- © 2008 American Society of Plant Biologists
During flowering, the CENTRORADIALIS (CEN) gene of Antirrhinum majus and its homolog TERMINAL FLOWER1 (TFL1) in Arabidopsis (Arabidopsis thaliana) are required to maintain inflorescence identity of the shoot apical meristem (SAM) while flower meristems are produced on its flanks (Bradley et al., 1996, 1997). Consistent with this role, both genes are expressed at high level in the inflorescence SAM and their mutation converts the SAM into a terminal flower. By contrast, the homologous SELF-PRUNING (SP) gene of tomato (Solanum lycopersicum)—which together with CEN and TFL1 defines the CETS gene family (Pnueli et al., 1998, 2001)—might have a different function because the sp mutation does not alter inflorescence structure, but rather changes whole plant architecture (Yeager, 1927; MacArthur, 1932).
Tomato exhibits a sympodial growth pattern (Fig. 1A ); after the production of some leaves by the SAM, the growth of the primary shoot is terminated by the initiation of the first inflorescence, which is displaced from its terminal position by activation of the meristem at the axil of the last initiated leaf. The latter so-called sympodial meristem (SYM) continues shoot growth, carrying up the subtending leaf until it occupies a position above the inflorescence, which then develops laterally. The SYM undergoes a vegetative phase—producing most often three leaves—then initiates the second inflorescence, which is once again displaced laterally by the active outgrowth of the next SYM. The process is indefinitely reiterated and growth is thus indeterminate. We hereafter refer to: (1) the initial segment as the shoot portion produced by the SAM; (2) sympodial segments as the additional, successive shoot portions; (3) SYM as the meristems in the axil of the last leaf formed before an inflorescence; (4) AXM as the meristems present in the axil of other leaves and whose fate is not to form sympodial segments but to initiate secondary shoots. In sp mutants, the number of leaves initiated in successive sympodial segments is gradually reduced until production of two successive inflorescences and growth is thus determinate. The SP gene was therefore hypothesized to prevent early flowering of SYM (Pnueli et al., 1998).
Detection of SP transcripts by in situ hybridization in wild-type (Ailsa Craig) tomato plant. A, Schematic representation of plant architecture. Black shows the initial segment with its final number of leaves indicated at the last node; white and gray are the successive sympodial segments; circles, flowers; lanceolate forms, leaves; arrow, growth continuation. B to G, In situ hybridizations. B and C, Longitudinal sections of shoot apices harvested before (B) and after (C) floral transition. D, Negative control hybridized with a sense probe. E and F, Serial longitudinal sections in an AXM before floral transition of the SAM. G, Transverse section in an AXM before floral transition of the SAM. H, 3-D model of the SP expression domain, in external and sectioned views. I, Transverse section in an older AXM hybridized with an EXPANSIN probe. FM, Flower meristem; IM, inflorescence meristem; L, leaf; lp, leaf primordium; S, stem. Expression zones are indicated by red arrows.
Contrary to CEN and TFL1, whose expression is up-regulated at flowering and precisely limited to a subapical domain in the inflorescence SAM (Bradley et al., 1996, 1997), the SP gene was found to be expressed in both the SAM and leaves from very early stages of development, and later in inflorescences and flowers (Pnueli et al., 1998). Such a widespread pattern was unexpected because the sp mutation has no pleiotropic effects on the architecture of the initial segment, leaves, or inflorescences. This curious situation has been stressed by several authors comparing CETS homologs in different species (e.g. Amaya et al., 1999). The discovery that SP belongs to a multigene family (Carmel-Goren et al., 2003) led us to revisit its function and to test its expression pattern with a specific probe (Supplemental Fig. S1). Careful examination was needed because formation of the inflorescence in tomato is complex (for review, see Quinet and Kinet, 2007; Samach and Lotan, 2007). At floral transition, the vegetative SAM (Fig. 1B) swells before dividing into two parts, which give rise to a first floral meristem on one side and restore a mound of dividing cells on the other side (Fig. 1, C and D). This latter meristem, which has a continuous existence and bifurcates repeatedly for the production of each new flower, will be referred to as an inflorescence meristem, consistent with the view that tomato inflorescences show a raceme-like ontogeny (Quinet and Kinet, 2007). Several mutants suggest that the SAM converts into an inflorescence meristem in tomato. For example, uniflora (uf) initiates a single flower, indicating that the UF gene controls inflorescence meristem identity (Dielen et al., 2004). Other mutants are apparently defective in the maintenance of this identity, as is the case for single flower truss (sft), which produces either a solitary flower, or inflorescences that may repeatedly revert to leaf initiation and sympodial growth (Molinero-Rosales et al., 2004; Quinet et al., 2006a). We therefore undertook a comparative study of the expression pattern of SP in wild-type tomato and in these two mutants.
SPATIAL EXPRESSION PATTERN OF SP IN THE WILD TYPE AND IN SFT AND UF MUTANTS
In our growing conditions, wild-type Ailsa Craig produced approximately nine leaves in the initial segment before the SAM initiated the first inflorescence and successive sympodial segments most often had three leaves (Fig. 1A). In these plants, SP transcripts were detected in all axillary meristems of the initial segment irrespective of their fate—either SYM or AXM—but not in the vegetative SAM (Fig. 1B). This pattern remained unchanged following floral transition of the SAM, because SP expression was still detected in SYM and AXM, but not in the inflorescence nor in the flower meristems (Fig. 1C).
Serial longitudinal and transverse sections of SYM and AXM were extrapolated to provide a three-dimensional (3-D) view of the SP domain (Fig. 1H). In longitudinal sections, the hybridization signal appeared as oval or as two strands in tangential (Fig. 1E) or more axial (Fig. 1F) sections, respectively. Interestingly, the expression zone excluded the L1 and L2 outer layers of the meristems and was limited in depth at the junction of the subtending leaf with the stem. On transverse sections, the SP expression area had the shape of two crescents facing each other and joined at their margins (Fig. 1G). Comparison of the picture 1G with the transverse sections of AXM hybridized with an EXPANSIN probe (Fig. 1I), used as a marker of nascent leaf primordia (Reinhardt et al., 1998), indicated that the highest expression of SP was on the flanks of the meristem where leaves will later arise.
The sft and uf mutants initiated more leaves in the initial segment than wild-type Ailsa Craig but only uf was markedly delayed in the floral transition of sympodial segments, which produced up to nine leaves instead of three (Fig. 2, A and C ). As in the wild type, expression of SP was not detected in the vegetative SAM of sft and uf plants, nor in the inflorescence or flower meristems (Supplemental Fig. S2), but was observed in all kinds of axillary meristems, including AXM in the sft reverted inflorescences (Fig. 2, B and D).
Detection of SP transcripts by in situ hybridization in tomato sft and uf mutants. A and C, Schematic representation of the sft (A) and uf (C) plant architectures. Black shows the initial segment with its final number of leaves indicated at the last node; white and gray are the successive sympodial segments; circles, flowers; lanceolate forms, leaves; arrows; growth continuation. B and D, In situ hybridizations. B, Longitudinal section of a part of a sft reverted inflorescence. D, Longitudinal section of a uf shoot apex harvested after floral transition of the SAM. F, Flower; L, leaf. Expression zones are indicated by red arrows.
TEMPORAL EXPRESSION PATTERN OF SP IN THE SYM AND AXM OF WILD-TYPE PLANTS
The SP expression domain was followed in the initial segment during outgrowth of SYM and AXM following floral transition at the SAM. Similar changes were observed in both kinds of axillary meristems. As shown in Figure 3 , the meristems first increased in size (Fig. 3, A and B); the expression of SP was then observed to weaken progressively when the meristems initiated leaves (Fig. 3C), and remained “anchored” as two strands in a basal domain close to the subtending leaf/stem junction (Fig. 3D).
Detection of SP transcripts by in situ hybridization in developing AXM in wild-type (Ailsa Craig) tomato plant. A to D, Successive developmental stages. A, Young AXM. B, Bulging AXM. C, AXM initiating its first leaf primordium. D, AXM initiating a second leaf primordium. L, Leaf; lp, leaf primordium; S, stem. Expression zones are indicated by red arrows.
The expression pattern of SP reported here differs from the findings of Pnueli et al. (1998) who detected SP transcripts in the SAM, leaves, and flowers. Techniques probably account for this discrepancy, but the methodological information published by Pnueli and coworkers does not allow a detailed comparison. The specificity of the SP probe we used and the reproducibility—without exception—of our observations show that the expression domain of SP was strictly limited to axillary meristems. This pattern appeared as extremely robust because expression of SP was observed in axillary meristems irrespective of their fate (either SYM or AXM) and irrespective of the position of the subtending leaf (which can be located in an inflorescence as is the case in the sft mutant). Expression of SP in axillary meristems was also found to be independent of the fate of the SAM, which forms a flower rather than an inflorescence in the uf mutant (Fig. 2D; Supplemental Fig. S2D). These observations suggest that SP acts independently of UF and SFT. This hypothesis is further supported by the fact that the inflorescence defects caused by uf and sft mutations are additive to the determinate growth phenotype of sp mutants (Molinero-Rosales et al., 2004; Quinet et al., 2006b).
In conclusion, we observed that expression of SP in tomato is strictly limited to nongrowing axillary meristems. This pattern is similar in SYM and AXM, suggesting that SP function is not restricted to sympodial growth but concerns all axillary meristems. It is well known that, at floral transition of the SAM, axillary meristems are released from apical dominance (Bernier et al., 1981). We found here that this correlates with down-regulation of SP in out-growing SYM and AXM. Also consistent with a link between SP and apical dominance is the fact that sp mutants exhibit increased branching (Kinet and Peet, 1997). It is worth emphasizing that, although it might have been neglected in studies aimed at analyzing the function of CETS genes in inflorescence formation, the expression of these genes in axillary meristems seems to be the rule in all plant species examined so far; it was reported in Lotus japonicus, Impatiens balsamina, ryegrass (Lolium perenne), tobacco (Nicotiana tabacum), and Arabidopsis (Amaya et al., 1999; Ratcliffe et al., 1999; Jensen et al., 2001; Ordidge et al., 2005; Guo et al., 2006; Conti and Bradley, 2007). In tobacco, Amaya et al. (1999) reported that expression of the TFL1 homologs in axillary buds decreased after release from apical dominance by decapitation of the plants. All together, these results give, to our knowledge, new insights into a conserved function of TFL1 homologs in shoot branching.
Supplemental Data
The following materials are available in the online version of this article.
Supplemental Figure S1. Cross-hybridization test of the SP probe.
Supplemental Figure S2. Detection of SP transcripts by in situ hybridization in tomato sft and uf mutants.
Supplemental Materials and Methods S1. A supplemental Materials and Methods section.
Acknowledgments
We are very grateful to Professor Georges Bernier for his critical reading of this manuscript and to an anonymous reviewer for the thorough treatment he gave to the text. We acknowledge Professor Gérard Michel (Institut Supérieur des Beaux Arts Saint Luc, Liège, Belgium) for the 3-D drawings and Nathalie Detry for her excellent technical assistance.
Footnotes
-
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: Claire Périlleux (cperilleux{at}ulg.ac.be).
-
↵1 This work was supported by the Belgian Fonds National de la Recherche Scientifique, Fonds pour la Recherche Fondamentale et Collective (grant no. 2.4534.05). J.T. and M.Q. were awarded Ph.D. and research fellowships from the Fonds pour la Recherche dans l'Industrie et l'Agriculture and the Belgian Fonds National de la Recherche Scientifique, respectively.
-
↵2 These authors contributed equally to the article.
-
↵3 Present address: Imaging and Flow Cytometry Technological GIGA Facility, University of Liège (CHU), Bât. B23 Sart Tilman, Avenue de l'Hôpital 3, B–4000 Liège, Belgium.
-
↵[W] The online version of this article contains Web-only data.
- Received June 6, 2008.
- Accepted June 23, 2008.
- Published September 8, 2008.
LITERATURE CITED
