Logo of plntcellLink to Publisher's site
Plant Cell. 2005 Jun; 17(6): 1674–1684.
PMCID: PMC1143069

Mechanisms of Derived Unitegmy among Impatiens Species


Morphological transitions associated with ovule diversification provide unique opportunities for studies of developmental evolution. Here, we investigate the underlying mechanisms of one such transition, reduction in integument number, which has occurred several times among diverse angiosperms. In particular, reduction in integument number occurred early in the history of the asterids, a large clade comprising approximately one-third of all flowering plants. Unlike the vast majority of other eudicots, nearly all asterids have a single integument, with the only exceptions in the Ericales, a sister group to the other asterids. Impatiens, a genus of the Ericales, includes species with one integument, two integuments, or an apparently intermediate bifid integument. A comparison of the development of representative Impatiens species and analysis of the expression patterns of putative orthologs of the Arabidopsis thaliana ovule development gene INNER NO OUTER (INO) has enabled us to propose a mechanism responsible for morphological transitions between integument types in this group. We attribute transitions between each of the three integument morphologies to congenital fusion via a combination of variation in the location of subdermal growth beneath primordia and the merging of primordia. Evidence of multiple transitions in integument morphology among Impatiens species suggests that control of underlying developmental programs is relatively plastic and that changes in a small number of genes may have been responsible for the transitions. Our expression data also indicate that the role of INO in the outgrowth and abaxial-adaxial polarity of the outer integument has been conserved between two divergent angiosperms, the rosid Arabidopsis and the asterid Impatiens.


The angiosperm ovule has long been a subject of classical morphological study, and in recent years it has been used as a model for understanding the molecular genetic basis of morphogenesis (reviewed in Skinner et al., 2004). Although relatively conserved in overall structure, ovules have undergone distinct morphological changes during angiosperm diversification. An angiosperm ovule commonly consists of a funiculus (supporting stalk), one or two integuments (lateral sheathing structures), and a terminal nucellus in which megasporogenesis, megagametophyte development, fertilization, and embryogenesis occur. After fertilization, the integuments differentiate to become the seed coat.

One morphological change that has occurred during the evolution of angiosperm ovules is reduction in the number of integuments. Most basal angiosperm taxa are bitegmic, that is, their ovules possess two integuments (Endress and Igersheim, 2000), and the earliest angiosperms were likely bitegmic (Doyle and Endress, 2000). Ovules of several diverse angiosperm lineages have a single integument (Philipson, 1977), indicating that unitegmy has arisen several times during angiosperm evolution (Stebbins, 1974; Bouman, 1984).

In these lineages, unitegmy could have evolved via the loss of one integument or via the fusion of two ancestral integuments. Although also occurring in other taxa, unitegmy is a shared character of the euasterids (Endress, 2001), in which it has been proposed to result from congenital fusion (Stebbins, 1974). Within the broader asterid clade, there are several early branching lineages, notably among the Ericales, that include both unitegmic and bitegmic species (Albach et al., 2001). The Ericalean genus Impatiens (Balsaminaceae) (Morton et al., 1996; Magallon et al., 1999) is one such taxon that includes bitegmic species, unitegmic species, and species with an intermediate integument morphology.

Congenital fusion involves the loss of primordial boundaries and is associated with the diversification of a variety of floral organs (Sattler, 1977; Verbeke, 1992; Endress, 1994). Congenital fusion can result from two primordia moving closer together and merging to give rise to a single structure or from intercalary growth between primordia. A variation on the first of these mechanisms, referred to as integumentary shifting, has been proposed for the congenital fusion of integuments in Impatiens (Boesewinkel and Bouman, 1991). In this model, the outer integument primordium arises on top of the inner integument, and only the tip of the mature integument derives from the outer integument.

Another aspect of ovule development that varies among angiosperms is the orientation of ovule curvature relative to the gynoecium. In gynoecia with axial, parietal, marginal, or free central placentation, anatropous ovules may curve toward the base of the gynoecium (gynobasal curvature) or toward the insertion point of the stigma (gynoapical curvature). The orientation of ovule curvature may be conserved among large clades of angiosperms, such as the euasterids (Payer, 1966), or may vary within a family, as in the Ericaceae (Payer, 1966).

Ovule development has been studied in detail in model species such as Arabidopsis thaliana. Some aspects of the genetic and molecular regulation of integument development have been elucidated (reviewed in Skinner et al., 2004). One Arabidopsis gene, INNER NO OUTER (INO), is required for the development of only the outer integument. In the ino mutant, the outer integument is absent, and asymmetric growth and curvature of the ovule are also lost (Baker et al., 1997). INO has been cloned and found to encode a member of the YABBY family of putative transcription factors (Villanueva et al., 1999). Members of the YABBY gene family are implicated in the establishment of abaxial identity of lateral organs (Bowman, 2000). YABBY gene family members are expressed predominantly on the abaxial side of lateral organs, including leaves, sepals, petals, stamens, and carpels. INO is expressed on the gynobasal side of the ovule at the site of outer integument initiation and in the abaxial cell layer of the outer integument in Arabidopsis (Villanueva et al., 1999; Balasubramanian and Schneitz, 2000). Because INO is expressed only in the outer integument, we reasoned that INO might serve as a marker for outer integument identity in ovules that possess only one integument. Recently, Yamada et al. (2003) examined the expression of an INO ortholog in the basal angiosperm genus Nymphaea. They interpreted their results as indicating expression in both the inner and outer integuments. However, the relatively weak signal observed in the inner integument was similar to that in negative controls; thus, the pattern could also be interpreted as representing expression only in the outer integument. Further analysis will be required to fully resolve the expression pattern in Nymphaea.

To gain insight into the mechanism of integument reduction in Impatiens, which may reflect the derivation of unitegmy in other asterids, we have studied integument development in species of Impatiens representing three different ovule morphologies. As part of this investigation, we cloned orthologs of INO from Impatiens species and used these clones to examine the expression patterns of INO orthologs in Impatiens species representing the different ovule morphologies. These studies allow us to evaluate the conservation of expression patterns among INO orthologs and to propose a model for the mechanisms underlying diversification of integument morphology in Impatiens.


Ovule and Integument Development in Impatiens

Ovule morphology has previously been characterized in several Impatiens species. Ovule morphologies reported in the literature are listed in Table 1. Five of these species were chosen for further study and are denoted with boldface text in the table. In the examined species, the reported ovule morphology was either confirmed or is shown corrected in Table 1. Impatiens sodenii (Impatiens oliveri) was reported to have an intermediate morphology (Narayana, 1965) but was found to be unitegmic in this study. One species from each morphological class was then chosen for a more detailed developmental analysis, as illustrated in Figures 1 and and22.

Figure 1.
Scanning Electron Microscopy of Developing Impatiens Ovules.
Figure 2.
Light Microscopy of Developing Impatiens Ovules.
Table 1.
Reported and Confirmed Integument Morphology of Impatiens Species

Ovule primordia arose similarly in all three Impatiens species characterized (Figures 1A, 1D, and 1G). Ovule primordia were initiated basipetally in the five carpels of the gynoecium with axial placentation. The ovule primordia began to curve gynobasally before integument initiation. At anthesis, the ovules had curved gynobasally far enough to position the micropyle adjacent to the placenta. Ovules were crowded in the gynoecium and often twisted as they grew, so they no longer appeared to be organized in a neat row.

Impatiens hookeriana had both an inner and an outer integument at anthesis (Figures 1C and and2C),2C), as reported previously by Narayana (1965). The inner integument was initiated by a periclinal division in a ring of L1 cells on the ovule primordium (Figures 1A and and2A).2A). Subsequent anticlinal and periclinal divisions of the two resulting cell layers caused the integument to grow out from the ovule axis and eventually enclose the nucellus, so that the mature inner integument had approximately four cell layers (Figure 2C). The adaxial layer adjacent to the nucellus differentiated into an endothelium. The outer integument was also initiated by periclinal division in the L1 on the proximal (toward the placenta) side of the inner integument along the ovule axis (Figures 1B and and2B).2B). However, periclinal and oblique divisions in the L2 layer beneath the initial integument primordium provided much of the subsequent growth. These divisions extended the outer integument while the L1 cells divided anticlinally to accommodate the underlying growth (Figure 2C). The outer integument never grew so far as to completely cover the inner integument, and the micropyle was formed by the inner integument alone (Figures 1C and and2C2C).

Impatiens balsamina had a single bifid integument at anthesis (Figures 1F and and2F),2F), as reported previously by Takao (1966). An inner integument primordium initiated in much the same way as the inner integument of I. hookeriana, by a periclinal division followed by anticlinal divisions (Figures 1D and and2D).2D). An outer integument primordium again arose proximal to the inner integument primordium, first by a periclinal division and then by subsequent anticlinal divisions (Figure 2E). However, at this stage, the L2 layer began to divide periclinally and obliquely and push out both the outer and inner integument primordia. The physical remnants of the two primordia remained distinct at the tip of the single integument (Figures 1E and and2E).2E). The L1 of both the inner and outer integument primordia continued to divide anticlinally to accommodate the growth of the subdermal layers. Again, the micropyle was formed from the part of the integument derived from the inner integument (Figures 1F and and2F2F).

Impatiens niamniamensis had a single integument at anthesis (Figure 2I), as reported previously by Boesewinkel and Bouman (1991). The single integument initiated similarly to the inner integument of I. hookeriana, with a periclinal division followed by anticlinal divisions (Figures 1G and and2G).2G). Growth proceeded by anticlinal division of the L1-derived cell layers until the tip of the integument was approximately even with the tip of the nucellus (Figures 1H and and2H).2H). Subsequent periclinal and oblique divisions in both the cells of the integument and the subdermal layers underlying the integument primordium completed the formation of the single massive integument (Figures 1I and and2I).2I). At maturity, the innermost layer of this integument had differentiated into a distinct endothelium (Figure 2I).

Cloning of Impatiens INO Orthologs

A degenerate PCR approach was taken to identify and clone INO orthologs from Impatiens. After cloning and sequencing an initial putative INO ortholog from I. hookeriana, specific primers derived from this sequence were used in combination with degenerate primers to amplify and clone putative INO orthologs from four other Impatiens species. Multiple independent clones were sequenced, but all derived from a single putative ortholog from the genomic DNA of each of the five species. A nearly complete sequence was obtained for I. niamniamensis, and partial sequences were obtained for I. hookeriana, I. sodenii, I. walleriana, and I. balsamina.

Alignment of the Impatiens putative INO ortholog amino acid sequences with Arabidopsis INO (Villanueva et al., 1999), the other Arabidopsis members of the YABBY protein family (Bowman and Smyth, 1999), and Nymphaea INO orthologs (Yamada et al., 2003) is illustrated in Figure 3, which shows that the putative INO orthologs share amino acid residues unique to INO within the YABBY protein family.

Figure 3.
Alignment of Predicted Amino Acid Sequences for Arabidopsis YABBY Protein Family Members, Nymphaea Orthologs (Yamada et al., 2003), and Impatiens INO Orthologs.

Prior analysis (Yamada et al., 2003, 2004; Yamaguchi et al., 2004; R.K. Kuzoff and C.S. Gasser, unpublished data) has shown that the INO proteins constitute a well-supported clade among the other YABBY proteins. Figure 4 shows an unrooted tree of putative Impatiens INO amino acid sequences, Arabidopsis YABBY sequences (Bowman and Smyth, 1999; Villanueva et al., 1999), and Nymphaea INO sequences (Yamada et al., 2003). Bootstrap support for groupings is shown when it was >50%. Additional phylogenetic analyses were performed omitting the two least complete Impatiens sequences, using only regions present in all of the sequences, and a combination of these two (data not shown). The topology of the tree was unaltered by such changes in the data set, and groupings were supported by similar bootstrap values in all of these analyses, a further indication of the significance of the relationships indicated in Figure 4. The Impatiens sequences fell into a group with 91% bootstrap support that also included the Arabidopsis and Nymphaea INO sequences. The clustering with other INO sequences and the high level of similarity among the Impatiens sequences led us to conclude that the sequences we isolated represent Impatiens INO orthologs (although the possibility that additional INO orthologs exist in these species cannot be excluded). The sequence from the bitegmic I. hookeriana appeared to be sister to a branch comprising the INO proteins of the other four Impatiens species, each of which has a bifid or single integument. A similar result was achieved when Impatiens INO intron sequences were used to construct an unrooted tree (data not shown). Based on INO sequences, the four bifid and unitegmic species examined appeared to be more closely related to each other than to the bitegmic I. hookeriana.

Figure 4.
One of the Three Shortest Trees Showing Possible Relationships among Impatiens INO Proteins and Other Published YABBY Proteins.

Expression Patterns of INO Orthologs in Impatiens

An antisense probe from the I. niamniamensis INO ortholog was used for in situ hybridization to determine the pattern of INO mRNA accumulation in ovules of Impatiens species with different integument morphologies. Use of this single probe for all species ensures that the hybridization signals observed in different species will reflect the expression of similar genes.

Figure 5A shows that in the bitegmic I. hookeriana, INO mRNA was first detected just before the initiation of the outer integument in the two rows of L1 cells that would become the outer integument initials. Expression was confined to the outermost cell layer of the expanding outer integument (Figure 5B) and persisted in these layers as the ovule neared maturity (Figure 5C). Thus, INO expression was associated with the L1 layer of the outer integument from its inception and was excluded from the subdermal layers that gave rise to the bulk of the outer integument (Figures 5A to 5C). No signal was detected in any species examined when the sense I. niamniamensis INO sequence was used as a probe (Figures 5D, 5H, and 5L).

Figure 5.
In Situ Hybridization with Antisense I. niamniamensis INO and Sense I. niamniamensis INO Probes on Developing Impatiens Ovules.

In the bifid integument of I. balsamina, INO mRNA was again first observed just flanking the inner integument, at the site where the outer integument primordium would appear (Figure 5E). Hybridization did not extend to the tip of the outer lobe of the single integument but was confined to the proximal part of the single bifid integument (Figure 5F). As the ovule neared maturity, expression of INO was confined to the outer lobe of the single bifid integument (Figure 5G). INO mRNA levels decreased to nearly undetectable levels by anthesis (Figure 5G).

The single integuments in I. niamniamensis (Figures 5I and 5K) and I. sodenii (Figure 5J) showed expression of INO in the same pattern observed in I. balsamina. Expression of INO was confined to the abaxial cell layer of the proximal part of the single integument throughout integument development (Figures 5I to 5K). This same expression pattern was also observed in the unitegmic ovules of I. walleriana (data not shown).

The appearance of an apparently broader region of INO expression in Figures 5E and 5I compared with Figure 5A is attributable to variations in the plane of the section and does not reflect an actual difference in expression pattern among the species illustrated.


Although the embryology and integument morphology of several Impatiens species have been described previously (Table 1), we were not always able to confirm the reported morphology. Narayana (1965) reported that I. sodenii (I. oliveri) had an intermediate integument morphology, but it was found to be unitegmic in this study. This discrepancy could be attributable to the misidentification of the species in either study, or possibly to variability within the species. The reported integument morphology was confirmed for four of the five species in this study and in all three of the species chosen for the most extensive analysis.

Our observations lead us to propose a model for the diversification of integument morphology in Impatiens. The observations and model are summarized in Figure 6. We propose that congenital fusion of the inner and outer integuments and hence the evolution of different integument forms in Impatiens results from changes in the timing and location of primordium initiation and subsequent subdermal growth. In the bitegmic species of Impatiens examined, the outer integument was initiated later than the inner integument, and growth that occurred from subdermal layers contributed only to the outer integument (Figure 6A). In intermediate species, initiation of the outer integument closely followed initiation of the inner integument, and subdermal growth occurred in the area underlying both integuments, so that the primordia grew out as a unit rather than as separate structures (Figure 6B). In unitegmic species, we hypothesize that the remnants derived from ancestral inner and outer integument primordia initiate simultaneously and overlap each other so that only one primordium is evident. Subdermal growth beneath this single structure then gives rise to a single integument (Figure 6C). This model for congenital fusion involves both the merging of primordial boundaries and a change in the location and/or timing of subdermal growth (Figure 6). The occurrence of an intermediate ovule morphology in Impatiens is one piece of evidence supporting the hypothesis that unitegmy has arisen via congenital fusion rather than via the loss of one of the integuments.

Figure 6.
Model for the Derivation of Integument Morphologies in Impatiens.

Our model includes the hypothesis that the inner layers of the single integument in unitegmic species will have properties associated with the inner layers of the inner integument of bitegmic species and that the outer layers of the single and outer integuments will also share similar features. Prior work has shown that the integumentary layer adjacent to the nucellus and embryo sac differentiates to form a characteristic endothelial layer late in ovule development in bitegmic, unitegmic, and intermediate species (Narayana, 1965; Takao, 1966; Boesewinkel and Bouman, 1991) (visible in a unitegmic species in Figure 2I). The conservation of this layer is consistent with our hypothesis of an inner integument derivation for the layers adjacent to the nucellus in all three ovule forms.

We tested the hypothesis concerning the presence of outer integument tissue by examining expression patterns of Impatiens INO orthologs. In situ hybridization using the I. niamniamensis putative INO ortholog as a probe revealed a pattern of expression in the outer integument of I. hookeriana that was essentially identical to that observed for INO in Arabidopsis. In this bitegmic Impatiens species, INO was expressed only in the abaxial cell layer of the outer integument, and not in the inner integument. Furthermore, INO was expressed only in the cells directly derived from the initial outer integument primordium, not in any of the subdermal cell layers that subsequently gave bulk to the outer integument. This conservation of expression pattern between Impatiens and Arabidopsis may indicate a conservation of function between Arabidopsis and Impatiens INO orthologs. Independent of conservation of function and orthology, the pattern of INO mRNA accumulation can serve as a marker for outer integument identity in this genus because the pattern of hybridization observed with our putative Impatiens INO probe was specific to the outer integument in I. hookeriana. Thus, our observation of INO expression in a similar pattern in the single and bifid integuments of other Impatiens species supports our hypothesis that these structures include tissue deriving from an ancestral outer integument. Together, the presence of the endothelium and the expression pattern of INO support the hypothesis that the single integument primordium is derived from both ancestral outer and inner integument primordia. Our results suggest that unitegmy has evolved by a combination of the simple merging of integument primordia and a change in the location of subdermal growth rather than by integumentary shifting of the outer integument onto the inner integument.

Yuan et al. (2004) have assembled an extensive phylogeny of Impatiens using ribosomal internal transcribed spacer sequences from a sample of 111 species. A subset of this tree, including those species for which ovule morphology has been reported, is shown in Figure 7. The Yuan et al. (2004) phylogeny places the clade including I. hookeriana sister to the clade containing the other four species included in our study, with all five species included on a branch of the larger tree. Our phylogram based on Arabidopsis YABBY and Impatiens and Nymphaea INO amino acid sequences (Figure 4) and an unrooted tree constructed with Impatiens INO intron sequences (data not shown) also have I. hookeriana sister to a clade including the other species in our study. Thus, it is possible that the intermediate and unitegmic states are derived among the species we studied. However, in the Yuan et al. (2004) phylogeny, additional species that have been reported to have an intermediate integument morphology (e.g., Impatiens arguta [Narayana, 1963]) are placed more basally in the complete Impatiens clade. This finding suggests that transitions between bitegmic, unitegmic, and intermediate integument morphologies have occurred repeatedly during the diversification of Impatiens. The nearest outgroups to the Impatiens clade in the Yuan et al. (2004) phylogeny have been reported to be bitegmic (L. Watson and M.J. Dallwitz, http://delta-intkey.com), so the intermediate morphology of more basal Impatiens species is hypothesized to derive from bitegmy.

Figure 7.
Distribution of Integument Morphologies among Impatiens and Relatives.

The apparent relatively frequent transition between the bitegmic, bifid, and unitegmic states in the early diverging asterids, including the Ebenaceae and Primulaceae (Philipson, 1974), as well as Impatiens, implies a predisposition to such changes in this group. Our model for the evolution of unitegmy may provide a basis for this predisposition in the form of the inclusion of the process of subdermal growth in the latter stages of outer integument development (Figure 6). A simple change in the location of such growth from only beneath the outer integument primordium to the region beneath both integument primordia would lead to the bifid single integument. Fusion of the initial primordia into a single structure and subdermal growth beneath this compound structure would produce the fully unitegmic state. By contrast, a switch to the original pattern of subdermal growth would result in a transition to the bitegmic state. Thus, if the controls for the precise spatial location of subdermal growth are relatively plastic, the pattern of multiple independent transitions between the intermediate integument morphology, bitegmy, and unitegmy in early diverging asterids can be explained. In the euasterids, the unitegmic state had apparently become fixed in the common ancestor of this subgroup.

In Arabidopsis, INO expression was observed on the gynobasal (previously referred to as abaxial, posterior, or ventral) side of the ovule (Robinson-Beers et al., 1992; Schneitz et al., 1998; Villanueva et al., 1999), and it was hypothesized that one role of INO might be in establishing the identity of the gynobasal side of the ovule (Villanueva et al., 1999). However, initial INO ortholog expression in Impatiens was not on the gynobasal side of the ovule but on the gynoapical side. Thus, in both Arabidopsis and Impatiens, the location of INO ortholog expression was on the side of the ovule where maximal growth of the asymmetric outer (or single) integument would occur. INO ortholog expression is thus associated with the side of the ovule that undergoes the most outer integument growth, not with a particular direction relative to the gynoecium. These results support the function of INO as a promoter of outer integument growth rather than as a determinant of ovule orientation. ino mutant ovules do not curve properly, showing that asymmetric growth of the outer integument is required for normal ovule curvature. The pattern of promotion of outer integument growth by INO may direct whether the ovule will curve gynoapically, as in Arabidopsis, or gynobasally, as in Impatiens.

Members of the YABBY gene family have been shown to be expressed on the abaxial sides of primary lateral organs in plants and have been associated with the determination of abaxial identity in these organs (Siegfried et al., 1999). Although INO does not provide gynobasal or gynoapical identity in the ovule itself, INO may function to promote abaxial identity in the outermost layer of the outer integument because it is expressed in those cells in both Arabidopsis and Impatiens. The juxtaposition of abaxial and adaxial identity has been proposed to be necessary for the laminar growth of lateral organs (Waites and Hudson, 1995; Bowman et al., 2002). Such a role for INO would explain how INO could be necessary for laminar growth of the outer integument. The conserved expression pattern of INO in two diverse genera of eudicots suggests that the outer integument shares features of its development with other lateral organs, such as leaves, petals, stamens, and carpels, all of which exhibit abaxial expression of YABBY family members. YABBY expression is absent from shoots as well as from the ovule axis, and the ovule itself undergoes a developmental program that is distinct from that of lateral organs and is more analogous to (and possibly structurally homologous with) shoot development.

Changes in INO expression or function probably were not involved in the evolution of integument morphology in Impatiens. The expression pattern of INO has been conserved between the asterid Impatiens and the rosid Arabidopsis, suggesting that its function may also be conserved at least within the eudicots. The unitegmic phenotype of the ino mutant suggested the possible involvement of INO in the diversification of integument morphology. Although this does not appear to have been the case in Impatiens, the expression pattern of INO allowed us to formulate a model for how integument morphology has diversified in this genus. Analysis in Impatiens of other ovule development genes identified in Arabidopsis may shed further light on integument fusion. Mutations in ABERRANT TESTA SHAPE cause a single integument to arise in place of the two found in wild-type Arabidopsis (Leon-Kloosterziel et al., 1994), indicating a possible role for this gene class in the evolution of integument fusion.


Plant Material

Cuttings of Impatiens niamniamensis and Impatiens sodenii were taken and cultivated at the University of California (UC) Davis Botanical Conservatory. Cuttings of Impatiens hookeriana were acquired from the UC Berkeley Botanical Garden and cultivated at UC Davis. Impatiens walleriana seeds were acquired from the Universidade de Coimbra Jardin Botanico (Coimbra, Portugal), and Impatiens balsamina seeds were from Civico Orto Botanico (Trieste, Italy) via Holly Forbes of the UC Berkeley Botanical Garden. Plants were grown in the greenhouse at UC Davis at 15 to 27°C with a 16-h daylight regimen.


Pistils were dissected from flowers at various stages of development for fixation. Tissue fixation and plastic sectioning were done according to Baum and Rost (1996) with modifications. Tissue was infiltrated with Histocryl Resin (London Resin Company, Hampshire, UK) for 2 weeks at 4°C. Plastic sections of 3 μm were stained using the periodic acid Schiff's reagent reaction followed by a counterstain of 1% aniline blue-black in 7% acetic acid. A Zeiss (Oberkochen, Germany) Axioplan microscope was used to view stained sections with bright-field illumination and in situ hybridization sections with differential interference contrast optics. Images were acquired with an MDS290 digital camera (Kodak, New Haven, CT) and edited in Photoshop version 7.0 (Adobe, San Jose, CA).

Scanning electron microscopy was as described by Broadhvest et al. (2000).

Isolation of INO Orthologs

Genomic DNA was extracted according to the cetyl-trimethyl-ammonium bromide (CTAB) protocol (Doyle and Dickson, 1987). mRNA was extracted using the mRNA Direct kit (Dynal, Oslo, Norway), and cDNA was made using the Superscript II reaction (Gibco BRL, Invitrogen, Grand Island, NY). The degenerate primers Yab4oR (designed to amino acids in YABBY region AFSLAAKN; 5′-RTTYTTIGCIGCIRDRCTRAAIGC-3′) and ZnF1FB (designed to amino acids in the zinc finger region ICHVQCG; 5′-YAYGTICARTGYGGITWYTGYAC-3′) were used to amplify the INO ortholog from I. hookeriana genomic DNA. A band hybridizing to a tomato INO probe was extracted from a gel and reamplified with the nested degenerate primers Yab4iR (designed to amino acids MAHKEAFSLA; 5′-RCTRAAIGCYTSYTTRTGIGYCAT-3′) and ZnF2FB (designed to amino acids MVVTVRCG; 5′-AWRGTIGTIACIGTNAGRTGYGG-3′). The resulting PCR product was gel-purified and cloned using the TOPO-TA cloning kit (Invitrogen). Clones were sequenced using m13–21 and m13 reverse primers. The clone identified that contained the I. hookeriana INO fragment was designated pJMM18. The sequence of this clone was used to design the specific Impatiens INO primers ImpF1 (5′-ACTGCACCACTCTCCTCTCTG-3′) and ImpR1 (5′-ATAAGCGGATGGAGCCCTTTG-3′). ZnF1FB and ImpR1 were used to amplify the INO orthologs from I. niamniamensis, I. walleriana, and I. sodenii DNA. These PCR products were similarly cloned and sequenced (designated pJMM23, pJMM24, and pJMM25, respectively). The primers ImpF1 and ImpR1 were used to amplify the INO ortholog from I. balsamina DNA. This PCR product was sequenced directly using the ImpF1 and ImpR1 primers. The specific primers InF1 (5′-CTCGTCCCATATCATGTCGTCGG-3′) and InF2 (5′-CGGATGATGAGGACGACGATTTG-3′) were designed for 3′ rapid amplification of cDNA ends (RACE) with I. niamniamensis mRNA. The Generacer kit (Invitrogen) was used for 3′ RACE, and the product was cloned using the TOPO-TA kit. The clone containing the 3′ end of the I. niamniamensis INO ortholog was designated pJMM26. Primers were then designed to the end of the 3′ RACE product and the farthest 5′ end of the genomic DNA clone, INF3 (5′-TATGTGCAGTGCGGTTATTGC-3′) and INR3 (5′-TCTCGATATCGTTGTTGCCAC-3′), to amplify the nearly full-length cDNA from I. niamniamensis. This cDNA product was cloned and designated pJMM28.

Alignment and Phylogenetic Analysis

Impatiens INO amino acid sequences were aligned with the Arabidopsis thaliana YABBY protein sequences (Bowman and Smyth, 1999) and Nymphaea INO sequences (Yamada et al., 2003) using ClustalX version 1.81 (Thompson et al., 1997) followed by hand editing to optimize the alignment. Phylogenetic trees were generated in PAUP 4.0b10 (Swofford, 1999) using parsimony and the branch and bound search algorithm. In parsimony analysis, characters were weighted using a BLOSUM62 amino acid substitution matrix, which was generated from the published matrix (Henikoff and Henikoff, 1992) by substituting 15 – n for each value n of the matrix. This results in all nonnegative values, ranging from 0 to 15. An amino acid-to-gap transition was given a cost equal to the highest value in the matrix, 15. Truncations at the ends of Impatiens sequences were evaluated as “unknown” characters. The statistical significance of clades was evaluated with 500 bootstrap replicates using the same search criteria.

In Situ Hybridization

Pistils were dissected from flowers at various stages and fixed in either 2.5% paraformaldehyde plus 0.5% glutaraldehyde in 1× PBS or FAA (10% formaldehyde, 47.5% ethanol, and 5% glacial acetic acid [v/v]). After dehydration and embedding in Paraplast X-tra (Fisher Scientific, Pittsburgh, PA), tissue was sectioned to 10 μm and mounted on FisherBrand Probe-on Plus (Fisher Scientific) slides. Sections were prepared for hybridization, hybridized, and washed according to Ferrándiz et al. (2000), with the following modifications. The probe was diluted to ∼10 pg/μL and applied directly to the slides on a slide warmer at 50°C. Slide sandwiches were made and placed in an incubator at 53°C overnight.

The clone pJMM28, which contains the nearly full-length I. niamniamensis INO cDNA, was linearized using BamHI for the sense probe and XhoI for the antisense probe. The probe was synthesized using the DIG labeling mix from Roche (Mannheim, Germany) with T7 RNA polymerase for the sense probe and Sp6 RNA polymerase for the antisense probe. The probe was treated with RQ1 DNase (Promega, Madison, WI) and stored in 90% ethanol overnight at −20°C. The next day, the probe precipitate was collected by centrifugation and resuspended in hybridization buffer.

Detection of the hybridizing probe was achieved with the DIG nucleic acid detection kit from Roche.

Impatiens INO ortholog sequence data have been deposited with the EMBL/GenBank data libraries under the following accession numbers: I. sodenii genomic, AY550267; I. balsamina genomic, AY550268; I. niamniamensis genomic, AY550270; I. niamniamensis cDNA, AY550271; I. hookeriana genomic, AY550269; and I. walleriana genomic, AY550266.


We thank Holly Forbes of the UC Berkeley Botanical Garden and Ernesto Sandoval and Douglas Walker of the UC Davis Botanical Conservatory for assistance with the acquisition and cultivation of Impatiens species, Yong-Ming Yuan for sharing results before publication, Dan Potter and Neelima Sinha for advice and comments on the manuscript, Dior Kelley for help with gene characterization, and members of the Gasser laboratory for invaluable discussions. This work was supported by a grant from the National Science Foundation (IBN-9983354) and a National Science Foundation predoctoral fellowship to J.M.M.


The author responsible for distribution of materials integral to the findings presented in this article in accordance with the policy described in the Instructions for Authors (www.plantcell.org) is: Charles S. Gasser (ude.sivadcu@ressagsc).

Article, publication date, and citation information can be found at www.plantcell.org/cgi/doi/10.1105/tpc.104.029207.


  • Albach, D.C., Soltis, P.S., and Soltis, D.E. (2001). Patterns of embryological and biochemical evolution in the asterids. Syst. Bot. 26, 243–262.
  • Baker, S.C., Robinson-Beers, K., Villanueva, J.M., Gaiser, J.C., and Gasser, C.S. (1997). Interactions among genes regulating ovule development in Arabidopsis thaliana. Genetics 145, 1109–1124. [PMC free article] [PubMed]
  • Balasubramanian, S., and Schneitz, K. (2000). NOZZLE regulates proximal-distal pattern formation, cell proliferation and early sporogenesis during ovule development in Arabidopsis thaliana. Development 127, 4227–4238. [PubMed]
  • Baum, S.F., and Rost, T.L. (1996). Root apical organization in Arabidopsis thaliana. 1. Root cap and protoderm. Protoplasma 192, 178–188.
  • Boesewinkel, F.D., and Bouman, F. (1991). The development of bi- and unitegmic ovules and seeds in Impatiens (Balsaminaceae). Bot. Jahrb. Syst. Pflanzengesch. Pflanzengeogr. 113, 87–104.
  • Bouman, F. (1984). The ovule. In Embryology of the Angiosperms, B.M. Johri, ed (New York: Springer-Verlag), pp. 123–157.
  • Bowman, J.L. (2000). The YABBY gene family and abaxial cell fate. Curr. Opin. Plant Biol. 3, 17–22. [PubMed]
  • Bowman, J.L., and Smyth, D.R. (1999). CRABS CLAW, a gene that regulates carpel and nectary development in Arabidopsis, encodes a novel protein with zinc finger and helix-loop-helix domains. Development 126, 2387–2396. [PubMed]
  • Bowman, J.L., Eshed, Y., and Baum, S.F. (2002). Establishment of polarity in angiosperm lateral organs. Trends Genet. 18, 134–141. [PubMed]
  • Broadhvest, J., Baker, S.C., and Gasser, C.S. (2000). SHORT INTEGUMENTS 2 promotes growth during Arabidopsis reproductive development. Genetics 155, 895–907. [PMC free article] [PubMed]
  • Doyle, J.A., and Endress, P.K. (2000). Morphological phylogenetic analysis of basal angiosperms: Comparison and combination with molecular data. Int. J. Plant Sci. 161 (suppl.), S121–S153.
  • Doyle, J.J., and Dickson, E.E. (1987). Preservation of plant samples for DNA restriction endonuclease analysis. Taxon 36, 715–722.
  • Endress, P.K. (1994). Diversity and evolutionary biology of tropical flowers. (Cambridge, UK: Cambridge University Press).
  • Endress, P.K. (2001). Origins of flower morphology. In The Character Concept in Evolutionary Biology, G.P. Wagner, ed (San Diego: Academic Press), pp. 493–510.
  • Endress, P.K., and Igersheim, A. (2000). Gynoecium structure and evolution in basal angiosperms. Int. J. Plant Sci. 161 (suppl.), S211–S223.
  • Ferrándiz, C., Gu, Q., Martienssen, R., and Yanofsky, M.F. (2000). Redundant regulation of meristem identity and plant architecture by FRUITFULL, APETALA1 and CAULIFLOWER. Development 127, 725–734. [PubMed]
  • Henikoff, S., and Henikoff, J.G. (1992). Amino acid substitution matrices from protein blocks. Proc. Natl. Acad. Sci. USA 89, 10915–10919. [PMC free article] [PubMed]
  • Leon-Kloosterziel, K.M., Keijzer, C.J., and Koornneef, M. (1994). A seed shape mutant of Arabidopsis that is affected in integument development. Plant Cell 6, 385–392. [PMC free article] [PubMed]
  • Magallon, S., Crane, P.S., and Herendeen, P.S. (1999). Phylogenetic pattern, diversity, and diversification of eudicots. Ann. Mo. Bot. Gard. 86, 297–372.
  • Morton, C.M., Chase, M.W., Kron, K.A., and Swensen, S.M. (1996). A molecular evaluation of the monophyly of the order Ebenales based upon rbcL sequence data. Syst. Bot. 21, 567–586.
  • Narayana, L.L. (1963). Contributions to the embryology of Balsaminaceae (1). J. Indian Bot. Soc. 42, 102–109.
  • Narayana, L.L. (1965). Contributions to the embryology of Balsaminaceae (2). J. Jpn. Bot. 40, 104–116.
  • Narayana, L.L., and Sayeeduddin, M.S. (1959). A study of the gametophytes of Impatiens leschenaultii. J. Indian Bot. Soc. 38, 391–397.
  • Payer, J.-B. (1966). Traite d'Organogenie Comparee de la Fleur (reprinted from 1857 edition). (Lehre, Germany: J. Cramer).
  • Philipson, W.R. (1974). Ovular morphology and the major classification of the dicotyledons. Bot. J. Linn. Soc. 68, 89–108.
  • Philipson, W.R. (1977). Ovular morphology and the classification of dicotyledons. Plant Syst. Evol. Suppl. 1, 123–144.
  • Raghuveer, M., and Narayana, L.L. (1994). Embryology of Balsaminaceae. I. Feddes Repert. 105, 23–29.
  • Robinson-Beers, K., Pruitt, R.E., and Gasser, C.S. (1992). Ovule development in wild-type Arabidopsis and two female-sterile mutants. Plant Cell 4, 1237–1249. [PMC free article] [PubMed]
  • Sattler, R. (1977). Fusion and continuity in floral morphology. Notes R. Bot. Gard. Edinb. 36, 397–405.
  • Schneitz, K., Balasubramanian, S., and Schiefthaler, U. (1998). Organogenesis in plants: The molecular and genetic control of ovule development. Trends Plant Sci. 3, 468–472.
  • Siegfried, K.R., Eshed, Y., Baum, S.F., Otsuga, D., Drews, D.N., and Bowman, J.L. (1999). Members of the YABBY gene family specify abaxial cell fate in Arabidopsis. Development 128, 4117–4128. [PubMed]
  • Skinner, D.J., Hill, T.A., and Gasser, C.S. (2004). Regulation of ovule development. Plant Cell 16 (suppl.), S32–S45. [PMC free article] [PubMed]
  • Stebbins, G.L. (1974). Flowering Plants: Evolution above the Species Level. (Cambridge, MA: Belknap Press of Harvard University Press).
  • Swofford, D.L. (1999). PAUP*: Phylogenetic Analysis using Parsimony (and Other Methods), Version 4.0. (Sunderland, MA: Sinauer Associates).
  • Takao, S. (1966). Embryo sac development in Impatiens balsamina. Bot. Mag. Tokyo 79, 437–446.
  • Thompson, J.D., Gibson, T.J., Plewniak, F., Jeanmougin, F., and Higgins, D.G. (1997). The CLUSTAL_X Windows interface: Flexible strategies for multiple sequence alignment aided by quality analysis tools. Nucleic Acids Res. 25, 4876–4882. [PMC free article] [PubMed]
  • Venkateswarlu, J., and Lakshminarayana, L. (1957). A contribution to the embryology of Hydrocera triflora. Phytomorphology 7, 194–203.
  • Verbeke, J.A. (1992). Fusion events during floral morphogenesis. Annu. Rev. Plant Physiol. Plant Mol. Biol. 43, 583–598.
  • Villanueva, J.M., Broadhvest, J., Hauser, B.A., Meister, R.J., Schneitz, K., and Gasser, C.S. (1999). INNER NO OUTER regulates abaxial-adaxial patterning in Arabidopsis ovules. Genes Dev. 13, 3160–3169. [PMC free article] [PubMed]
  • Waites, R., and Hudson, A. (1995). phantastica: A gene required for dorsoventrality of leaves in Antirrhinum majus. Development 121, 2143–2154.
  • Yamada, T., Ito, M., and Kato, M. (2003). Expression pattern of INNER NO OUTER homologue in Nymphaea (water lily family, Nymphaeaceae). Dev. Genes Evol. 213, 510–513. [PubMed]
  • Yamada, T., Ito, M., and Kato, M. (2004). YABBY2-homologue expression in lateral organs of Amborella trichopoda (Amborellaceae). Int. J. Plant Sci. 165, 917–924.
  • Yamaguchi, T., Nagasawa, N., Kawasaki, S., Matsuoka, M., Nagato, Y., and Hirano, H.Y. (2004). The YABBY gene DROOPING LEAF regulates carpel specification and midrib development in Oryza sativa. Plant Cell 16, 500–509. [PMC free article] [PubMed]
  • Yuan, Y.-M., Song, Y., Geuten, K., Rahelivololona, E., Wohlhauser, S., Fischer, E., Smets, E., and Küpfer, P. (2004). Phylogeny and biogeography of Balsaminaceae inferred from ITS sequences. Taxon 53, 391–403.

Articles from The Plant Cell are provided here courtesy of American Society of Plant Biologists
PubReader format: click here to try


Save items

Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...