SEARCH

SEARCH BY CITATION

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

The Norway spruce (Picea abies) gene DAL2 shows distinct structural similarities to angiosperm MADS-box genes which act in the control of the development of the sexual organs of the flower. Transcription of DAL2 is restricted to the reproductive organs, the unisexual cones, of Norway spruce. In this paper we show that DAL2 in the compound female cone is exclusively expressed in the developing ovule-bearing organ, the ovuliferous scale. When expressed constitutively in transgenic Arabidopsis the gene causes developmental alterations very similar to those observed in plants ectopically expressing the Arabidopsis gene AGAMOUS and the closely related Brassica napus gene BAG1. These alterations include homeotic transformations of floral organs. On the basis of these data and analysis of the phylogeny of the plant MADS-box gene family, we propose that DAL2 acts to control reproductive organ development in spruce. We also propose that DAL2 shares a common origin with AGAMOUS and related genes from other angiosperms, in an ancestral MADS-box gene that was active in the control of ontogeny of ovule-bearing organs in the unknown last common ancestor of conifers and angiosperms.


Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

MADS-box genes encode transcription factors which are key components in developmental control systems of angiosperms, including the mechanisms by which the identity of the floral organs is determined. As modelled by Coen & Meyerowitz (1991) distinct combinations of three homeotic gene functions (A, B and C) act in primordia of the four whorls of flower organs to determine their identity. Sepal development in whorl one is directed by function A alone. The combination of A and B confers petal identity to the second whorl organs. B and C regulate stamen identity in whorl three and C activity alone directs carpel identity to the fourth whorl. MADS-box genes carry out all or part of each of these activities. In Arabidopsis, APETALA 1 (AP1) is part of the A activity (Gustafson-Brown et al. 1994;Mandel et al. 1992b), besides its function in the transition from inflorescence to flower meristem identity. The APETALA3 (AP3) and PISTILLATA (PI) pair of MADS-box genes is sufficient to provide the B function (Bowman et al. 1991;Krizek & Meyerowitz 1996b). The C function is executed by the AGAMOUS (AG) gene (Bowman et al. 1991;Mizukami & Ma 1992;Yanofsky et al. 1990). The expression patterns of these flower homeotic genes correlate closely both in time and space with the proposed activity of their gene products (Rounsley et al. 1995).

The highly conserved sequence element shared by these genes, the MADS box, encodes a 60 amino acid MADS domain responsible for sequence-specific DNA binding, and part of the protein–protein interaction properties of the transcription factor. In addition to this conserved element, all plant MADS-box genes described so far, in contrast to the more distantly related yeast and animal MADS-box genes, also contain a second conserved sequence element; the K box, located distal to the MADS box (Ma et al. 1991) and important for the functional specificity of the class B genes (Krizek & Meyerowitz 1996a).

The MADS-box gene family in Arabidopsis contains at least 24 different members which are active at distinct and different stages of the sporophytic phase of the life cycle, as judged from their spatially and temporally different and well defined expression patterns (Rounsley et al. 1995). Phylogenetic analyses show the plant MADS-box gene family to be derived from an ancestral gene by a series of duplications, followed by structural and functional divergence (Purugganan et al. 1995;Tandre et al. 1995;Theißen et al. 1996). In an attempt to trace the duplication history of the family in the seed plants, we have previously screened for genes containing the MADS-box element expressed in reproductive organs of a non-angiosperm plant, the conifer Norway spruce, Picea abies. This screen resulted in the identification of three different MADS-box genes, DAL1, 2 and 3, expressed in both female and male cones. Phylogenetically, each spruce gene is more closely related to genes in angiosperms than to the other spruce genes (Tandre et al. 1995). Thus, the genome of the last common ancestor of the conifers and the angiosperms must have contained a minimum of three different MADS-box genes. Therefore, the series of duplications leading to the present day complexity of this gene class must have been initiated before the split between the lineages that developed into angiosperms and conifers, respectively, at least 325 million years ago (Stewart & Rothwell 1993).

One of the spruce genes, DAL2, is of special interest as its expression is specific to the male and female cones, as judged by Northern analysis. The sequence of the DAL2 gene shows distinct similarities to the angiosperm class C MADS-box. The MADS domain of DAL2 is nearly to identical to those of AG, and the Antirrhinum majus protein PLENA (PLE) (Bradley et al. 1993;Yanofsky et al. 1990) and a phylogenetic analysis based on nucleotide sequence suggested DAL2 to be the sister gene of a set of angiosperm genes including known class C genes (Tandre et al. 1995). In this report we present additional data on the structure and expression of DAL2 which support the hypothesis of a close phylogenetic relationship between DAL2 and the class C genes in angiosperms. We also show that DAL2, when expressed constitutively in transgenic Arabidopsis, causes developmental alterations very similar to those that result from ectopic expression of class C MADS-box genes, but distinct from the effects of ectopic expression of other major classes of MADS-box genes. This indicates that DAL2 also shares major functional characteristics with the class C genes in angiosperms.

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

DAL2 is structurally related to angiosperm MADS-box genes exclusively expressed in the sexual organs

Phylogeny of MADS-box genes. A phylogenetic analysis based on the nucleotide sequences of the MADS-and K-boxes of a selection of well characterised plant MADS-box genes shows that the genes share a common origin, and were grouped into phylogenetically distinct subclasses (Fig. 1). Genes with similar functions in different angiosperms consistently group together in the tree; the flower meristem identity genes AP1 and SQUA, the class B genes AP3/PI and DEFICIENS/GLOBOSA (DEF/GLO), and the class C genes AG and PLE, in Arabidopsis and A. majus, respectively. The Norway spruce gene DAL3 is associated with the tomato gene TM3, of unknown function. DAL1 is associated with a complex clade which includes the petunia gene FBP2, TM5 from tomato and AGL6, AGL2, AGL4 from Arabidopsis. The third spruce gene, DAL2, intercalates in the phylogenetic tree at a position basal to a large clade including a group of ovule-specific genes, AGL11, FBP7 and FBP11; the female organ-specific genes AGL1, AGL5 and ZAG2; and the class C genes PLE, FBP6, PMADS3, AG, BAG1 and ZAG1. This association of DAL2 with class C genes from different angiosperms is consistent with the high degree of amino acid sequence similarity within the MADS domain found between DAL2, AG and PLE, and indicates that DAL2 is phylogenetically most closely related to this subclass of MADS-box genes.

image

Figure 1. Phylogenetic hypothesis for the relationship of seed plant MADS-box genes.

The figure shows the single most parsimonious solution to the problem of relationships of MADS-box genes resulting from an analysis by PAUP (Swofford 1993) of the nucleotide sequences of the MADS-and K-boxes of genes from angiosperms and the conifer Norway spruce. Total branch length is 2276, branches are drawn to scale. The consistency index is 0.290. Numbers indicate percentage of bootstrap resamplings, when higher than 50%, that support indicated branches. Sequences included are from the following accession numbers and references AG: Yanofsky et al. 1990;AGL1, 2, 4, 5, 6: Ma et al. 1991;AGL8:Mandel & Yanofsky (1995b);AGL11: U20182;AP1: Mandel et al. 1992b, AP3: Jack et al. (1992);BAG1: Mandel et al. (1992a);CAL: L36925;DAL1, 2, 3: X80902, X79280, X79281;DEF: Sommer et al. (1990);FBP1, 2: Angenent et al. (1992);FBP6: Angenent et al. (1993);FBP7: X81651;FBP11: Angenent, personal communication;GLO: Tröbner et al. 1992;PI: Goto & Meyerowitz (1994);PLE: Bradley et al. 1993;PMADS1, 2: X69946, X69947;PMADS3: Tsuchimoto et al. (1993);SQUA;Huijser et al. (1992);TM3, 4, 5, 6, 8: Pnueli et al. (1991);ZAG1: Schmidt et al. (1993);ZAG2: X80206. The tomato gene TM8 is oriented as an outlier based on the result of an analysis including also animal and yeast genes (Tandre et al. 1995)

Download figure to PowerPoint

Exon/intron organisation.To further examine the structural relationship between DAL2 and angiosperm MADS-box genes we analysed the exon/intron organisation of the part of the gene which encodes the K domain and part of the C-terminal domain, of DAL2 and angiosperm genes for which such information is available.

Using genomic DNA from Norway spruce as a template in PCR reactions, sequences containing five different introns were cloned and sequenced from DAL2. The positions of these introns in relation to the deduced amino acid sequence of the gene is shown in Fig. 2, together with the corresponding angiosperm-gene data. The alignment was made according to the amino acid sequence of the K domain, no gaps or insertions being introduced. The number of introns within this part of the gene is largely conserved among the angiosperm genes of the different subclasses; four or five introns in each gene. The intron positions are partly, but not entirely, conserved among the genes. A first intron (A in Fig. 2) at the 5′-end of the K box has a conserved position in all genes except AP1, SQUA, and ZAG2, all of which have the intron shifted by 3 bp towards the proximal end of the gene. Intron B has a conserved position in all genes except in AGL5 where it is missing. Intron C is perfectly conserved in position in all genes. Intron D, at the 3′-end of the K box is absent in AGL2 and AGL4 and shifted by 12 bp in GLO and PI towards the proximal end, as compared to the remaining eight genes in which the intron position is identical. Intron E, distal to the K box, is conserved in position in eight genes but, as in intron D, is proximally displaced by 12 bp in the GLO and PI genes. The similar displacement of both introns D and E in these two genes suggests a small deletion to have taken place in exon three of the K box in an ancestor of GLO and PI. In the second pair of B class genes, AP3 and DEF, intron E is distally displaced by 3 bp. In total, similarities and differences in intron positions found among these angiosperm genes are consistent with the phylogenetic hypothesis in Fig. 1.

image

Figure 2. Exon/intron organisation of plant MADS-box genes.

The presence of introns in the K box and part of the 3′-end of MADS-box genes are indicated by triangles. Nucleotide sequences were aligned according to the amino acid sequence deduced from the genomic sequences of angiosperm MADS-box genes and PCR-amplified genomic fragments of DAL2, with no insertions or deletions introduced. References for the angiosperm genes are as for Fig. 1, except for DEF data, which originate from Schwarz-Sommer et al. (1992).

Download figure to PowerPoint

In DAL2 the positions of introns A, B, C, D and E are identical to those of AG and AGL1, although DAL2 differs in the position of one intron from the class A and B genes, AGL2 and AGL4, as well as the female organ-specific AGL5 and ZAG2 genes. The exon/intron organisation of DAL2 thus supports the phylogenetic association with the class C genes and the female organ-specific class of genes represented by AGL1.

DAL2 is specifically expressed in the ovule-bearing scale of the female cone

Previous Northern blot analyses (Tandre et al. 1995) have shown DAL2 to be expressed in the male and female cones, but not in other organs examined. To investigate the spatial distribution of DAL2 expression we used a DAL2 probe in in situ hybridisation experiments to sections of developing female cones. In the cone, the ovule-bearing organs, the ovuliferous scales (each subtended by a sterile bract) are arranged spirally along the axis forming a developmental series from the meristem at the apex towards the base (Fig. 3a). The results show the DAL2 transcript to be present in the developing ovuliferous scale of the cone (Fig. 3b), but not in the subtending bract, in the primary axis of the cone or in the apical meristem. In control hybridisations using a complementary sense probe no hybridisation signal was detected (Fig. 3c). Expression of DAL2 is detectable in the ovuliferous scale very early in development. A hybridisation signal is observed in the axil of the bract primordium at the flanks of the cone apical meristem, at a stage where ovuliferous scale primordia are not yet discernible (Fig. 3d). Expression is maintained in the primordia throughout their development. In the older scales at the base of the cone DAL2 transcript is unevenly distributed. Expression is maintained at high levels predominantly in the proximal-lateral part of the scale where ovules will develop later (Fig. 3e).

image

Figure 3. Anatomy and expression of DAL2 in developing female cones of Norway spruce as detected by in situ hybridisation.

The pictures, except the bright-field micrograph in (a), are dark field micrographs where signal appears as white grains. Bars represent 100 μm.

(a) A longitudinal section through a cone stained with Toluidine Blue. The apical meristem (M); ovuliferous scale (O); and the sterile bract (B) are indicated. A developmental series of scales are seen along the cone, ontogenetically young units being close to the meristem.

(b) Longitudinal section through an ovuliferous scale and the subtending bract, hybridised to an antisense DAL2 probe.

(c) Sense probe control to (b).

(d) The distal part of the cone, including the apical meristem and young scales, hybridised to an antisense DAL2 probe. The apparent signal in cells in the cone axis is artifactual, and observed also in hybridisations to a control sense probe (data not shown). The apical meristem (M); ovuliferous scale (O); the sterile bract (B), and the axil of the bract (A) are indicated.

(e) Cross-section of an ovuliferous scale and bract hybridised to an antisense DAL2 probe.

Download figure to PowerPoint

DAL2 is functionally related to the angiosperm class C MADS-box genes

To examine whether the structural similarities between DAL2 and angiosperm class C MADS-box genes is coupled to similarities in function, we have constructed transgenic Arabidopsis plants expressing DAL2 under the control of the CaMV 35S promoter, and analysed the effects of expression on the development of the plants. Two different DAL2 constructs were used in these experiments. One contained the intact coding sequence of the gene and is referred to as DAL2. In the second construct, NDAL2, the 5′-end of the coding sequence of the Brassica napus class C gene BAG1, encoding an amino-terminal extension of the MADS domain, was fused in frame to the DAL2 coding sequence. This construct was included to test whether this 5′ extension, which is present in class C genes from dicotyledonous plants and maize, but absent from DAL2, were of importance for the function of the gene in Arabidopsis. The DNA constructs were introduced into Arabidopsis by Agrobacterium-mediated transformation, and phenotypic properties scored in the T1 and T2 generations.

Expression of DAL2 alters development of transgenic Arabidopsis.Eleven independent kanamycin-resistant T1 plants harbouring DAL2 and 15 harbouring the NDAL2 construct were collected. One NDAL2 plant was indistinguishable from the wild-type control or from control plants transformed with the vector alone.

The remaining 25 DAL2 and NDAL2 plants all showed phenotypic alterations to different degrees in vegetative as well as reproductive parts. Two classes of transformant plants, class I and II, were distinguished on the basis of a difference in the degree to which morphology differed from wild-type. Plants of both classes had a reduced stem length and leaves were small and curled or folded from the margins upwards/inwards, and initiated flowering earlier than wild-type (data not shown).

The nine DAL2 and 13 NDAL2 plants constituting class I also differed from wild-type in their floral anatomy (data not shown). Flower buds were rounded or triangular in appearance and opened prematurely. In some flowers the sepals were of aberrant shape or bent outwards. Petal development in 17 of the class I plants appeared normal. In five of the class I plants (two DAL2 and three NDAL2 plants), some flowers had narrow or curled petals, or irregularly shaped petal margins, and stomata were present on the abaxial side of the petals.

The class II plants (one NDAL2 and two DAL2 plants) were defined by a more severely aberrant floral morphology and homeotic transformations of floral organ identity. In the two DAL2 plants, organs in sepal positions had gained carpel characteristics; stigmatic tissue was present at the apex, and ectopic ovules with a wild-type appearance formed on these organs in some flowers (Fig. 4b). Organs in petal position were partly or entirely stamen-like (Fig. 4c), whereas the androecium and gynoecium of the flowers were essentially like the wild-type (Fig. 4a). The number of organs in the second whorl was variable between different flowers. The single class II NDAL2 plant showed very similar homeotic transformations of floral organ identity (data not shown).

image

Figure 4. Phenotype of wild-type and class II transgenic DAL2 and NDAL2 Arabidopsis plants.

Scanning electron micrographs are shown. Bars in (a) and (b) represent 500 μm; in (c) 100 μm; in (d), (e), (f) and (g) 20 μm.

(a) Mature wild-type flower.

(b) Mature DAL2 flower. The arrow points to ovule-like structures along a sepal margin.

(c) Detail of a mature DAL2 flower. An anther of a stamen (left) and the anther-like proximal part of a transformed second-whorl organ (right).

(d)-(g) Epidermal cell patterning of second and third whorl organs of a mature NDAL2 flower; a filament of a stamen (d); the proximal part of a second-whorl organ (e); the anther of a stamen (f); and the distal part of a second-whorl organ (g).

Download figure to PowerPoint

The epidermal cell morphology of the proximal filament-like part of the second whorl organs in the class II plants (Fig. 4e) was very similar to the stamen filament epidermis in the wild-type as well as in the DAL2 and NDAL2 plants (Fig. 4d). The anther epidermis in the NDAL2 plants (Fig. 4f), as well as in the wild-type and DAL2 plants (data not shown), was characterised by the presence of stomata and interdigitated cells with a rugose surface. The distal anther-like part of the transformed second whorl organs of the DAL2 and the NDAL2 plants (Fig. 4g) had very similar cell patterning.

The severity of the alterations in flower phenotype increased acropetally in the transgenic plants, most prominently in the second whorl organs where the morphology gradually changed from wild-type petal appearance to stamen-like organs. The class II plants in contrast to wild-type Arabidopsis had determinate inflorescences consisting of unordered carpel-like structures, typically carrying ectopic stigmatic papillary cells and ovules (not shown). Seven of the class I transformant plants had similar terminal inflorescence structures.

Seed set in the transgenic plants varied from rich to none. Only one of the class II plants produced viable seed, 1% of which were kanamycin sensitive, indicating the presence of multiple T-DNA inserts in this plant. Eleven class I plants examined were all fertile. Phenotypic aberrations of vegetative and reproductive development observed in T1 were inherited to the T2 generation. The degree to which plant morphology differed from wild-type differed among the offspring of each plant, indicating a segregation of phenotype.

The offspring of 10 different plants was tested by Northern blot experiments for the presence of a DAL2 transcript. One transcript of the size expected from the sequence of the construct (950 nt) was detected in each sample (data not shown). Seedling samples from three class I plants showed expression of the transgene at high levels, relative to the remaining samples. These progeny plants included ones that were more severely abnormal than the T1 parent plants. The progeny of one of these plants included a small fraction of plants with homeotically transformed floral organs, i.e. class II phenotype. The offspring of the remaining seven class I lines examined showed lower levels of expression and developmental abnormalities similar to or weaker than their parent T1 plants. The probe used did not detect any transcript in wild-type plants or plants transformed with a cDNA construct unrelated to DAL2.

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

Evidence presented in this report, and previously by Tandre et al. (1995), supports the notion that DAL2 is structurally related to the class C MADS-box genes in angiosperms. The degree of sequence similarity within the MADS domain between DAL2 and AG/PLE is remarkably high, considering the large-scale differences in reproductive organ morphology and development, and the long evolutionary distance between conifers and angiosperms. We infer from this extreme similarity that the function of this sequence motif must be similar in DAL2 and the angiosperm class C proteins, and that the structural constraints on the sequence imposed by this function must be strict.

The MADS domain is known to have a primary function in the sequence-specific DNA binding of the angiosperm proteins (Schwarz-Sommer et al. 1992;Tröbner et al. 1992; reviewed by Riechmann & Meyerowitz 1997). Similarities in amino-acid sequence of the MADS domain in AG, AGL1 and AGL2 correlate with similarities in the consensus DNA sequence to which the MADS-domain proteins binds in vitro (Huang et al. 1996). It appears likely that DAL2 and the class C genes share DNA binding properties.

This conclusion is in apparent conflict with data on the functional specificity of chimeric MADS-domain proteins in transgenic Arabidopsis presented by Krizek & Meyerowitz (1996a). These data show that although substantial differences in primary structure exist between MADS domains of different subclasses, MADS domains can be exchanged between proteins without major effects on the specificity in their activity (B.A. Krizek, personal communication;Riechmann et al. 1996). We have no explanation for this apparent conflict, but note that the exchangeability of MADS domains was demonstrated under ectopic expression conditions, which may not precisely mimic the conditions under which the proteins normally function in the plant.

An obvious structural difference between DAL2 and the class C angiosperm proteins is the absence of a protein domain amino-terminal to the MADS domain in DAL2. Such amino-terminal extensions are present in class C proteins in angiosperms. Under ectopic expression conditions in transgenic Arabidopsis, this sequence motif is not essential for protein function, since a hybrid DAL2 gene encoding a DAL2 protein fused to the amino-terminal motif from the Brassica napus class C protein BAG1 conferred very similar developmental defects to the plant as the wild-type DAL2 gene. This latter result is in agreement with the results of Mizukami et al. (1996) on the AG gene. We cannot exclude, however, that the amino-terminal extension may be essential for some specific aspect of the function of the proteins under the conditions in which the class C proteins act in the wild-type flower.

In addition to DNA-binding, the MADS domain of the class C proteins in angiosperms is an essential determinant of the protein–protein interaction properties of the molecule. Homodimer formation of AG proteins is mediated by the C-terminal part of the MADS domain and the linker region distal to the MADS domain (Riechmann et al. 1996). Interestingly, the similarities in primary structure between DAL2 and AG extends from the MADS domain into the linker region. The degree of identity over the 34 amino acids is 56%. Among the conserved residues are those that constitute the TIERYKK motif found in class C proteins. Furthermore, the K domain of DAL2 is significantly more similar to those of the class C proteins (58% identity between DAL2 and AG) than K domains of class C proteins are to those of other major classes of angiosperm MADS-proteins (e.g. 24% identity between AG and AP3). This sequence motif is also thought to mediate protein–protein interactions (Ma et al. 1991;Zachgo et al. 1995). This extended sequence similarity suggests that DAL2 may share not only DNA-binding properties, but also protein–protein interaction properties with the class C proteins.

This suggestion is supported by our results from the experiments using transgenic Arabidopsis plants ectopically expressing DAL2. The alterations in vegetative development in these are similar to the effects observed by us and others in plants ectopically expressing AG or BAG1. However, very similar vegetative developmental abnormalities have been observed in plants ectopically expressing MADS-box genes belonging to other classes;AP1 (Mandel & Yanofsky 1995a), and AP3/PI (Krizek & Meyerowitz 1996a). This developmental abnormality, although characteristic of MADS-box genes, does not distinguish between the different classes of MADS-box genes. The homeotic transformations in floral organs we have observed in the plants expressing DAL2, however, appear to uniquely result from ectopic expression of class C MADS-box genes, like AG (Mizukami et al. 1996;Mizukami & Ma 1992) and BAG1 (Mandel et al. 1992a). In such plants, the class C gene acts by suppressing A function allowing carpels and stamens to develop in place of the perianth organs through the action of the class C gene. Therefore, the effects of DAL2 expression on floral development are similar to the effects of ectopic expression of class C MADS-box genes, but different from the known effects of ectopic expression of other major classes of MADS-box genes. As regards mechanisms, the simplest explanation for our results is that DAL2 activity in the perianth organs can functionally substitute for AG activity in suppressing A gene activity, in directing carpel identity to the outermost whorl, and by interacting with AP3 and PI in directing stamen identity to the second whorl of organs in the transgenic flowers. Alternatively, it is possible that a suppression of class A gene function, mediated by DAL2, may result in an extension of AG expression into the perianth whorls of the transgenic plants, and thus the redirection of identity of the perianth organs might result from the activity of AG, rather than of DAL2, or from the formation of active heterodimers of the two proteins. At present we cannot distinguish between these possibilities. In either case, our data indicate that DAL2 has the capacity to interact with other regulatory protein components and, at least in part, can substitute for class C function in an angiosperm context. It therefore appears likely that the high degree of sequence similarity in the MADS-, linker- and K-domains in part is a result of structural constraints imposed by the requirement for protein–protein interaction.

The results of our phylogenetic analyses imply that the similarities that exist in structure and function between DAL2 and the class C angiosperm genes reflect a common ancestry of the genes, and result from a conservation of features already present in a common progenitor MADS-box gene. In this analysis DAL2 is sister to a clade including known class C genes from mono- and dicots, two genes expressed specifically in carpels (AGL1 and AGL5), and genes expressed specifically in the ovule (FBP7, FBP11 and AGL11). The major branching pattern of the phylogenetic tree is relatively stable to perturbations due to the inclusion of different selections of genes. It is also entirely consistent with functional data, known members of different functional subclasses from different species grouping together in the tree. Furthermore, it is consistent with the information presented in this report on the intron/exon structure of genes, in all cases where such information is available. In fact, the intron positions within the K box and 3′-end of the gene in the case of four major classes of genes appear to provide a reliable indication on the function of the gene. The class A, the different subclasses of B, and class C genes each appear uniquely distinguished by a characteristic intron pattern.

Since available structural and functional evidence supports the tree, we conclude that it represents a valid hypothesis for the evolution by gene duplication of the family of MADS-box genes in seed plants. Based on the assumption that the rooting of the tree is correct, this hypothesis has some interesting implications. First, the gene duplication which resulted in the distinction between class B and C genes must have taken place previous to the speciation event that gave rise to the separate angiosperm and conifer lineages. Previous to that the duplication leading to a B/C branch and a second branch giving rise to the class A genes, as well as the AGL2/4-FBP2 class, must have already taken place. This is consistent with the position in the tree of the spruce DAL1 gene, being positioned in the sister clade to that in which the class A gene AP1 is found. Thus, the three important classes of reproductive control genes, classes A, B and C, according to this hypothesis were already present in the ancestor of both the conifers and the angiosperms. At present there is no evidence for the existence of either A- or B-related genes in conifers, but the notion is interesting in view of the possibility that DAL2 retains the functional capacity to act in an angiosperm, where interactions between A, B and C genes are essential. It is possible, therefore, that the activity of DAL2 in transgenic Arabidopsis reflects a true functional conservation of the mechanisms by which these classes of genes interact since the early seed plants.

Second, the duplications resulting in the divergence of the class C genes and the ovule- and carpel-specific genes included in the clade defined by DAL2, would have occurred after the speciation event leading to the establishment of the separate conifer and angiosperm lineages. These genes are thus candidates for genes that function in the control of morphological features that are specific for the angiosperm lineage, but absent in the early seed plants as well as in the conifers. This is interesting since the morphology of both the ovule-bearing organ and the ovule differ distinctly between conifers and angiosperms. In view of the data on DAL2 function in Arabidopsis, it is a tempting hypothesis that AG has retained the essential functions of the ancestral gene, whereas AGL1 and AGL5 after sequence divergence have been recruited to other functions associated with novel morphological or physiological features of the angiosperm flower. The petunia genes FBP7 and FBP11 are active in the ovule, and are of clear significance for ovule development (Angenent et al. 1995;Colombo et al. 1995).

Taken together, these data suggest that DAL2 may have a role similar to that of class C MADS-box genes in angiosperms, in the control of reproductive organ development in spruce. Our data on the expression of the gene, specifically in the ovuliferous scale is well in accordance with this proposal. This highly specific expression pattern may be quite significant as an indicator of function of the gene since, in angiosperms, reproductive development is tightly controlled at the transcriptional level, the expression pattern of each class of MADS-box gene in the developing flower reflecting its function very closely. The timing of DAL2 expression, in relation to organ ontogeny, highly resembles that of AG in Arabidopsis. The fact that DAL2 expression marks the cells that will form the ovuliferous scale, even before the physical emergence of a primordium, supports the idea of DAL2 being active in the determination of organ identity.

Thus, available data are consistent with DAL2 acting as a determinant of reproductive organ identity in the female cone of Norway spruce. The similarities in structure between DAL2 and the class C angiosperm MADS-box genes very likely result from a conservation of features of an ancestral MADS-box gene in both the conifer and angiosperm lineages. The role of this gene, present in the last common ancestor of angiosperms and conifers in middle carboniferous more than 325 million years ago (Stewart & Rothwell 1993), would have been to control the ontogeny of the ovule-bearing organ in this unknown early seed plant.

Experimental procedures

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

Phylogenetic analysis of nucleotide sequence by parsimony

Sequence alignments were made as described previously (Tandre et al. 1995), with the addition of a gap corresponding to the nucleotides 166–177 in the K box inserted in the sequences of GLO, FBP1, PMADS2 and PI. The parsimony analysis was performed by use of PAUP 3.1.1 (Swofford 1993) as described earlier (Tandre et al. 1995). A separate bootstrap analysis with 1000 bootstrap resamplings of a single replicate with RANDOM addition using TBR was performed.

Exon/intron mapping

Total DNA was extracted from Norway spruce needles as described by Kvarnheden et al. (1995). Partial genomic sequences were amplified by use of thermostable DNA polymerase (Advanced Biotechnologies, London, UK) or AmpliTaq Gold (Perkin Elmer, New Jersey, USA) and the following primers, which were directed towards potential intron flanking sites of DAL2: (1) 5′-ACGATTGAGAGGTACAAGAAG-3′; (2) 5′-AACCTCAAGCTGCTTGAGTTC-3′; (3) 5′-GCTTGAGGTTCGACTTGAAAAAG-3′; (4) 5′-TCTCTGCATGATGTCGATCTCC-3′; (5) 5′-TGTTGCTTGAGGAGATCG-3′; (6)5′GCGAAGAATCTCATTCTCC-3′; (7) 5′-GAGAATGAGATTCTTCGCAGC-3′; and (8) 5′-CTCCTGATGTGCATAGTGATG-3′.

Resulting PCR products were subcloned and the inserts were sequenced.

In situ hybridisation

The in situ hybridisations were performed essentially as described by Jackson (1991). Female cones collected in October 1996 in Uppsala, Sweden were fixed in paraformaldehyde (4%) and glutaraldehyde (0.25%), and 7 μm sections hybridised to 35S-labelled RNA probes, prepared from a 3′ non-coding subclone of DAL2. The slides were coated with NBT2 emulsion (Kodak, NY, US), and exposed for 4–7 weeks. After development the sections were stained in 20% Gill’s hematoxylin (BDH, Poole, UK).

Plastic embedding for light microscopy

Embedding in Technovit 7100 methacrylate was made essentially as recommended by the manufacturer (Heraeus Kulzer, Wehrheim, Germany). Female cones (collected in November in Luleå, Sweden) were fixed in paraformaldehyde/glutaraldehyde (4%/0.25%) supplemented with 0.1 m NaCl, dehydrated, sectioned and stained with 1% Toluidine Blue.

Transgenic plants

Three plasmid constructs were used for plant transformation. The DAL2 clone used was reconstituted from N-terminal and C-terminal fragments of the cDNA clone 3AE, joined at a HindIII site (nucleotides 515–520) introduced as silent muations at the ends of the fragments by site-directed mutagenesis. A HindIII fragment corresponding to the nucleotides 1–270 of the BAG1 cDNA clone pCIT750, (GenBank accession number M99415;Mandel et al. 1992a) was ligated to the C-terminal PCR fragment of DAL2, to produce the NDAL2 construct. The BAG1 clone was used in its entirety as a positive control for the induction of class C over-expression phenotype.

The three cDNA constructs were cloned downstream of the 35S promoter in pHTT202 (Elomaa et al. 1993) and introduced to the Agrobacterium tumefaciens strain C58::pGV2260 by use of tri-parental mating. Resulting Agrobacterium strains were used to transform Arabidopsis thaliana ecotype Wassilewskija by an infiltration protocol (Bechtold et al. 1993), and transformants selected on 50 μg ml–1 kanamycin.

The phenotypic properties of T1 and T2 plants grown under long day conditions at 20–22°C were studied. Samples collected for SEM were fixed in ethanol, formaldehyde and acetic acid, or fixed in glutaraldehyde followed by OsO4 treatment and dehydrated. After critical point drying and gold sputter coating samples were analysed by SEM. Brightness and contrast were adjusted by the Photoshop software (Adobe Inc., CA, USA).

Acknowledgements

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

We thank Charlotta Thornberg and Erik Ullerås for assistance with cloning and plant care, Gary Wife for valuable support and advice on electron microscopy, and Eva Sundberg and Annika Sundås for critical comments on the manuscript. The pHTT202 vector was a gift from Dr Teemu Teeri. Dr Martin F. Yanofsky kindly provided the BAG1 clone. This work was supported by grants from The Foundation for Strategic Research and from The Swedish Natural Sciences Research Council.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  • Angenent, G.C., Busscher, M., Franken, J., Mol, J.N.M. & Van Tunen, A.J. 1992 Differential expression of two MADS box genes in wild-type and mutant petunia flowers. Plant Cell, 4, 983993.
  • Angenent, G.C., Franken, J., Busscher, M., Colombo, L. & Van Tunen, A.J. 1993 Petal and stamen formation in petunia is regulated by the homeotic gene Fbp1. Plant J., 4, 101112.
  • Angenent, G.C., Franken, J., Busscher, M., Van Dijken, A., Van Went, J.L., Dons, H.J.M. & Van Tunen, A.J. 1995 A novel class of MADS box genes is involved in ovule development in petunia. Plant Cell, 7, 15691582.
  • Bechtold, N., Ellis, J. & Pelletier, G. 1993 In planta Agrobacterium mediated gene transfer by infiltration of adult Arabidopsis thaliana plants. C.R. Acad. Sci. Paris, Life Sci., 316, 11941199.
  • Bowman, J.L., Smyth, D.R. & Meyerowitz, E.M. 1991 Genetic interactions among floral homeotic genes of Arabidopsis. Development, 112, 120.
  • Bradley, D., Carpenter, R., Sommer, H., Hartley, N. & Coen, E. 1993 Complementary floral homeotic phenotypes result from opposite orientations of a transposon at the plena locus of Antirrhinum. Cell, 72, 8595.
  • Coen, E.S. & Meyerowitz, E.M. 1991 The war of the whorls: genetic interactions controlling flower development. Nature, 353, 3137.
  • Colombo, L., Franken, J., Koetje, E., Van Went, J., Dons, H.J.M., Angenent, G.C. & Van Tunen, A.J. 1995 The petunia MADS box gene FBP11 determines ovule identity. Plant Cell, 7, 18591868.
  • Elomaa, P., Honkanen, J., Puska, R., Seppänen, P., Helariutta, Y., Mehto, M., Kotilainen, M., Nevalainen, L. & Teeri, T.H. 1993 Agrobacterium-mediated transfer of antisense chalcone synthase cDNA to Gerbera hybrida inhibits flower pigmentation. Bio/Technology, 11, 508511.
  • Goto, K. & Meyerowitz, E.M. 1994 Function and regulation of the Arabidopsis floral homeotic gene PISTILLATA. Genes Devel., 8, 15481560.
  • Gustafson-Brown, C., Savidge, B. & Yanofsky, M.F. 1994 Regulation of the Arabidopsis floral homeotic gene APETALA1. Cell, 76, 131143.
  • Huang, H., Tudor, M., Su, T., Zhang, Y., Hu, Y. & Ma, H. 1996 DNA binding properties of two Arabidopsis MADS domain proteins: binding consensus and dimer formation. Plant Cell, 8, 8194.
  • Huijser, P., Klein, J., Lönnig, W.-E., Meijer, H., Saedler, H. & Sommer, H. 1992 Bracteomania, an inflorescence anomaly, is caused by the loss of function of the MADS-box gene squamosa in Antirrhinum majus. EMBO J., 11, 12391249.
  • Jack, T., Brockman, L.L. & Meyerowitz, E.M. 1992 The homeotic gene APETALA3 of Arabidopsis thaliana encodes a MADS box and is expressed in petals and stamens. Cell, 68, 683697.
  • Jackson, D.P. 1991 In situ hybridization in plants. In: Molecular Plant Pathology: a Practical Approach ( Bowles, D.J. Gurr, S.J. & McPhereson, M., eds.). Oxford: Oxford University Press, pp. 163174.
  • Krizek, B.A. & Meyerowitz, E.M. 1996a Mapping the protein regions responsible for the functional specificities of the Arabidopsis MADS domain organ-identity proteins. Proceedings Natl Acad. Sci. USA, 93, 40634070.
  • Krizek, B.A. & Meyerowitz, E.M. 1996b The Arabidopsis homeotic genes APETALA3 and PISTILLATA are sufficient to provide the B class organ identity function. Development, 122, 1122.
  • Kvarnheden, A., Tandre, K. & Engström, P. 1995 A cdc2 homologue and closely related processed retropseudogenes from Norway spruce. Plant Mol. Biol., 27, 391403.
  • Ma, H., Yanofsky, M.F. & Meyerowitz, E.M. 1991 AGL1-AGL6, an Arabidopsis gene family with similarity to floral homeotic and transcription factor genes. Genes Devel., 5, 484495.
  • Mandel, M.A., Bowman, J.L., Kempin, S.A., Ma, H., Meyerowitz, E.M. & Yanofsky, M.F. 1992a Manipulation of flower structure in transgenic tobacco. Cell, 71, 133143.
  • Mandel, M.A., Gustafson-Brown, C., Savidge, B. & Yanofsky, M.F. 1992b Molecular characterization of the Arabidopsis floral homeotic gene APETALA1. Nature, 360, 273277.
  • Mandel, M.A. & Yanofsky, M.F. 1995a A gene triggering flower formation in Arabidopsis. Nature, 377, 522524.
  • Mandel, M.A. & Yanofsky, M.F. 1995b The Arabidopsis AGL8 MADS box gene is expressed in inflorescence meristems and is negatively regulated by Apetala1. Plant Cell, 7, 17631771.
  • Mizukami, Y., Huang, H., Tudor, M., Hu, Y. & Ma, H. 1996 Functional domains of the floral regulator AGAMOUS: characterization of the DNA binding domain and analysis of dominant negative mutaions. Plant Cell, 8, 831845.
  • Mizukami, Y. & Ma, H. 1992 Ectopic expression of the floral homeotic gene AGAMOUS in transgenic Arabidopsis plants alters floral organ identity. Cell, 71, 119131.
  • Pnueli, L., Abu-Abeid, M., Zamir, D., Nacken, W., Schwarz-Sommer, Z. & Lifschitz, E. 1991 The MADS box gene family in tomato: temporal expression during floral development, conserved secondary structures and homology with homeotic genes from Antirrhinum and Arabidopsis. Plant J., 1, 255266.
  • Purugganan, M.D., Rounsley, S.D., Schmidt, R.J. & Yanofsky, M.F. 1995 Molecular evolution of flower development: diversification of the plant MADS-box regulatory gene family. Genetics, 140, 345356.
  • Riechmann, J.L., Krizek, B.A. & Meyerowitz, E.M. 1996 Dimerization specificity of Arabidopsis MADS domain homeotic proteins APETALA1, APETALA3, PISTILLATA and AGAMOUS. Proceedings Natl Acad. Sci. USA, 93, 47934798.
  • Riechmann, J.L. & Meyerowitz, E.M. 1997 MADS domain proteins in plant development. Biol. Chem., 378, 10791101.
  • Rounsley, S.D., Ditta, G.S. & Yanofsky, M.F. 1995 Diverse roles for MADS box genes in Arabidopsis development. Plant Cell, 7, 12591269.
  • Schmidt, R.J., Veit, B., Mandel, M.A., Mena, M., Hake, S. & Yanofsky, M.F. 1993 Identification and molecular characterization of ZAG1, the maize homolog of the Arabidopsis floral homeotic gene AGAMOUS. Plant Cell, 5, 729737.
  • Schwarz-Sommer, Z., Hue, I., Huijser, P., Flor, P.J., Hansen, R., Tetens, F., Lönnig, W.-E., Saedler, H. & Sommer, H. 1992 Characterization of the Antirrhinum floral homeotic MADS-box gene deficiens: evidence for DNA binding and autoregulation of its persistent expression throughout flower development. EMBO J., 11, 251263.
  • Sommer, H., Beltrán, J.-P., Huijser, P., Pape, H., Lönnig, W.-E., Saedler, H. & Schwarz-Sommer, Z. 1990 Deficiens, a homeotic gene involved in the control. of flower morphogenesis in Antirrhinum majus: the protein shows homology to transcription factors. EMBO J., 9, 605613.
  • Stewart, W.N. & Rothwell, G.W. 1993Paleobotany and the Evolution of Plants. Cambridge: Cambridge University Press.
  • Swofford, D.L. 1993PAUP Phylogenetic Analysis Using Parsimony, Version 3.1.1. Champaign: Illinois Natural History Survey.
  • Tandre, K., Albert, V.A., Sundås, A. & Engström, P. 1995 Conifer homologues to genes that control flower development in angiosperms. Plant Mol. Biol., 27, 6978.
  • Theißen, G., Kim, J.T. & Saedler, H. 1996 Classification and phylogeny of the MADS-box multigene family suggest defined roles of MADS-box gene subfamilies in the morphological evolution of eukaryotes. J. Mol. Evol., 43, 484516.
  • Tröbner, W., Ramirez, L., Motte, P., Hue, I., Huijser, P., Lönnig, W.-E., Saedler, H., Sommer, H. & Schwarz-Sommer, Z. 1992 GLOBOSA: a homeotic gene which interacts with DEFICIENS in the control. of Antirrhinum floral organogenesis. EMBO J., 11, 46934704.
  • Tsuchimoto, S., Van Der Krol. A.R. & Chua, N.-H. 1993 Ectopic expression of pMADS3 in transgenic petunia phenocopies the petunia blind mutant. Plant Cell, 5, 843853.
  • Yanofsky, M.F., Ma, H., Bowman, J.L., Drews, G.N., Feldmann, K.A. & Meyerowitz, E.M. 1990 The protein encoded by the Arabidopsis homeotic gene agamous resembles transcription factors. Nature, 346, 3539.
  • Zachgo, S., De Andrade Silva, E., Motte, P., Tröbner, W., Saedler, H. & Schwarz-Sommer, Z. 1995 Functional analysis of the Antirrhinum floral homeotic DEFICIENS gene in vivo and in vitro by using a temperature-sensitive mutant. Development, 121, 28612875.