls of organs differentiate like the lower organs of that whorl in normal flowers. This phenotype is consistent with the conclusion that localized CYC+ expression serves to differentiate the upper portion from the rest of the floral meristem. Variation among CYC+ gene homologues has been associated with variation in symmetry among related species in some (Cubas et al., 1999a
), but not all (Citerne et al., 2000
) taxa.
Analyses of additional peloric mutations of snapdragon, such as Dichotoma (Luo et al., 1999
), Radialis (Luo et al., 1996
), and Divaricata (Almeida et al., 1997
), support the conclusion that zygomorphic development in flowers is the result of the inhibition of growth in the adaxial portion of the floral meristem and later in developing floral organs. Interestingly, a similar hypothesis regarding polar development of the leaf was put forth by Wardlaw (1949)
, who suggested that the shoot meristem inhibits growth on the adaxial side of the initially radial leaf primordium, thereby setting up a physical system in which growth occurs in a new (medial-lateral) plane.
wcr mutants
Peloric mutations affect how various members of a whorl become different from each other in generating zygomorphy, but have no effect on the actual orientation of the flower. Wandering carpel is, so far, unique in that it alters both floral symmetry and orientation. The aberrant kernel orientation phenotype arises, at least in part, from aberrant orientation of spikelets, in which the two florets maintain their normal orientation relative to each other, but the spikelet itself is not as it should be, parallel to the long axis of the ear. Surprisingly, the glumes do not appear to be included in the aberrant orientation of the spikelet; just the florets are affected. Another developmental defect that results in aberrant kernel orientation is the occasional failure to suppress the lower floret of a spikelet. The fertilization of these florets results in kernels with a completely reversed orientation. A third probable cause of aberrant kernel orientation is loss of zygomorphy: if the distinction between the indeterminate, sterile carpels and the determinate, fertile carpel is lost, it is likely that some ovules could arise in a carpel other than the adaxial carpel. Whereas we have not proved that some aberrantly oriented kernels result from ovules forming in the "wrong" carpel, this type of event is likely the cause of formation of two ovules in a single ovary, which can result in the formation of a kernel with two embryos. Whether a carpel can both form an ovule and contribute to a silk is not known.
An alternative hypothesis to explain multiple indeterminate carpels and multiple ovules is that fasciation occurs in the floral meristem of wcr mutants. This fasciation would have to occur just before carpel initiation because stamen number remains unchanged at three in wcr tassels (unpublished data). Evidence for fasciation would be increased size of the floral meristem (Szymkowiak and Sussex, 1992
). Whereas careful measurements have not yet been done, preliminary examinations provide no evidence of an increase in floral meristem size in the mutant. Also, fasciation is not sufficient to explain the altered orientation of spikelets in the mutants that have failed to suppress lower florets. For these reasons, we favor the hypothesis that the wcr phenotype is the result of loss of polarity.
It is not immediately apparent why an affected spikelet might show either altered orientation or loss of zygomorphy, or in some cases both. We suspect that these two phenotypes are the consequence of a single developmental defect, in much the same way the leaf polarity mutants can show both loss of the blade as well as altered patterns of differentiation of leaf tissues (Waites and Hudson, 2001
). In the case of wcr, the lack of some factor that establishes polarity in a spikelet may result in a normal spikelet forming with random orientation or in a normal spikelet with radially symmetric florets.
It is interesting to note the similarities between maize and snapdragon flowers with regard to zygomorphy. The missing lodicule, the wide gap between lateral stamens, and the fertile carpel are all on the same, adaxial radius. This correlation suggests that growth is inhibited in the portion of the floral meristem closest to the spikelet rachilla, i.e., the adaxial side, in a manner similar to what is seen in snapdragon, in which growth is down-regulated by the action of Cycloidea in the upper, adaxial side of the developing flower (Luo et al., 1996
; Gaudin et al., 2000
). The similarities in growth inhibition and its location between maize and snapdragon suggest that the processes underlying zygomorphic development in these distantly related species share some common elements.
Function of WCR+
We hypothesize that WCR+ is needed for normal production of a factor needed for polar development or its distribution to both spikelet meristems. Normal polarity encompasses both zygomorphic floral development and correct orientation of the spikelet, so that the unadorned side of the upper florets will be basipetal on the ear, as well as suppression of the lower floret in ears. This gene functions in a quantitative manner so that one normal allele is nearly, but not quite, sufficient for normal development. In absence of both normal alleles, spikelet meristems lack the cues for normal polarity. Loss of zygomorphy in the carpel whorl results in supernumerary silks and/or ovules.
One way to envision how WCR+ bestows both correct orientation and bilateral symmetry in developing spikelets is to consider a growing primordium with the shape of a hemisphere. If growth is equal throughout the primordium, the hemisphere will simply become larger, while maintaining its radial symmetry. If, alternatively, the hemisphere is somehow tethered at two opposite points, as it grows it will convert from radial to bilateral symmetry and the axis of that symmetry will be defined by the two tether points. In this scenario, wcr mutants could be thought of as having lost function of one or both of the tether points. Variation in genetic redundancy directing the formation or stability of the tethers could account for differences in wcr penetrance between W23 and W22.
An unusual feature of wcr ears is the lack of full expressivity. Even in its most extreme expression (homozygous in W23), roughly half of the kernels on mutant ears have normal, acropetal-embryo orientation. Developmental studies revealed that this apparent lack of full expressivity is the result of only one of the two spikelets in most pairs having aberrant orientation, with the other spikelet being normal with respect to symmetry and orientation. Whereas some spikelet pairs are both normal or both aberrant, pairs of a normal with an aberrantly oriented spikelet were far more common than expected from a random distribution. As the two spikelets of a pair originate from a common spikelet pair meristem, this normal/aberrant pairing suggests that the defect in wcr mutants occurs during the formation of the two spikelet meristems from the spikelet pair meristem.
Although at later stages the two spikelets that arise from a single spikelet pair meristem are indistinguishable on ears, at the time of initiation there are visible as well as historical differences between the two spikelet pair meristems. The sessile spikelet meristem arises laterally on the flank of the spikelet pair meristem and appears first (Bonnett, 1948
; Irish, 1997
). The pedicellate meristem, while also arising laterally (Chuck et al., 1998
; Irish, 1998
), is composed primarily of cells that most recently had the identity of spikelet pair meristem. Thus, if wcr mutants fail to distribute some putative polarizing factor in sufficient quantities to both spikelet meristems, it seems plausible that the first-formed meristem, the sessile spikelet meristem, receives enough of this factor, leaving the later-formed pedicellate spikelet meristem without sufficient polarizing factor. In an equally plausible alternative, the polarizing factor may be limited to the cells that compose the spikelet pair meristem. In this case, the sessile spikelet meristem would not receive this factor, but the pedicellate spikelet would receive it because it comes more directly from the spikelet pair meristem.
Evolution of the caryopsis
Although very rare, the two-seeded kernel phenotype suggests that WCR+ has played an important role in development of the caryopsis. If normal developmental constraints limit the grass fruit to a single ovule, and thus a single embryo, by asymmetric development of the carpel whorl, then loss-of-function mutations that prevent zygomorphy may result in the differentiation of additional fertile carpels. The low level of phenotypic expression may indicate the existence of redundantly acting genes that almost completely mask this phenotype. The development of two seeds in wcr kernels is the result of the development and fertilization of two ovules in a single syncarpellate pistil. Given the low number of kernels found with two embryos, it is perhaps surprising that several pistils with two ovules were found during histological studies. The frequency of kernels with two seeds may underrepresent the strength of the two-ovule phenotype, as the result of failure to fertilize both ovules. Ears with a high number of aberrantly oriented kernels tend to have poor seed set (not shown). As the mutants show misoriented flowers in SEMs, the silks may be growing the wrong direction within the husk leaves, and thus are never exposed to pollen.
The two-embryo phenotype of wcr mutants may also indicate some role of WCR during the evolution of the caryopsis, or grain. All grasses have this fruit type, or a type derived from the caryopsis (Sendulsky et al., 1988
). In the caryopsis, a dry, monospermic, indehiscent fruit, a single ovule containing a single embryo sac is enclosed by a tricarpellary pistil. How the caryopsis evolved, most likely from a capsule (Dahlgren et al., 1985
, p. 71), is not well understood. This process is thought to involve a transition from a three-loculate-three-ovule gynoecium to a single-loculate-single-ovulate condition (Philipson, 1985
). A reduction in ovule number is also considered to be a feature of the wind pollination syndrome (Dahlgren et al., 1985
, p. 84). The wcr mutation may provide a key to unraveling the evolution of this special fruit type.
|
FOOTNOTES |
|---|
3 It is common to observe a small number of kernels with oblique orientation in most inbreds. These kernels are typically found at the tip or base of the ear. Because it was impossible to distinguish between these and the kernels with aberrant orientation resulting from the wcr mutation, all aberrantly oriented kernels were counted when scoring the wcr phenotype. Similar counts of aberrantly oriented kernels on six normal W23 ears, averaging 600 kernels each, gave an average of 1.3% a.o.k.
|
LITERATURE CITED |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Ambrose B. D. Lerner P. Ciceri C. Padilla M. F. Yanofsky R. J. Schmidt 2000 Molecular and genetic analyses of the silky1 gene reveal conservation in floral organ specification between eudicots and monocots. Molecular Cell 5: 569-579[CrossRef][ISI][Medline]
Beckett J. 1993 Locating recessive genes to chromosome arm with B-A translocations. In M. Freeling and V. Walbot [eds.], The maize handbook, 315–317. Springer-Verlag, New York, New York, USA
Bonnett O. T. 1940 Development of the staminate and pistillate inflorescences of sweet corn. Journal of Agricultural Research 60: 25-37
Bonnett O. T. 1948 Ear and tassel development in maize. Annals of the Missouri Botanical Garden 35: 269-287[CrossRef]
Bonnett O. T. 1953 Developmental morphology of the vegetative and floral shoots of maize. Illinois Agricultural Experiment Station Bulletin 568
Carlson W. 1988 The cytogenetics of maize. In G. F. Sprague and J. W. Dudley [eds.], Corn and corn improvement, 259–344. American Society of Agronomy, Crop Science Society of America, and Soil Science Society of America, Madison, Wisconsin, USA
Cheng P. C. R. I. Greyson D. B. Walden 1983 Organ initiation and the development of unisexual flowers in the tassel and ear of Zea mays. American Journal of Botany 70: 450-462[CrossRef][ISI]
Chuck G. R. Meeley S. Hake 1998 The control of maize spikelet meristem fate by the APETALA2-like gene indeterminate spikelet1. Genes and Development 12: 1145-1154
Citerne H. L. M. Moller Q. B. C. Cronk 2000 Diversity of cycloidea-like genes in Gesneriaceae in relation to floral symmetry. Annals of Botany 86: 167-176
Clifford H. T. 1988 Spikelet and floral morphology. In T. Soderstrom, K. Hilu, C. Campbell, and M. Barkworth [eds.], Grass systematics and evolution, 28. Smithsonian Institution Press, Washington, D.C., USA
Clifford H. T. L. Watson 1977 Identifying grasses: data, methods and illustrations. University of Queensland Press, St. Lucia, Queensland, Australia
Cocucci A. E. A. M. Anton 1988 The grass flower: suggestions on its origin and evolution. Flora 181: 353-362[ISI]
Cubas P. N. Lauter J. Doebley E. Coen 1999a The TCP domain: a motif found in proteins regulating plant growth and development. Plant Journal 18: 215-222[CrossRef][ISI][Medline]
Cubas P. C. Vincent E. Coen 1999b An epigenetic mutation responsible for natural variation in floral symmetry. Nature 401: 157-161[CrossRef][Medline]
Dahlgren R. H. Clifford P. Yeo 1985 The families of the monocotyledons. Springer-Verlag, New York, New York, USA
Dodoens R. (Lyte, H.) 1578 A Niewe Herball. ... nowe first translated out of French into English, by Henry Lyte Esquyer. The fourth Booke of the Historie of Plantes, 4623–4624. G. Dewes, London
Doebley J. A. Stec L. Hubbard 1997 The evolution of apical dominance in maize. Nature 386: 485-488[CrossRef][Medline]
Emerson R. A. 1920 Heritable characters in maize. II. Pistillate flowered maize plants. Journal of Heredity 11: 65-76
Endress P. 1999 Symmetry in flowers: diversity and evolution. International Journal of Plant Sciences 160: S3-S23[CrossRef][ISI][Medline]
Gaudin V. P. A. Lunness P. R. Fobert M. Towers C. Riou-Khamlichi J. A. H. Murray E. Coen J. H. Doonan 2000 The expression of D-cyclin genes defines distinct developmental zones in snapdragon apical meristems and is locally regulated by the cycloidea gene. Plant Physiology 122: 1137-1148
Hudson A. 2000 Development of symmetry in plants. Annual Reviews of Plant Physiology and Plant Molecular Biology 51: 349-370[CrossRef][ISI]
Irish E. E. 1997 Class II tassel seed mutations provide evidence for multiple types of inflorescence meristems in maize. American Journal of Botany 84: 1502-1515[Abstract]
Irish E. E. 1998 Grass spikelets: a thorny problem. BioEssays 20: 789-793[CrossRef][ISI]
Irish E. E. T. Nelson 1991 Identification of multiple stages in the conversion of maize meristems from vegetative to floral development. Development 112: 891-898[Abstract]
Irish E. E. T. Nelson 1993 Development of tassel seed 2 inflorescences in maize. American Journal of Botany 80: 292-299[CrossRef][ISI]
Jones D. F. 1934 Unisexual maize plants and their bearing on sex differentiation in other plants and in animals. Genetics 19: 552-567
Luo D. R. Carpenter L. Copsey C. Vincent J. Clark E. Coen 1999 Control of organ asymmetry in flowers of Antirrhinum. Cell 99: 367-376[CrossRef][ISI][Medline]
Luo D. R. Carpenter C. Vincent L. Copsey E. Coen 1996 Origin of floral asymmetry in Antirrhinum. Nature 383: 794-799[CrossRef][Medline]
McSteen P. S. Hake 2001 barren inflorescence2 regulates axillary meristem development in the maize inflorescence. Development 128: 2881-2891
McSteen P. D. Laudencia-Chingcuanon J. Colasanti 2000 A floret by any other name: control of meristem identity in maize. Trends in Plant Science 5: 61-67[CrossRef][ISI][Medline]
Philipson W. R. 1985 In the grass gynoecium monocarpellary?. American Journal of Botany 72: 1954-1961[CrossRef][ISI]
Randolph F. R. 1926 A cytological study of two types of variegated pericarp in maize. Cornell Agricultural Experiment Station Memoirs 102
Sachan J. K. S. K. R. Sarkar 1978 Reversed germ orientation. Maize News Letter 52: 119-120
Sendulsky T. T. S. Filgueiras A. G. Burman 1988 Fruits, embryos, and seedlings. In T. Soderstrom, K. Hilu, C. Campbell, and M. Barkworth [eds.], Grass systematics and evolution, 31–35. Smithsonian Institution Press, Washington, D.C., USA
Stebbins G. L. 1982 Major trends of evolution in the Poaceae and their possible significance. In J. Estes, R. Tyrl, and J. Brunken [eds.], Grasses and grasslands: systematics and ecology, 3–36. University of Oklahoma Press, Norman, Oklahoma, USA
Szymkowiak E. J. I. M. Sussex 1992 The internal layer (L3) determines floral meristem size and carpel number in tomato periclinal chimeras. Plant Cell 4: 1089-1100
Theissen G. 2000 Evolutionary developmental genetics of floral symmetry: the revealing power of Linnaeus' monstrous flower. BioEssays 22: 209-213[CrossRef][ISI][Medline]
Tucker S. 1999 Evolutionary lability of symmetry in early floral development. International Journal of Plant Sciences 160: S25-S39[CrossRef][ISI][Medline]
Waites R. A. Hudson 2001 The Handlebars gene is required with Phantastica for dorsoventral asymmetry of organs and for stem cell activity in Antirrhinum. Development 128: 1923-1931
Wardlaw C. 1949 Experiments on organogenesis in ferns. Growth 9: 93-131
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
