Skip to main content
NCBI home page
Plant Physiology logoLink to Plant Physiology
. 2007 Feb;143(2):697–706. doi: 10.1104/pp.106.087957

Analysis of the DECREASED APICAL DOMINANCE Genes of Petunia in the Control of Axillary Branching1,[C],[OA]

Joanne L Simons 1, Carolyn A Napoli 1, Bart J Janssen 1, Kim M Plummer 1,2, Kimberley C Snowden 1,*
PMCID: PMC1803742  PMID: 17158589

Abstract

Control of branch development is a major determinant of architecture in plants. Branching in petunia (Petunia hybrida) is controlled by the DECREASED APICAL DOMINANCE (DAD) genes. Gene functions were investigated by plant grafting, morphology studies, double-mutant characterization, and gene expression analysis. Both dad1-1 and dad3 increased branching mutants can be reverted to a near-wild-type phenotype by grafting to a wild-type or a dad2 mutant root stock, indicating that both genes affect the production of a graft-transmissible substance that controls branching. Expression of the DAD1 gene in the stems of grafted plants, detected by quantitative reverse transcription-polymerase chain reaction correlates with the branching phenotype of the plants. The dad2-1 mutant cannot be reverted by grafting, indicating that this gene acts predominantly in the shoot of the plant. Double-mutant analysis indicates that the DAD2 gene acts in the same pathway as the DAD1 and DAD3 genes because the dad1-1dad2-1 and dad2-1dad3 double mutants are indistinguishable from the dad2-1 mutant. However, the dad1-1dad3 double mutant has an additive phenotype, with decreased height of the plants, delayed flowering, and reduced germination rates compared to the single mutants. This result, together with the observation that the dad1-1 and dad3 mutants cannot be reverted by grafting to each other, suggests that the DAD1 and DAD3 genes act in the same pathway, but not in a simple stepwise fashion.

Determination of aerial growth patterns and architecture is an important developmental process within a plant. Control of the type, position, and timing of shoot growth allows plants to develop in a dynamic manner that is defined genetically, but modulated environmentally (Bell, 1991). Patterns of plant branching are diverse and despite the importance of branching in plant development, and many years of study, the controls of lateral branching are still not fully understood (Napoli et al., 1999; Angenent et al., 2005; Beveridge, 2006).

One approach to studying branching is the induction of mutations in endogenous genes. A group of relatively nonpleiotropic mutants that cause increased lateral branching in plants has been identified in several species. These include the decreased apical dominance (dad) mutants of petunia (Petunia hybrida; Napoli, 1996; Napoli and Ruehle, 1996; Napoli et al., 1999), the ramosus (rms) mutants of pea (Pisum sativum; Beveridge, 2000; Morris et al., 2001; Rameau et al., 2002), and the more axillary growth (max) mutants of Arabidopsis (Arabidopsis thaliana; Stirnberg et al., 2002; Turnbull et al., 2002; Sorefan et al., 2003).

Three independent DAD loci controlling branching in petunia have been identified and mutations in any of these loci result in plants with an increase in basal axillary branches and a decrease in height (Napoli, 1996; Napoli and Ruehle, 1996; Snowden and Napoli, 2003). Recently, the DAD1 gene has been cloned and encodes a putative carotenoid cleavage dioxygenase (CCD). DAD1 is an ortholog of the MAX4 and RMS1 genes (Sorefan et al., 2003; Snowden et al., 2005) and belongs to a family of CCD genes of which MAX3 and RMS5 are also members (Booker et al., 2004; Johnson et al., 2006). The increased branching phenotype of the dad2-1 mutant is very similar to that of dad1-1, especially during long-day conditions. Although the dad3 mutant shows increased branching and decreased plant height, the phenotype is not as extreme as in either dad1-1 or dad2-1 (Snowden and Napoli, 2003). Cloning of DAD3 is under way (this laboratory), but final confirmation of the identity of the gene is ongoing.

To understand the role of the DAD genes in the control of plant branching, their relationship to each other needs to be clarified. Genetic analyses have shown that the three DAD loci are independent (Napoli and Ruehle, 1996), but it is unknown whether they act in the same biochemical or signaling pathways and, if so, in what order. Although the components of the DAD branching pathway are conserved across species, there are some significant variations in the way the genes appear to function in mutants of different plants. These include differences in the patterns of gene expression (Sorefan et al., 2003; Bainbridge et al., 2005; Foo et al., 2005; Snowden et al., 2005) and some differences in response to grafting (for review, see Angenent et al., 2005).

In previous experiments, dad1 mutant scions grafted over wild-type root stocks reverted the increased branching phenotype to that of a wild-type plant (Napoli, 1996). This result indicated that the DAD1 gene product was involved in the control of a graft-transmissible branching signal. A similar graft-revertible phenotype was also demonstrated for rms1 (Beveridge et al., 1997) and max4 (Sorefan et al., 2003). This article investigates the effects of grafting on the lateral branching phenotype of dad2-1 and dad3 mutant petunias, as well as their interactions with the dad1-1 mutant. The phenotype of double mutants between the DAD genes is also described and the results used to place the DAD genes in a model for branching control.

RESULTS

Grafting at Nodes Above the Cotyledons Show That Movement of a Branching Signal Occurs Acropetally

Napoli (1996) reported that dad1-1 mutant scions could be reverted to a near-wild-type phenotype by hypocotyl grafting of a wild-type interstock between a dad1-1 mutant scion and a dad1-1 mutant stock either in the hypocotyl or epicotyl region of mutant seedlings. To test whether this reversion could be effected in both an acropetal and basipetal manner, a grafting experiment where wild-type interstocks were grafted several nodes above the cotyledons in dad1-1 mutant seedlings was performed. This experiment required that axenic dad1-1 seedlings were grown for at least 3 weeks on one-half-strength Murashige and Skoog agar, which helped elongate the stem to where longer internode distances could be observed and severed to accept the wild-type hypocotyl interstock (it was not possible to obtain sufficient wild-type epicotyl material to form a successful interstock). Whereas the dad1-1 mutant nodes above the graft reverted to a wild-type phenotype, the dad1-1 mutant nodes below the graft did not revert, but instead produced primary, secondary, and tertiary lateral branches (Fig. 1).

Figure 1.

Figure 1.

Open in a new tab

Interstock grafting of dad1-1. dad1-1 grafted mutant with a wild-type interstock inserted several nodes acropetally to the cotyledons. Brackets indicate branches emerging from dad1-1 nodes below the interstock (Y) and not emerging from dad1-1 tissue above the graft (X), which has reverted to a wild-type phenotype. [See online article for color version of this figure.]

As shown in Figure 1, the overall morphology of the grafted plants was very different from either a dad1-1 mutant or a wild-type plant (e.g. see Fig. 3A) in that there is a combination of the two phenotypes with the wild-type flower shoot overgrowing the highly branched and compact lower portion of the grafted plant. The wild-type appearing shoot flowered sooner than the highly branched part of the plant as expected for a wild-type phenotype. The wild-type shoot could be traced to the mutant nodes above the internodal hypocotyl insertion, whereas the highly branched and compact nodes were below the interstock insertion. This experiment demonstrates that a graft-transmissible substance controlling branching can also be produced from a small interstock or produced in the roots and then acted upon by the interstock, from whence it moves in an acropetal, but not a basipetal, manner.

Figure 3.

Figure 3.

Open in a new tab

Response of dad1-1 mutant scions to grafting over wild-type and dad2-1 mutant stocks. A, Representative 3-month-old plants of those used in the experiment. B, Height of the grafted plants (cm). C, Number of branches present on grafted plants. D, Relative expression level of DAD1 gene in basal portions of stems (close to the graft union; black bars) and roots (shaded bars) of grafted plants. Relative gene expression was normalized against expression of the three internal reference genes in each tissue sample. Data shown are ±sem (n ≥ 8 for all samples). Data are representative of two independent datasets. [See online article for color version of this figure.]

The dad3 Mutant Can Be Reverted by Grafting to Wild-Type Root Stocks, But the dad2 Mutant Cannot

Given that the dad1-1 mutant phenotype could be reverted by grafting over wild-type root stocks (Napoli, 1996) and by insertion of wild-type interstocks, the effect of grafting on the dad3 mutant phenotype was investigated. The grafting technique alone did not affect the phenotype of either wild-type self grafts or dad3 self-grafted plants. However, when grafted onto wild-type stocks, dad3 mutant scions were converted to a branching phenotype that was not significantly different from wild-type self grafts (see Fig. 2A for branch number comparisons, Fig. 2C for visual phenotype comparisons, and Table I for statistical comparisons). This result suggests that, like DAD1, the DAD3 gene controls the production of a graft-transmissible factor that can control branching. Visual observations suggested a small effect in the reciprocal graft where wild-type scions were grafted onto dad3 stocks (Fig. 2C), warranting further investigation. A second experiment was carried out using a greater number of wild-type over dad3-grafted plants (Fig. 2B). In this experiment, grafting of wild-type scions over dad3 stocks caused a statistically significant increase in the number of lateral branches over wild-type, self-grafted controls (see Table I).

Figure 2.

Figure 2.

Open in a new tab

Effects of grafting on dad mutants dad1 (dad1-1), dad2 (dad2-1), dad3 (dad3), and wild type (wt). The notation indicates scion/stock (i.e. dad3/wt is a plant with a dad3 scion grafted over a wild-type stock). A, Number of branches present on grafted plants involving dad3 mutants (dad3, n = 11; dad3/dad3, n = 5; dad3/wt, n = 12; wt/dad3, n = 8; wt, n = 6; wt/wt, n = 9). B, Number of branches present on plants in the second dad3 grafting experiment (dad3/dad2, n = 8; dad3/dad3, n = 9; wt/dad3, n = 25; wt/wt, n = 8). C, Representative plants of those used in the wild-type and dad3 grafting experiment shown in A. D, Number of branches present on plants in the dad2-1 grafting experiment (dad2/dad1, n = 6; dad2/dad3, n = 3; dad2/wt, n = 10; dad2, n = 2; wt/dad2, n = 8; wt, n = 2). E, Reciprocal grafting experiment with dad1-1 and dad3. Number of branches present on the grafted plants (dad1/dad1, n = 6; dad3/dad3, n = 7; dad1/dad3, n = 9; dad3/dad1, n = 12). Data shown are means ± se of means (sem). [See online article for color version of this figure.]

Table I.

Statistical analysis of dad3 and dad2-1 grafting results

Mean number of branches and statistical analysis for the four independent grafting experiments in Figure 2. Mean separation tests were performed using Fisher's protected lsd test at the 5% level of significance and comparisons were only within experiments. Mean values within an experiment that have the same letter (letters a–d) are not significantly different at the P = 0.05 level.

Graft Type Branches (F < 0.001) lsd 5% = 2.19 Graft Type Branches (F < 0.001) lsd 5% = 2.67
dad3 10.18a dad2/dad1 10.00a
dad3/dad3 10.80a dad2/dad3 10.33a
dad3/wt 5.58b dad2/wt 11.20a
wt/dad3 6.00b dad2 11.00a
wt 3.83b wt/dad2 1.62b
wt/wt 4.89b wt 3.00b
Graft Type Branches (F < 0.001) lsd 5% = 5.58 Graft Type Branches (F < 0.001) lsd 5% = 4.16
dad3/dad2 3.25b dad1/dad1 18.86a
dad3/dad3 6.44a dad3/dad3 13.29b
wt/dad3 1.96c dad1/dad3 17.67a
wt/wt 0.62d dad3/dad1 13.83b
Open in a new tab

A similar grafting experiment was carried out to investigate how grafting affected dad2-1 mutants. In contrast to dad1-1 (Napoli, 1996) and dad3 (Fig. 2, A–C), the phenotype of the dad2-1 scion was not converted to a wild-type branching pattern by grafting over a wild-type stock (Fig. 2D; Table I). When dad2-1 was grafted over dad1-1 or dad3 stocks, the phenotype of the scion was also not reverted and was not significantly different to ungrafted dad2-1 plants. These data indicate that the function of the DAD2 gene is likely to be limited to the shoot.

dad1-1 and dad3 Mutants Cannot Revert Each Other by Grafting, But dad2 Can Revert Both dad1 and dad3

Because the branching phenotype of both dad1-1 and dad3 can be reverted to wild type by grafting, an experiment was carried out to determine whether either of these mutants could revert the other by grafting. When dad1-1 or dad3 scions were grafted over a stock of the other mutant, the branching phenotype did not revert to wild type (Fig. 2E; Table I). The grafted plants displayed the phenotype of the scion and were unaffected by the genotype of the stock. This result indicates that it is likely that the action of both the DAD1 and DAD3 genes is required in the same tissue and that these genes control the production of the same graft-transmissible signal.

The interactions between the dad1-1 and dad3 mutants (revertible by grafting over wild-type stocks) and dad2-1 (not revertible by grafting) were also investigated. In separate experiments, dad1-1 and dad3 scions were grafted over a dad2-1 stock. The dad3 phenotype of the scion was only partially converted to wild type; the number of branches measured was intermediate between dad3 and wild-type plants (Fig. 2B; Table I). By contrast, the dad1-1 scions were reverted to a phenotype of near-wild-type appearance by a dad2-1 stock (Fig. 3; Table II), similar to the reversion observed by grafting dad1-1 scions over wild-type stocks. The reversion of the mutant scions was observed for both plant height (Fig. 3B) and branching (Fig. 3C). In the reciprocal grafts, when dad2-1 scions were grafted over dad1-1 root stocks, no reversion of the increased branching phenotype was observed (Figs. 2D and 3C). This experiment was repeated with similar results (Fig. 3A; data not shown). When wild-type scions were grafted over dad2-1 root stocks, the branching was not significantly different to that of wild-type plants (Fig. 2D; Table II). These experiments reiterate our conclusion that the action of the DAD2 gene in the suppression of shoot branching is likely to be restricted to the shoot.

Table II.

Means and statistical analysis for branch and plant height data from dad1-1/dad2-1 grafting experiment (Fig. 3)

Mean separation tests were performed using Fisher's protected lsd test at the 5% level of significance. Values within a column that have the same letter (a–c) are not significantly different at the P = 0.05 level.

Graft Type Plant Height (F < 0.001) lsd 5% = 5.58 Branches (F < 0.001) lsd 5% = 4.16
cm
wt/wt 31.85a 7.90c
dad1/dad1 20.90b 37.80a
dad1/dad2 31.67a 7.22c
dad1/wt 35.65a 9.00c
dad2/dad1 17.86b 31.00b
Open in a new tab

Expression of the DAD1 Gene in Stem Correlates with the Phenotype of the Scion

Quantitative reverse transcription (qRT)-PCR was used to measure DAD1 gene expression in the stems and roots of the grafted plants from the dad1-1, dad2-1 grafting experiment. Stem tissue samples were taken from above the graft union and so were of the same genotype as the scion, whereas root samples were of the root stock genotype. Expression of DAD1 in root tissue did vary between wild-type and dad1-1 self-grafted plants (Fig. 3D); however, this variation did not correlate with any observed phenotypic differences or with the genotype of the root stock or scion. Expression of DAD1 in shoot tissue, however, generally correlated with the branching phenotype of the plant. The expression of DAD1 was up-regulated in dad1-1 mutant scions in self-grafted plants and in dad2-1 mutant scions grafted to dad1-1 stocks when compared with wild type, but not in dad1-1 scions that were grafted over wild-type or dad2-1 stocks (Fig. 3D). The qRT-PCR levels in the reverting grafted plants showed a typically wild-type DAD1 gene expression phenotype in tissue that was genetically dad1-1 mutant.

Analysis of dad Double-Mutant Phenotypes Suggest dad1-1dad3 Is Additive

The physical characteristics of the single and double dad mutants were quantified and compared against those of wild-type petunias to study the interactions of these genes. Plants were grown until 3 months old in controlled conditions and then their morphology was characterized. The main characters that showed variation between the dad mutants were plant height, number of basal branches, and number of flowers (sympodial shoots) produced after the termination of the main shoot axis, which was used as an estimate of flowering time (Fig. 4; Table III).

Figure 4.

Figure 4.

Open in a new tab

Phenotypes of the single and double dad mutants. A, Representative plants of those used in the study. B, Number of primary branches (black bars) and secondary branches arising from primary branches (shaded bars) present on single- and double-mutant plants. C, Plant height (cm) of single and double dad mutants. D, Number of flowering nodes produced after 3 months growth. E, Flower buds present on dad1-1 mutants after 3 months growth in a second experiment. F, Absence of flowers on dad1-1dad3 double mutants grown together with the dad1 plants shown in E. Data shown are means ± sem (n ≥ 8 for all samples). Data are representative of two independent datasets.

Table III.

Means and statistical analysis for branch and plant height data from the dad double-mutant phenotype study (Fig. 4, A–D)

Mean separation tests were performed using Fisher's protected lsd test at the 5% level of significance. Values within a column that have the same letter (a–e) are not significantly different at the P = 0.05 level.

Genotype Primary Branches (F < 0.001) lsd 5% = 0.40 Secondary Branches (F < 0.001) lsd 5% = 7.51 Plant Height (F < 0.001) lsd 5% = 4.16 Flowers (F < 0.001) lsd 5% = 1.49
cm
Wild type 0.92d 0.0c 50.93a 10.87a
dad1-1 3.33a,b 14.3b 22.80c 5.20c
dad2-1 3.25b 20.5a,b 23.50c 4.50c,d
dad3 2.54c 0.0c 38.22b 8.67b
dad1-1dad2-1 3.70a 23.5a 22.50c 4.00c,d
dad1-1dad3 3.59a,b 18.4a,b 13.00d 0.44e
dad2-1dad3 3.42a,b 26.2a 23.28c 3.33d
Open in a new tab

In this study, the three single dad mutants displayed a decrease in plant height, an increased amount of basal branching, and a delay in the onset of flowering when compared to wild-type plants (Fig. 4), as has been observed previously (Napoli, 1996; Napoli and Ruehle, 1996; Snowden and Napoli, 2003). The dad3 mutant did not display as severe a phenotype as the other single dad mutants, but dad1-1 and dad2-1 were very similar in morphology (Fig. 4A). This raises the possibility that the dad3 mutant phenotype may not be due to a complete loss of function of DAD3. However, from previous work, it is probable that the dad1-1 mutant is a complete loss-of-function mutant (Snowden et al., 2005). The phenotype of the dad1-1dad3 mutant was more severe than either single mutant (dad1-1 or dad3). The double dad1-1dad3 mutant was reduced in height (Fig. 4, A and C) and was delayed in flowering (Fig. 4, D–F), although the double mutant was not additive with regard to the number of basal branches (Fig. 4B; see also Table III). In contrast to dad1-1dad3, the dad2-1dad3 double mutant was not additive for any of the characters measured and was indistinguishable from the dad2-1 mutant, indicating that the DAD2 and DAD3 genes are likely to be acting in the same pathway. The dad1-1dad2-1 double mutant was not able to be differentiated by morphology from either the dad1-1 or dad2-1 single mutants with respect to the characters measured (i.e. the phenotype was not additive), indicating that DAD1 and DAD2 are also likely to be acting in the same pathway. Similar results were obtained when this experiment was replicated (data not shown).

Fertility of dad1-1 and dad3 and Germination of dad1-1dad3 Double-Mutant Seed

Isolation and identification of the dad1-1dad3 double mutant was at first difficult, suggesting either linkage between the two genes or a problem with fertility and/or germination of this double mutant. The dad1-1dad3 double mutant was eventually successfully isolated (at expected frequencies for unlinked loci) when seeds from self pollination of a plant with the genotype dad1-1/dad1-1 DAD3/dad3 were sown on one-half-strength Murashige and Skoog agar plates. As an estimate of fertility, self pollinations were carried out on dad1-1, dad3, dad1-1dad3, and wild-type plants, and the seed yield per capsule was measured. Whereas the number of seeds produced from each cross was considerably lower than that of wild type for both dad1-1 and dad3 plants, there was no additive effect observed in the dad1-1dad3 plants, with similar amounts of seed produced compared with the crosses on the single dad mutant plants (Table IV).

Table IV.

Seed germination and seed yield of dad1-1, dad3, and dad1-1dad3 mutants compared with wild-type petunia

Self pollinations were performed and the number of seed produced calculated for each genotype (mean seed per capsule; n ≥ 8). Germination was tested in three conditions; surface-sterilized seed on one-half-strength Murashige and Skoog agar, surface-sterilized seed sown on soil, and nonsterilized seed sown on soil. Fifty seeds were sown under each condition for each of three replicates of the experiment. Results are presented as percentage of seed germinating. Mean separation tests were performed using Fisher's protected lsd test at the 5% level of significance. Values within either the germination or fecundity (seed yield) data that have the same letter (a–e) are not significantly different at the P = 0.05 level.

Genotype Germination and Treatment (F = 0.002) lsd 5% = 11.8
Fecundity (Mean Seed per Capsule; F = 0.006) lsd 5% = 57
Sterilized Seed, Murashige and Skoog Sterilized Seed, Soil Nonsterilized Seed, Soil
% % %
Wild type 65.3a,b 73.3a 61.3b 250a
dad1-1 55.3b,c 44.7c,d 44.0c,d 112b
dad3 74.7a 54.0b,c 40.7d 127b
dad1-1dad3 22.7e 22.0e 26.7e 106b
Open in a new tab

The germination of the dad1-1dad3 double mutant was then investigated. The rates of seed germination of dad1-1, dad3, and dad1-1dad3 mutants were compared with that of wild-type seed on various growth media (Table IV). When sown on soil, the germination rate was considerably lower in both dad1-1 and dad3 than that observed in wild-type seed. When sown in tissue culture on one-half-strength Murashige and Skoog plates, however, the germination rate of dad3 was very similar to, if not greater than, that of wild type, and the germination rate of dad1-1 seed was much closer to that of wild type than when sown on soil. In all cases, the germination of dad1-1dad3 seed was significantly lower than that of any other genotype, reiterating the additive effect of this double mutant on the dad phenotype. Surface sterilization of seed did not have a large effect on the germination rate of any seed type.

DISCUSSION

DAD1 and DAD3 Act at the Same Step to Produce a Branching Signal and Act in the Same Pathway as DAD2 to Control Branching

The data presented here demonstrate that both DAD1 and DAD3 regulate or modify the production of the same graft-transmissible substance that acts as a signal from the roots to regulate branching in the shoot. The DAD1 gene has been cloned and is a member of the CCD gene family (Snowden et al., 2005). Members of the CCD gene family have been shown to have activity on polyene compounds in general and in particular many of the genes catalyze cleavage of carotenoids (or apocarotenoids) to produce aldehydes and/or ketones (Schwartz et al., 1997, 2004; Booker et al., 2004; Schmidt et al., 2006). The AtCCD8 (also known as MAX4) gene product has been shown to cleave a carotenoid-derived molecule in vitro (Schwartz et al., 2004), suggesting that the unidentified branching signals may belong to this class of compounds. This signal appears to move only in an acropetal direction in petunia (Fig. 1). Acropetal movement of branching signals has also been demonstrated in pea and Arabidopsis using plants with a single graft, where two shoot systems were allowed to develop (Foo et al., 2001; Turnbull et al., 2002). Together, these results indicate that the movement of the branching signal is limited to the acropetal direction.

In the case of the DAD2 gene, our double-mutant analysis indicates that it acts in the same pathway as both DAD1 and DAD3. This result, together with the data from grafting experiments, suggests that DAD2 acts in the reception or transduction of the branching signal. This is consistent with its activity in the shoot and its placement downstream of DAD1 and DAD3. The identity of the DAD2 gene has not yet been determined.

Given that our results imply that DAD2 acts in the same pathway as DAD1 and that DAD2 acts in the same pathway as DAD3, an obvious prediction is that DAD1 and DAD3 act together in the same pathway. However, characterization of plant height, germination, and flowering for the dad1-1dad3 double mutant showed a strong additive phenotype compared with either the dad1-1 or dad3 single mutants. Whereas with respect to dad3 this additive effect could indicate that the dad3 mutation is only a partial loss of function, this is not the case for dad1-1 (Snowden et al., 2005). Whereas the dad1-1dad3 double mutant was not additive for the branching phenotype, the upper limit of branching for these plants may have already been reached in the single dad1-1 mutants where most, if not all, of the basal nodes on these plants already form branches (Snowden and Napoli, 2003). These results suggest the possibility that DAD1 and DAD3 act in separate pathways to produce the effects on plant height, flowering, and seed germination, although the possibility remains that they are not additive for the branching trait. If DAD1 and DAD3 act in separate pathways, then dad1 and dad3 root stocks should be able to revert scions of the other mutant to wild type in reciprocal grafting experiments. However, in the reciprocal grafts between dad1-1 and dad3, neither graft combination caused a reversion of the scion phenotype to wild type (Fig. 2E), indicating that it is unlikely that the DAD1 and DAD3 genes act in completely separate pathways.

The combined results of reciprocal grafting experiments and double-mutant analysis suggest that DAD1 and DAD3 interact in a relatively complex manner. One explanation for these results is that DAD1 and DAD3 act together at the same step in combination to process the branching signal. However, other models are possible, for example, where DAD1 and DAD3 act at sequential steps with a nonmobile intermediate requiring colocalization of DAD1 and DAD3 (or close localization at least in the same tissue), or where DAD1 or DAD3 regulate the activity of each other. We know that the expression of the DAD1 gene is altered in the stem of the dad3 mutant compared with wild type (Snowden et al., 2005) so there is some interaction between the DAD1 and DAD3 genes at this level. Given that DAD1 has been shown to be a member of the CCD gene family, we propose that DAD1 and DAD3 act in concert to process a compound that is likely to be carotenoid derived in the roots and stem of the plant. This modified compound then moves in an acropetal direction (through grafts) where the signal is received and transduced through a pathway involving DAD2.

Interactions of the dad3 Mutant Suggest Potential Candidates for the Gene's Identity

As discussed above, the dad3 mutant is graft revertible. This suggests similarities to the Arabidopsis branching mutants, max1 and max3, which are also graft revertible (Turnbull et al., 2002). When max4 (the dad1 ortholog) scions are grafted over max1 root stocks, the branching phenotype is reverted to wild type (Booker et al., 2004). This indicates that max1 acts downstream of max4 in the branching pathway. This phenomenon is not seen in the grafting of dad1-1 scions over dad3 root stocks (Fig. 2E), indicating that DAD3 and MAX1 are unlikely to be orthologous.

The more likely possibility is that the MAX3 gene is orthologous to DAD3. We predict that DAD3 is acting at the same step in the pathway as DAD1 and it is possible that DAD3 is a related enzyme. In Arabidopsis, both the MAX3 and MAX4 genes have been shown to encode CCD enzymes, AtCCD7 and AtCCD8, respectively (Sorefan et al., 2003; Booker et al., 2004; Schwartz et al., 2004), and the pea orthologs of these genes are RMS5 and RMS1 (Sorefan et al., 2003; Johnson et al., 2006). Schwartz et al. (2004) showed that AtCCD7 and AtCCD8 can act sequentially or in concert to cleave carotenoid substrates and produce different products. If DAD1 and DAD3 are both CCD enzymes and are acting on the same substrate in a slightly different manner, this could account for some of the differences observed between the mutants in the grafting experiments. For example, a dad3 root stock is capable of causing some changes in branching in wild-type scions under some growth conditions (Fig. 2B), which has not been observed for dad1-1 root stocks. A similar (small, branch-inducing) effect of a mutant root stock on a wild-type scion has been reported for pea (rms5; Morris et al., 2001), but not for Arabidopsis. There are some significant differences in the phenotypes of plants carrying max3 or dad3 mutations. Whereas the dad1-1dad3 double mutant is additive in several aspects of its phenotype, it has been reported that the max3max4 mutant is not additive in phenotype (Auldridge et al., 2006). This indicates that either DAD3 is not orthologous to MAX3 or that they act in different ways in different species. However, the equivalent pea double mutant (rms1rms5) does show a slight additive phenotype (Morris et al., 2001), which would be consistent with the additive phenotype of dad1-1dad3. Furthermore, in reciprocal grafting experiments, dad1-1 and dad3 act similarly to max3 and max4 in Arabidopsis and also to rms1 and rms5 in pea, where in each case these mutants cannot revert the other by grafting (Morris et al., 2001; Booker et al., 2005). Together, these comparisons indicate that it is likely that DAD3 is orthologous to RMS5 and MAX3, although the phenotypes of the corresponding mutants in petunia and pea appear to be more similar to each other than to Arabidopsis.

A Model for Interaction of the DAD Genes to Control Branching

We propose a model for the control of axillary meristem outgrowth (Fig. 5) based on the data presented in this article and previously published data (Napoli, 1996; Snowden et al., 2005). This model allows us to place genes and putative signal molecules in context and suggests further areas for study. From the above analysis, we have placed DAD1 and DAD3 at the same step in the model where they are involved in the conversion of a root-derived substrate A (possibly a carotenoid or an apocarotenoid) into a product B (Angenent et al., 2005; Snowden et al., 2005). This reaction occurs in the roots and can also occur in the stem, given that the DAD1 gene is expressed in both of these tissues.

Figure 5.

Figure 5.

Open in a new tab

Model of interaction of DAD gene products in control of branching in petunia. A positive branching signal A is modified by DAD1 and DAD3 into B, which may be a negative signal to control branching. Both of these compounds are graft transmissible and may be formed in the roots and lower stem of the plant. The signals move up into the shoot of the plant where DAD2 is involved with their reception or signal transduction to control outgrowth of axillary buds into axillary branches. The black arrow indicates conversion of the positive signal compound to an inhibition signal; dotted arrows indicate passage of the branching signal through the plant; dashed arrows indicate feedback signals.

The DAD2 gene product acts in the shoot and can be placed downstream of DAD1 and DAD3, possibly in a signal reception or signal transduction pathway. There is one Arabidopsis branching mutant (ore9, also known as max2, cloned and shown to be an F-box protein; Woo et al., 2001) and two pea branching mutants (rms3 and rms4; Beveridge et al., 1996), where rms4 is orthologous to max2 (Johnson et al., 2006) that cannot be reverted by grafting, similar to dad2. Preliminary data suggest that dad2 is not an ortholog of max2 (data not shown). It is tempting to postulate that these genes make up the signal reception and transduction pathway that leads to meristem outgrowth in response to the graft-transmissible signal.

Graft analysis indicates that the branching signal is produced in both root and stem tissue. The signal can cross graft unions and moves acropetally and not basipetally. Two possibilities exist for the type of signal produced. First, the product of the DAD1 and DAD3 gene activity B (or a derivative of B) could be an inhibitor of branching; loss of the inhibitor in the mutants would then result in branch production. The second possibility is that the substrate of DAD1 and DAD3 gene activity A (or a derivative of A) could be a promoter of branching. In this case, loss of gene activity in the mutants could result in accumulation of A and promotion of branching. Napoli (1996) noted that when dad1-1 mutant roots were allowed to form above the graft, union of a graft between a wild-type root stock and a dad1-1 scion, the dad1-1 scions were not reverted to a wild-type phenotype and were highly branched; if these dad1-1 mutant roots were removed, subsequent nodes reverted to wild type and did not branch. This indicates that the substrate of DAD1 and DAD3 or a derivative (A in Fig. 5) is a promoter of branching. Moreover, because in those grafts the wild-type roots are still producing B, then even if B is an inhibitor, A is able to induce branching in the presence of B. The results presented here give further support to the presence of a branch-promoting compound because a dad3 root stock is capable of causing an increase in branching in a wild-type scion under some circumstances (Fig. 2B). In the case of a graft between a dad1-1 root stock and a wild-type scion (Napoli, 1996), no promotion of branching has been observed. This result indicates that any branch-promoting signal (A) must be able to be acted upon by the wild-type DAD1 gene product in the stem of the plant. This prediction is supported by the fact that the DAD1 gene is expressed and under regulatory control in the stem of the plant (Fig. 3D; Snowden et al., 2005).

The two possible activities for the signal molecules are not mutually exclusive (i.e. A could be a promoter of branching and B could be an inhibitor of branching). Interestingly, the balance between A and B may be important in altering branch structure and may vary between species. For example, in petunia, mutant roots produce a promoter of branching that is dominant over any inhibitor present; in Arabidopsis and pea, published data suggest that, if a promoter is present, it does not appear to be dominant.

Studies of the expression of DAD1 in the dad mutant plants (Snowden et al., 2005) have indicated that there is probably a feedback mechanism involved in the DAD pathway because expression of DAD1 is up-regulated in the stems (but not the roots) of dad1-1, dad2-1, and dad3 mutant plants compared to wild-type plants. This phenomenon is also observed in pea (Foo et al., 2005), but not in Arabidopsis (Bainbridge et al., 2005). In this current study, DAD1 gene expression in the roots of the plant did not appear to correlate in any way with either the phenotype or genotype of the scion or stock. In contrast, expression of the DAD1 gene in the shoot of the plant correlates with the phenotype of the scion of grafted plants, rather than the genotype, suggesting that the levels of DAD1 gene expression in the shoot are important in the control of branching and respond to the branching phenotype (Fig. 3D).

Because a small interstock of wild-type tissue can revert the branching phenotype of a dad1-1 mutant scion (Fig. 1; Napoli, 1996), this raises interesting questions as to the mode of transport of any signal molecules. If the only signal molecules are products of this reaction, hydrophilic transport pathways would suffice. However, it is possible that the substrates of DAD1 and DAD3 are hydrophobic (e.g. carotenoids are predominantly hydrophobic). If, indeed, the substrates or a precursor of the DAD1 and DAD3 enzymes are a positive branching signal, then that implies a transport pathway for hydrophobic compounds. In addition, AtCCD7 and AtCCD8 have been reported to be plastid localized (Auldridge et al., 2006), suggesting that the DAD1 and DAD3 substrates must be able to move efficiently into the plastid from the transport pathway.

It is possible that branching is not the sole process controlled by this pathway. The max2 increased branching mutant, for which there is currently no orthologous mutant in petunia, was originally isolated from Arabidopsis by its delayed senescence phenotype (Woo et al., 2001) and the dad1-1 mutant has been shown to have delayed leaf senescence as well as alterations in leaf color, flowering time, root mass, plant height, and seed germination and yield (Napoli, 1996; Snowden et al., 2005; Table IV). The AtCCD7 gene products have been shown to be active on several substrates and there is preliminary evidence that the AtCCD8 gene product is involved in the catabolism of carotenoids (Booker et al., 2004; Auldridge et al., 2006). It is possible that the activity of the CCD enzymes in the branching pathway also results in the production of other signal molecules that regulate other aspects of plant growth and senescence. In addition, it is possible that mutations in these CCD enzymes could lead to altered flux in polyene metabolism pathways, leading to alteration of other compounds with biological activity. From these observations, it is likely that the model presented here is a simplification of what may prove to be a complex regulatory network.

MATERIALS AND METHODS

Petunia Genetic Stocks and Growth Conditions

The single dad mutants were induced with ethyl methanesulfonate in V26, an inbred genetic stock of petunia (Petunia hybrida; Napoli and Ruehle, 1996). Double mutants were generated by crossing the single mutants and self pollinating the F1 progeny, and then confirmed by backcrosses to the single mutants. The growth conditions of the plants were as previously described in Snowden et al. (2005). Numbers of leaf nodes and branches were defined and recorded as in Snowden and Napoli (2003). Characters measured included number and length of primary, secondary, and tertiary axillary branches, plant height, height on the stem to the first flower produced, and the number of flowering nodes after the 3-month growth period, which was used as an estimate of flowering time. Data presented in graphs are means ± se (sem). ANOVAs were performed for statistical analyses using the GenStat statistical software package (8th ed.). The ANOVA model assumes that the error terms are independent and normally distributed with zero mean and constant variance. Appropriate transformations were used to ensure that model assumptions were met where necessary. Mean separation tests were performed using Fisher's protected least significant difference (lsd) test at the 5% level of significance.

Hypocotyl Grafting of Seedlings

All graft procedures employed surface sterilized seeds to produce axenic seedlings grown on tissue culture agar. All manipulations were carried out in a sterile transfer hood as described in Napoli (1996), with the exception that cotyledons were not removed from seedlings during graft alignment. Hypocotyl grafting was used as the method of choice to produce grafted plants. The seedlings that provided the scion and stock material were severed in the hypocotyl region and then the blunt ends of the appropriate combinations were united. Any roots forming above the graft union were removed as the graft healed. Grafted seedlings were transferred to a greenhouse environment between 1 and 2 weeks after grafting. The phenotypes of the grafted plants were scored 3 months after grafting and this was also the time when tissue was harvested for qRT-PCR analysis. Plant phenotypes are affected by variations in light and temperature, so direct comparisons between plant types are only made within single experiments.

Assessments of Fertility and Germination

Controlled self pollinations were performed on wild-type, dad1-1, dad3, and dad1-1dad3 petunias. Capsules were allowed to mature and were collected approximately 1 month after pollination to determine seed yield.

Three growth conditions were examined; seeds were either surface sterilized and grown on one-half-strength Murashige and Skoog agar (Murashige and Skoog, 1962), surface sterilized and sown onto soil, or not sterilized and sown onto soil. Seeds sown onto soil in the greenhouse were germinated in humidity chambers. Three replicates of 50 seeds of each genotype were sown on each type of growth medium.

Expression of DAD1 in Grafted Petunia Plants by qRT-PCR

Total RNA was isolated from tissues using the RNeasy plant mini kit (Qiagen) and treated with DNAse (using the Ambion DNA-free kit). RT reactions were carried out in triplicate as described in Snowden et al. (2005). qRT-PCR with Sybr Green detection was carried out for DAD1 and the internal control genes Actin, EF-1α, and Histone 4 using the primers and methods also described in Snowden et al. (2005). The primers for the DAD1 gene anneal to a portion of the second exon, downstream of the position of the insertion located in the dad1-1 allele. In previous work, expression of the mutant dad1-1 allele is readily detected by these primers (Snowden et al., 2005). Relative expression was calculated using the comparative cycle threshold method (Pfaffl, 2001) with normalization of data to the geometric average of the internal control genes (Vandesompele et al., 2002).

Sequence data from this article can be found in the GenBank/EMBL data libraries under accession number AY743219.

Acknowledgments

We would like to acknowledge Kerry Templeton for his assistance with characterization of the double mutants, Padmaja Ramankutty for help with statistical analyses, as well as Revel Drummond and Susan Ledger for helpful discussions.

1

This work was supported by the Foundation for Research, Science, and Technology, New Zealand (contract no. C10X0404).

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.plantphysiol.org) is: Kimberley C. Snowden (ksnowden@hortresearch.co.nz).

[C]

Some figures in this article are displayed in color online but in black and white in the print edition.

[OA]

Open Access articles can be viewed online without a subscription.

References

  1. Angenent GC, Stuurman J, Snowden KC, Koes R (2005) Use of Petunia to unravel plant meristem functioning. Trends Plant Sci 10 243–250 [DOI] [PubMed] [Google Scholar]
  2. Auldridge ME, Block A, Vogel JT, Dabney-Smith C, Mila I, Bouzayen M, Magallanes-Lundback M, DellaPenna D, McCarty DR, Klee HJ (2006) Characterisation of three members of the Arabidopsis carotenoid cleavage dioxygenase family demonstrates the divergent roles of this multifunctional enzyme family. Plant J 45 982–993 [DOI] [PubMed] [Google Scholar]
  3. Bainbridge K, Sorefan K, Ward S, Leyser O (2005) Hormonally controlled expression of the Arabidopsis MAX4 shoot branching regulatory gene. Plant J 44 569–580 [DOI] [PubMed] [Google Scholar]
  4. Bell AD (1991) Plant Form: An Illustrated Guide to Flowering Plant Morphology. Oxford University Press, Oxford
  5. Beveridge CA (2000) Long distance signalling and a mutational analysis of branching in pea. Plant Growth Regul 32 193–203 [Google Scholar]
  6. Beveridge CA (2006) Axillary bud outgrowth: sending a message. Curr Opin Plant Biol 9 35–40 [DOI] [PubMed] [Google Scholar]
  7. Beveridge CA, Ross JJ, Murfet IC (1996) Branching in pea—action of genes Rms3 and Rms4. Plant Physiol 110 859–865 [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Beveridge CA, Symons GM, Murfet IC, Ross JJ, Rameau C (1997) The rms1 mutant of pea has elevated indole-3-acetic acid levels and reduced root-sap zeatin riboside content but increased branching controlled by graft-transmissible signal(s). Plant Physiol 115 1251–1258 [Google Scholar]
  9. Booker J, Auldridge M, Wills S, McCarty D, Klee H, Leyser O (2004) MAX3/CCD7 is a carotenoid cleavage dioxygenase required for the synthesis of a novel plant signaling molecule. Curr Biol 14 1232–1238 [DOI] [PubMed] [Google Scholar]
  10. Booker J, Sieberer T, Wright W, Williamson L, Willett B, Stirnberg P, Turnbull C, Srinivasan M, Goddard P, Leyser O (2005) MAX1 encodes a cytochrome P450 family member that acts downstream of MAX3/4 to produce a carotenoid-derived branch-inhibiting hormone. Dev Cell 8 443–449 [DOI] [PubMed] [Google Scholar]
  11. Foo E, Bullier E, Goussot M, Foucher F, Rameau C, Beveridge CA (2005) The branching gene RAMOSUS1 mediates interactions among two novel signals and auxin in pea. Plant Cell 17 464–474 [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Foo E, Turnbull CGN, Beveridge CA (2001) Long-distance signaling and the control of branching in the rms1 mutant of pea. Plant Physiol 126 203–209 [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Johnson X, Brcich T, Dun E, Goussot M, Haurogné K, Beveridge CA, Rameau C (2006) Branching genes are conserved across species: genes controlling a novel signal in pea are co-regulated by other long-distance signals. Plant Physiol 142 1014–1026 [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Morris SE, Turnbull CG, Murfet IC, Beveridge CA (2001) Mutational analysis of branching in pea: evidence that Rms1 and Rms5 regulate the same novel signal. Plant Physiol 126 1205–1213 [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Murashige T, Skoog F (1962) A revised medium for rapid growth and bio-assays with tobacco tissue cultures. Physiol Plant 15 473–497 [Google Scholar]
  16. Napoli CA (1996) Highly branched phenotype of the petunia dad1-1 mutant is reversed by grafting. Plant Physiol 111 27–37 [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Napoli CA, Beveridge CA, Snowden KC (1999) Reevaluating the concepts of apical dominance and the control of axillary bud outgrowth. Curr Top Dev Biol 44 127–169 [DOI] [PubMed] [Google Scholar]
  18. Napoli CA, Ruehle J (1996) New mutations affect meristem growth and potential in Petunia hybrida Vilm. J Hered 87 371–377 [Google Scholar]
  19. Pfaffl MW (2001) A new mathematical model for relative quantification in real-time RT-PCR. Nucleic Acids Res 29 e45. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Rameau C, Murfet IC, Laucou V, Floyd RS, Morris SE, Beveridge CA (2002) Pea rms6 mutants exhibit increased basal branching. Physiol Plant 115 458–467 [DOI] [PubMed] [Google Scholar]
  21. Schmidt H, Kurtzer R, Eisenreich W, Scwab W (2006) The carotenase AtCCD1 from Arabidopsis thaliana is a dioxygenase. J Biol Chem 281 9845–9851 [DOI] [PubMed] [Google Scholar]
  22. Schwartz SH, Qin X, Loewen MC (2004) The biochemical characterization of two carotenoid cleavage enzymes from Arabidopsis indicates that a carotenoid-derived compound inhibits lateral branching. J Biol Chem 279 46940–46945 [DOI] [PubMed] [Google Scholar]
  23. Schwartz SH, Tan BC, Gage DA, Zeevart JAD, McCarty DR (1997) Specific oxidative cleavage of carotenoids by VP14 of maize. Science 276 1872–1874 [DOI] [PubMed] [Google Scholar]
  24. Snowden KC, Napoli CA (2003) A quantitative study of lateral branching in petunia. Funct Plant Biol 30 987–994 [DOI] [PubMed] [Google Scholar]
  25. Snowden KC, Simkin AJ, Janssen BJ, Templeton KR, Loucas HM, Simons JL, Karunairetnam S, Gleave AP, Clark DG, Klee H (2005) The decreased apical dominance1/Petunia hybrida CAROTENOID CLEAVAGE DIOXYGENASE8 gene affects branch production and plays a role in leaf senescence, root growth and flower development. Plant Cell 17 746–759 [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Sorefan K, Booker J, Haurogne K, Goussot M, Bainbridge K, Foo E, Chatfield SP, Ward S, Beveridge CA, Rameau C, et al (2003) MAX4 and RMS1 are orthologous dioxygenase-like genes that regulate shoot branching in Arabidopsis and pea. Genes Dev 17 1469–1474 [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Stirnberg P, van de Sande K, Leyser HMO (2002) MAX1 and MAX2 control shoot lateral branching in Arabidopsis. Development 129 1131–1141 [DOI] [PubMed] [Google Scholar]
  28. Turnbull CGN, Booker JP, Leyser HMO (2002) Micrografting techniques for testing long-distance signalling in Arabidopsis. Plant J 32 255–262 [DOI] [PubMed] [Google Scholar]
  29. Vandesompele J, De Preter K, Pattyn F, Poppe B, Van Roy N, De Paepe A, Speleman F (2002) Accurate normalization of real-time quantitative RT-PCR data by geometric averaging of multiple internal control genes. Genome Biol 3 research0034.1–0034.11 [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Woo HR, Chung KM, Park J-H, Oh SA, Ahn T, Hong SH, Jang SK, Nam HG (2001) ORE9 an F-box protein that regulates leaf senescence in Arabidopsis. Plant Cell 13 1779–1790 [DOI] [PMC free article] [PubMed] [Google Scholar]

Articles from Plant Physiology are provided here courtesy of Oxford University Press

RESOURCES