Intercellular signaling plays an important role in controlling cellular behavior in apical meristems and developing organs in plants. One prominent example in Arabidopsis is the regulation of floral organ shape, ovule integument morphogenesis, the cell division plane, and root hair patterning by the leucine-rich repeat receptor-like kinase STRUBBELIG (SUB). Interestingly, kinase activity of SUB is not essential for its in vivo function, indicating that SUB may be an atypical or inactive receptor-like kinase. Since little is known about signaling by atypical receptor-like kinases, we used forward genetics to identify genes that potentially function in SUB-dependent processes and found recessive mutations in three genes that result in a sub-like phenotype. Plants with a defect in DETORQEO (DOQ), QUIRKY (QKY), and ZERZAUST (ZET) show corresponding defects in outer integument development, floral organ shape, and stem twisting. The mutants also show sub-like cellular defects in the floral meristem and in root hair patterning. Thus, SUB, DOQ, QKY, and ZET define the STRUBBELIG-LIKE MUTANT (SLM) class of genes. Molecular cloning of QKY identified a putative transmembrane protein carrying four C2 domains, suggesting that QKY may function in membrane trafficking in a Ca2+-dependent fashion. Morphological analysis of single and all pair-wise double-mutant combinations indicated that SLM genes have overlapping, but also distinct, functions in plant organogenesis. This notion was supported by a systematic comparison of whole-genome transcript profiles during floral development, which molecularly defined common and distinct sets of affected processes in slm mutants. Further analysis indicated that many SLM-responsive genes have functions in cell wall biology, hormone signaling, and various stress responses. Taken together, our data suggest that DOQ, QKY, and ZET contribute to SUB-dependent organogenesis and shed light on the mechanisms, which are dependent on signaling through the atypical receptor-like kinase SUB.
Plant organs, such as flowers or leaves, are made up of distinct cell layers. Although communication across these cell layers is essential for organ development, we have only recently gained some insight into the underlying mechanisms. Receptor-like kinases are cell-surface receptors that perceive and relay intercellular information. In Arabidopsis, the receptor-like kinase STRUBBELIG is required for inter–cell-layer communication during floral development, amongst other functions; little is known, however, concerning its exact signaling mechanism. Here, we identified three new genes called DETORQUEO, QUIRKY, and ZERZAUST. Plants defective in any of these genes strongly resemble the strubbelig mutant, both at the whole-organ and cellular levels. Thus, all four genes may share or contribute to a common signaling pathway essential for plant morphogenesis. Analyses revealed complex interactions between the genes, indicating that each has additional and distinct activities. We provide the molecular nature of QUIRKY; the encoded protein is likely membrane-localised and predicted to require Ca2+ for activity. In light of analogous animal models, we speculate that QUIRKY facilitates transport of molecules to the cell boundary and may support a STRUBBELIG-related extracellular signal. These results open new inroads into a molecular understanding of inter-cellular communication during flower development.
Citation: Fulton L, Batoux M, Vaddepalli P, Yadav RK, Busch W, et al. (2009) DETORQUEO, QUIRKY, and ZERZAUST Represent Novel Components Involved in Organ Development Mediated by the Receptor-Like Kinase STRUBBELIG in Arabidopsis thaliana. PLoS Genet 5(1): e1000355. doi:10.1371/journal.pgen.1000355
Editor: Joseph R. Ecker, The Salk Institute for Biological Studies, United States of America
Received: May 1, 2008; Accepted: December 23, 2008; Published: January 30, 2009
Copyright: © 2009 Fulton et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This work was funded by grants SCHN 723/1-2 and SCHN 723/1-3 from the German Research Council (DFG) to KS and by the Free State of Bavaria. JUL was supported by a Career Development Award of the Human Frontier Science Program and the Max Planck Society. SUA and WB were recipients of Carlsberg Foundation and Cusanuswerk fellowships, respectively. JUL and KS were also supported by EMBO Young Investigator grants.
Competing interests: The authors have declared that no competing interests exist.
How intercellular communication mechanisms coordinate the activities of cells during organogenesis is an important topic in biology. In higher plants shoot apical meristems and floral meristems are the ultimate source of above-ground lateral organs, such as leaves, flowers, and floral organs . Meristems are organised into three distinct meristematic or histogenic layers, called L1, L2, and L3 , and cells of all histogenic layers contribute to organogenesis ,. The L1 layer gives rise to the epidermis while the L2 and L3 layers contribute to internal tissues. In Arabidopsis ovules, for example, the integuments that eventually develop into the seed coat are entirely made up of L1-derived cells, while L2 cells generate the inner tissue .
Classic studies have demonstrated that meristematic layers communicate ,, but it is only recently that the biological relevance and the molecular mechanisms are being elucidated –. For example, work on the receptor-like kinase (RLK) BRASSINOSTEROID INSENSITIVE 1 (BRI1) has provided evidence that the epidermis both promotes and restricts organ growth . Furthermore, microsurgical experiments indicated that the epidermis also maintains cell division patterns in subtending layers . These are but two examples that highlight the importance of the epidermis and inwards-oriented signaling in this inter-cell-layer cross-talk required for correct organ size and shape. At the same time, radial outward-oriented signaling also takes place during organogenesis. Known scenarios include transcription factors or small proteins that are synthesized in inner layers and move outwards into overlaying cell layers in a controlled fashion –. The so far best-characterised case of such a movement underlies radial patterning of the root ,. In addition, the epidermally-expressed RLKs CRINKLY4 (CR4) from corn or its Arabidopsis homolog ACR4 are necessary for epidermis development and may receive signals from underlying cell layers –.
Inter-cellular communication during floral morphogenesis in Arabidopsis also depends on signaling mediated by the leucine-rich repeat transmembrane receptor-like kinase (LRR-RLK) STRUBBELIG (SUB) . Analysis of sub mutants indicated that SUB is required for proper shaping of floral organs such as carpels, petals and ovules. At the cellular level SUB participates in the control of cell shape and/or the orientation of the cell division plane in floral meristems and ovules. In addition, SUB, also known as SCRAMBLED (SCM), affects specification of hair cells in the root epidermis ,. Recent evidence suggests that the SUB protein is confined to interior tissues in floral meristems, developing ovules and young roots although SUB mRNA is monitored throughout those organs . In particular functional SUB:EGFP fusion protein is absent from cells that show a mutant phenotype in sub mutants, but can either be found in adjacent cells, as in floral meristems and ovules, or in cells that are separated from mutant cells by two cell diameters, like in the root. The non-cell-autonomous effects of SUB were corroborated by an analysis of sub-1 plants expressing a functional SUB:EGFP transgene under the control of different tissue-specific promoters. Thus the data indicate that SUB undergoes posttranscriptional regulation, acts in a non-cell-autonomous fashion and mediates cell morphogenesis and cell fate across clonally distinct cell layers in an inside-out fashion . The SUB protein is a member of the LRRV/STRUBBELIG-RECEPTOR FAMILY (SRF) family of receptor-like kinases ,. It is predicted to carry an extracellular domain with six leucine-rich repeats, a transmembrane domain, and a cytoplasmic intracellular domain with the juxtramembrane and kinase domains. Interestingly, phosphotransfer activity of the SUB kinase domain is not essential for its function in vivo  and thus SUB seems to belong to the family of atypical or “dead” receptor kinases ,.
Very little is known regarding signaling through atypical receptor-like kinases in plants . In addition, it remains to be understood how cellular morphogenesis is coordinated across cell layers –. It is therefore of great interest to investigate the molecular basis of SUB signaling and function. Here we present the identification and analysis of three genetic factors that may relate to SUB signaling. Our results show that mutations in QUIRKY (QKY), ZERZAUST (ZET), and DETORQUEO (DOQ) result in a sub-like phenotype. Molecular cloning of QKY revealed that the predicted QKY protein is likely a transmembrane protein with four C2 domains indicating a role for QKY in Ca2+-dependent signaling. Global gene expression profiling of the mutants corroborates the morphological analysis but also suggests additional and distinct roles for each gene. Furthermore, the data indicate that SUB signaling plays previously unknown roles in cell wall and stress biology.
Isolation of sub-Like Mutants
We applied a forward genetic approach to isolate additional factors of the SUB signaling pathway, based on the hypothesis that mutations in some of the genes that are part of the SUB pathway should result in sub-like (slm) mutant phenotypes. We thus screened M2 families of an ethylmethane sulfonate-mutagenized Ler population for slm mutants (see Materials and Methods). In this experiment we identified two new sub alleles  as well as several novel mutants with sub-like phenotypes (Figures 1–3). These fell into three different complementation groups, which map to distinct positions on chromosome 1 (Table 1). We termed two of the genes DETORQUEO (DOQ) and ZERZAUST (ZET), respectively. DETORQUEO refers to a latin term that means “to twist out of shape”. ZERZAUST is a German term for “disheveled”. We also isolated three mutant alleles of QUIRKY (QKY), which plays a role in fruit dehiscence (L.F., unpublished results; S.J. and Martin F. Yanofsky, unpublished observations) and the numbering of qky alleles was coordinated. Thus, our genetic approach resulted in the identification of three loci, DOQ, QKY, and ZET, mutations in which result in a sub-like phenotype and that, together with SUB, define the STRUBBELIG-LIKE MUTANT (SLM) class of genes.
Figure 1. Comparison of the overall above-ground morphology of sub-1, doq-1, qky-8, and zet-2 mutants.
The given stages correspond across equivalent panels. (A–E) Wild-type Ler. (F–K) sub-1. (L–Q) doq-1. (R–W) qky-8. (X-Ac) zet-2. (A, F, L, R, X) An open stage 13 flower from a 30-day old plant. Note the misorientation of petals due to twisting in the basal end of the petal structure (arrows). (F, R, X) Petals can also show small notches. (B, G, M, S, Y) Top view of a 30-day inflorescence. (G, M, S, Y) Flower phyllotaxis is irregular. Arrows mark prematurely opened flower buds. (C, H, N, T, Z) Top view of a 12-day rosette. (H, N) Leaf petioles can be twisted (arrows). (N) doq-1 leaves have longer petioles and narrow blades (arrow heads). (D, I, O, U, Aa) Morphology of mature siliques. Typical twisting observed in (I) sub-1, (U) qky-8 and (Aa) zet-2 mutants. doq-1 exhibits more subtle twisting in both (O) siliques and (P) stems. (E, J, P, V, Ab) A lateral view of a section of stems from a 30-day plant. (V) qky-8 and (Ab) zet-2 mutant show twisting in stems equivalent to (J) sub-1 plants. (K, Q, W, Ac) Plant height of mutants (left) in comparison to Ler (right). (K) sub-1 and (W) qky-8 mutants show a clearly reduced plant height. (Q) The doq-1 and (Ac) zet-2 mutants show only a slight reduction. Scale bars: (A, D, E, F, I, J, L, O, P, R, U, V, X, Aa, Ab) 0.5 mm, (B, C, G, H, M, N, S, T, Y, Z) 2 mm, (K, Q, W, Ac) 3 cm.doi:10.1371/journal.pgen.1000355.g001
Figure 2. Comparison of ovule morphology in sub-1, doq-1, qky-8, and zet-2 mutants.
(A) Wild-type Ler. The arrow marks one of the elongated cells of the distal outer integument. (B) sub-1. A mild phenotype is shown. Note the irregular size and shape of cells at the distal outer integument (arrow heads, compare to (A)). (C, D) sub-1. Strong phenotypes are depicted. Note the half-formed outer integument. (D) shows an example where the outer integument shows several gaps. (E) doq-1. This mutant shows a mild ovule phenotype comparable to the one depicted in (B). The arrow heads highlight the disruption of the regular cell files of the distal outer integument. (F) qky-8. Note the variability of the phenotype. The specimen to the left shows only a mild disorganisation of the cell shape (arrow head). The one to the right shows a strong phenotype (compare to (C)). (G) zet-2. This mutant shows a variable phenotype. The ovule to the left shows gaps in the outer integument while the one to the right exhibits only mild alterations in cell shape and size (arrow head). Abbreviations: fu, funiculus; ii, inner integument; mp, micropyle; oi, outer integument. Scale bars: 20 µm.doi:10.1371/journal.pgen.1000355.g002
Figure 3. Analysis of cellular defects in 4-day old main roots and stage 3 floral meristems of sub-1, doq-1, qky and zet-2 mutants.
(A–E) Expression of the GL2::GUS reporter in whole-mount main roots. (A) Wild-type Ler root. The reporter is detected in regular files of non-hair cells. (B–E) GL2::GUS reporter expression is patchy. (F–K) Mid-optical longitudinal sections through stage 3 floral meristems obtained by confocal microscopy of propidium-iodide-stained specimen. (F) Wild-type Ler. Note the regular arrangement of cells in the L1 and L2. (G) sub-1. The arrow marks a periclinal cell division event. (H) doq-1. The region marked by the square (i) is shown at higher magnification in (Hi). The arrows highlight aberrant oblique and periclinal cell divisions in the L1 and L2, respectively. (I) qky-8. The region marked by the square (i) is shown at higher magnification in (Ii). The arrow labels a periclinal cell division in the L2. (J–K) zet-2. (J) The region marked by the square (i) is shown at higher magnification in (Ji). The arrows highlight periclinal cell divisions. A cell undergoing cell separation is indicated. (K) Disintegrating cells are marked. Abbreviations: cd, cell disintegration; cs, cell separation; L1, L1 cell layer; L2, L2 cell layer. Scale bars: (A–E) 25 µm, (F–K) 20 µm.doi:10.1371/journal.pgen.1000355.g003
Table 1. Summary of sub-like mutants identified.doi:10.1371/journal.pgen.1000355.t001
Comparison of sub-1, doq-1, zet-2, and qky-8 Phenotypes
We identified one mutant allele of DOQ and two and three independent alleles of ZET and QKY, respectively (Table 1). All sub, doq, zet and qky alleles were recessive and behaved in a Mendelian fashion (not shown). The various zet and qky mutants did not noticeably differ in their respective phenotypes and the three qky alleles are likely to be nulls (see below). Thus, zet-2 and qky-8 or qky-9 were used as reference alleles for further analysis. In addition, we used the well-characterised sub-1 mutant for comparison . This mutation likely represents a null-allele since it results in a stop codon and a predicted shorter SUB protein lacking the transmembrane and intracellular kinase domains. Thus, it is expected that SUB-dependent signaling across the plasma membrane is blocked in sub-1 mutants.
At the macroscopic level sub-1 mutants are known to be affected in several above-ground organs  (Figures 1–4) (Tables 2, 3). Inflorescences are characterised by reduced height, an irregularly twisted stem, and an aberrant phyllotaxis of flowers. Flowers open prematurely and show a large percentage of twisted and often notched petals. Furthermore, all flowers exhibit twisted carpels and about 70 percent of sub-1 ovules showed aberrant initiation of the outer integument (Figure 2) (Table 2). This results in outer integuments with gaps that often resemble “multifingered clamps” or “scoops”. Also ovules with a fully developed outer integument show defects. In particular the distal or micropylar cells of the outer integument can show aberrant size and shape. In addition, about 40 percent of sub-1 plants show at least one leaf per rosette with twisted petioles.
Figure 4. Comparison of above-ground morphology of sub-1, doq-1, qky-8, and zet-2 double mutants.
The given stages are identical across panels. (A–F) sub-1 doq-1. (G–L) sub-1 qky-8. (M–R) sub-1 zet-2. (S–X) doq-1 qky-8. (Y-Ad) doq-1 zet-2. (Ae–Aj) qky-8 zet-2. (A, G, M, S, Y, Ae) An open stage 13 flower from a 30-day old plant. Note the petal misorientation due to twisting along the longitudinal axis of petals. (M) sub-1 zet-2, (S) doq-1 qky-8 and (Y) doq-1 qky-8 flowers show particularly aberrant perianth morphology, with thin petals and heavy notching evident. (B, H, N, T, Z, Af) Top view of a 30-day inflorescence. (B) sub-1 doq-1, (H) sub-1 qky-8 and (N) sub-1 zet-2 inflorescences show aberrant phyllotaxy. (B, H, N, Z) Premature opening of flower buds is apparent. (C, I, O, U, Aa, Ag) A lateral view of a stem segment from a 30-day plant. (D, J, P, V, Ab, Ah) Morphology of a mature silique. Stem and silique twisting is exaggerated as compared to single mutant parentals, with the exception of the qky-8 zet-2 double mutant (Ag, Ah). (D, P, V, Ab) Note floral organs have not abscised. (E, K, Q, W, Ac, Ai) Top view of a 16-day rosette. Arrows indicate exaggerated petiole twisting in leaves. (Ai) qky-8 zet-2 leaf petioles are not twisted. (F, L, R, X, Ad, Aj) Plant height comparisons between Ler (left), single mutant parental (middle) and double mutant (right) 30-day plants. (F) sub-1 doq-1 and (L) sub-1 qky-8 mutants are significantly dwarfed, whereas (R) sub-1 qky-8 and (X) doq-1 qky-8 plants exhibit modest reductions in height. (Ad, Aj) doq-1 zet-2 and qky-8 zet-2 plant height is not significantly different from parental mutants. Scale bars: (A, C, D, G, I, J, M, O, P, S, U, V, Y, Aa, Ab, Ae, Ag, Ah) 0.5 mm, (B, H, N, T, Z, Af) 2 mm, (E, K, Q, W, Ac, Ai) 1 cm, (F, L, R, X, Ad, Aj) 3 cm.doi:10.1371/journal.pgen.1000355.g004
Table 2. Ovule defects observed in sub and slm single and double mutants.doi:10.1371/journal.pgen.1000355.t002
Table 3. Cellular defects in sub and sub-like mutant floral meristems.doi:10.1371/journal.pgen.1000355.t003
At the cellular level sub-1 exhibits aberrant cell shape and robustly scorable numbers of periclinal, rather than anticlinal cell division planes in cells of the L2 layer of stage 3 floral meristems  (Figure 3) (Table 3). Interestingly, in this analysis we could also observe a previously unnoticed low number of periclinal divisions in the L1 of sub-1 floral meristems. Furthermore, sub/scm mutants develop root hairs at discordant positions in the epidermis ,. This defect in root hair patterning can be followed by expressing the bacterial β-glucuronidase gene under the control of the Arabidopsis GLABRA2 (GL2) promoter (GL2::GUS) , (Figure 3). This reporter conveniently labels the regular files of non-hair cells in the epidermis of wild-type roots and exhibits an irregular expression pattern in sub/scm mutants.
Upon examination, many aspects of the phenotypes of sub-1 and the other slm mutants were comparable. The doq-1, qky-8 and zet-2 mutants showed aberrant floral phyllotaxis and flowers with twisted petals (Figure 1). In addition, we observed premature floral bud opening in doq-1 and zet-2. Petals of zet-2 mutants showed notches similar to sub-1 petals but zet-2 floral organs were generally more misshapen. In rosettes, nearly all plants showed examples of leaf petiole twisting. In particular, qky-8 mutants showed leaf twisting comparable to sub-1, whereas doq-1 rosette leaves showed in addition elongated petioles and narrow blades. In contrast, zet-2 rosette leaves did not show major defects at the gross morphology level. Irregular twisting of siliques and stems was apparent in doq-1, qky-8 and zet-2 mutants, although the twisting in doq-1 siliques and stems was more subtle. Plant height was most affected in qky-8 while doq-1 and zet-2 showed only a slight reduction. Mature ovules of doq-1 mutants were mildly but consistently affected (Figure 2) (Table 2). The outer integument did not fully extend to the funiculus (reduced campylotropy). In addition, its distal cells were shorter and of irregular shape. Mature ovules of qky-8 and zet-2 closely resembled ovules of sub-1 mutants with respect to gaps and altered cell shape in the outer integument. Interestingly, zet-2 showed a slightly higher percentage of malformed outer integuments. As was the case for sub  the inner integument in all other slm single mutants appeared to be unaffected.
Further, doq-1, qky-9 and zet-2 mutants were investigated for cellular defects in the root epidermis and in L1/L2 cells of stage 2 to 4 floral meristems (Figure 3) (Table 3). In 4 day old main roots, all three mutants showed misregulation of GL2::GUS reporter expression comparable to sub-1, indicating that root hair patterning was similarly affected. The three mutants also exhibited cell shape and periclinal cell division plane defects in the L1 and L2 cells of floral meristems. Floral meristems of doq-1 showed a higher percentage of those defects while in zet-2, cell separation and disintegration could also be observed. The latter finding indicates that ZET is also required for cell viability.
Taken together the analysis of the slm single mutants revealed that there is a large functional overlap of the corresponding genes and that individual SLM genes participate in subsets of SUB-dependent processes, such as ovule development, cellular behavior of L2 cells of stage 3 floral meristems, and root hair patterning. The results also suggest, however, that SLM genes have additional functions unrelated to each other. For example, ZET has a particular function in cell survival in floral meristems as has DOQ in the regulation of leaf shape.
Double Mutant Analysis
To investigate further the genetic relationship between SLM genes we generated all possible double-mutant combinations and analysed the respective mutant phenotypes. The results are summarized in Figures 4 and 5 and Table 2.
Figure 5. Comparison of ovule morphology in different slm double mutants.
SEM micrographs of stage-4-V ovules are depicted. (A–C) sub-1 doq-1. (D–F) sub-1 qky-8. (G–I) sub-1 zet-2. (J–L) doq-1 qky-8. (M–O) doq-1 zet-2. (P,Q) qky-8 zet-2. Each double mutant shows a range of ovule phenotypes. (A, D, J, M and P) Mild phenotypes are depicted. Note the irregular size and shape of cells at the distal outer integument (arrow heads). (B, C, D, G K, N Q) Stronger ovule phenotypes are observed whereby specimens have half formed outer integuments. (C, F, I, L, O) In extreme cases, sub-1 doq-1, sub-1 qky-8, sub-1 zet-2, doq-1 qky-8 and doq-1 zet-2 ovules can show malformation of the inner integument. (C) This specimen is an example where the inner integument shows several gaps. (F, I, L, O) Inner integuments can also appear as finger-like protrusions. The qky-8 zet-2 double mutant is an exception, whereby inner integument defects are rarely observed. (E, H) Severely malformed ovules can also appear as integument-like and nucellus-like structures emerging from placental tissue. (R) Fertility was assayed as the average seed number per silique (n = 20). Abbreviations: fu, funiculus; ii, inner integument; mp, micropyle; oi, outer integument. Scale bars: 20 µm.doi:10.1371/journal.pgen.1000355.g005
Petals of sub-1 doq-1 double mutants mostly resembled doq-1 petals (Figure 4A), however, stem twisting was more similar to sub-1. Other aspects of the sub-1 doq-1 phenotype, such as silique twisting, plant dwarfism and rosette leaf petiole twisting were more exaggerated when compared to either single mutant. Ovules of sub-1 doq-1 plants showed an increase in outer integument defects, although not as strong as in some other double mutant combinations. In addition, about 11% of ovules of sub-1 doq-1 plants showed defects in inner integument morphology (Figure 5C) (Table 2), with gaps of variable sizes and finger-like protrusions. The fertility of sub-1 doq-1 plants was severely reduced (Figure 5R).
Petal and carpel twisting in sub-l qky-8 double mutants was similar to that shown by each single mutant (Figure 4G). In contrast, twisting of siliques, stems, and leaf petioles was more pronounced as was the reduction in plant height (Figure 4I–L). In addition, ovule development was more heavily affected compared to the single mutants (Figure 5D–F) (Table 2), with 13% of ovules showing inner integument defects and a corresponding reduction in fertility (Figure 5R). About 5% of ovules were hardly recognizable as such, but rather resembled a mass of cells with integument-like outgrowths (Figure 5E).
Overall, the sub-1 zet-2 double mutants showed the most disturbed morphology (Figure 4M–R). Perianth organs were twisted, narrower and notched, while carpels were heavily twisted. As a result the overall structure of the flower was irregular. The stem phenotype was knotted rather than twisted. Leaf morphology was misshapen with more pronounced leaf petiole twisting and plants showed prominent dwarfism. Consistent with this exaggerated phenotype ovules of sub-1 zet-2 plants showed severe defects with 21% of ovules resembling a mass of cells with integument-like outgrowths (Figure 5H) (Table 2). About 18% of sub-1 zet-2 ovules exhibited short gapped outer integuments and similarly malformed inner integuments (Figure 5I). Again, fertility was reduced in sub-1 zet-2 plants (Figure 5R).
Perianth organs and carpels in doq-1 qky-8 mutants were more twisted as compared to the parental lines (Figure 4S). Twisting of leaf petioles, stems, and siliques was also more pronounced as was dwarfism. Ovules of doq-1 qky-8 mutants showed a less exaggerated phenotype compared to some of the other double mutant combinations. Still, a large proportion of doq-1 qky-8 ovules showed gaps in the outer integument (Figure 5J–L) (Table 2) and aberrant inner integument morphology was seen in 9% of ovules (Figure 5L). Fertility was strongly reduced in doq-1 qky-8 double mutants (Figure 5R).
Perianth morphology in doq-1 zet-2 mutants was about equivalent to that observed in sub-1 zet-2 and doq-1 qky-8 flowers (Figure 4Y). Twisting of stems appeared slightly more pronounced compared to each single mutant while siliques were drastically more twisted. Plant height was comparable to zet-2 single mutants. Overall a higher percentage of doq-1 zet-2 ovules showed gaps in the outer and inner integuments (Figure 5M–O), with 4% of ovules exhibiting severe malformations comparable to sub-1 zet-2 mutants (Table 2). The reduction in fertility was comparable to that of doq-1 qky-8 double mutants (Figure 5R).
The qky-8 zet-2 double mutants largely phenocopied zet-2 single mutants in above-ground morphology (Figure 4Ae–Aj) including ovules (Figure 5P,Q) (Table 2). One exception to this was leaf petiole twisting as this aspect was most similar to the phenotype in qky-8 single mutants.
In summary, the pleiotropic phenotypes of slm single and double mutants complicated the double mutant analysis. Although an exaggerated phenotype was often observed in a double mutant combination it was usually difficult to decide whether a double mutant displayed an additive or synergistic phenotype, or whether a particular mutation was epistatic to another. Overall, the double mutant analysis reinforced the notion that SLM genes likely do not act in a single linear pathway but have both overlapping and separate functions.
Phenotypic Analysis of doq, qky, and zet Plants Carrying a 35S::SUB Transgene
To further explore the genetic relationship between SLM genes we first tested whether SUB expression was responsive to other genes of this group. We investigated SUB expression in inflorescence apices and stage 10–12 flowers (see below) from several slm mutants by quantitative real time PCR (qRT-PCR). As can be seen in Figure 6A only very moderate changes in SUB expression were detected with a slight but consistent reduction of SUB expression in flowers of doq-1, qky-8 and zet-2 mutants. In contrast, SUB transcript levels were unaltered in those tissues and mutants when assessed by a transcriptome analysis (see below). Given that SUB is expressed at very low levels to begin with  we reasoned that the observed mild effects may not be the result of direct effects but rather be a consequence of indirect influences due to tissue sampling or the altered morphology of the mutants. To test the relevance of the effects we asked whether rendering SUB independent of its normal transcriptional regulation, by ectopically expressing SUB using the cauliflower mosaic virus 35S promoter, could result in phenotypic rescue of doq, qky and zet mutants. We analysed the phenotypes of 35S::SUB doq-1, 35S::SUB qky-8 and 35S::SUB zet-2 plants (Figure 6B). In particular we scored stem twisting, flower morphology and silique twisting of individual transgenic T1 lines. Transgene functionality was demonstrated by the wild-type appearance of 35S::SUB sub-1 plants. Despite demonstrating comparable transgene expression (Figure 6C), the other tested combinations did not show phenotypic rescue (35S::SUB doq-1: 42 T1 lines scored, 35S::SUB qky-8: >100, and 35S::SUB zet-2: >100), indicating that SUB does not act as a downstream target gene of DOQ, QKY or ZET. Taken together our data suggest that SUB is not directly regulated at the transcriptional level by the other SLM genes.
Figure 6. Analysis of SUB expression in the slm plant lines.
(A) Quantitative expression analysis of SUB in the slm mutants. Steady-state mRNA levels were measured in apex and flower tissues by quantitative real-time PCR. UBC21 expression was used for normalization. (B) Phenotypes of 35S:SUB slm transgenic lines. A 35S::SUB:myc construct was transformed into each slm mutant background. T1 plants were analysed for rescue of flower, silique and stem mutant phenotypes. Note only sub-1 mutants exhibit rescue by SUB overexpression. (C) Transgenic 35S:SUB:myc T1 plants represented in (A) were assayed for SUB transgene expression using semi-quantitative RT-PCR (top panels). GAPC  served as internal control (bottom panels). Scale bars: 0.5 mm.doi:10.1371/journal.pgen.1000355.g006
Genome-Wide Transcriptome Analysis and Global Expression Profiling
To shed light on the molecular nature of SUB-dependent processes and to systematically investigate the similarities between the SLM genes at the mechanistic level, we applied a transcriptome analysis. Since mutants with defects in genes of overlapping functions should also exhibit overlapping alterations in transcript profiles we compared wild type (Ler) with plants of two sub alleles , as well as doq-1, zet-2 and qky-8 mutants using the Affymetrix ATH1 GeneChip platform (see Materials and Methods). The analysis of two independent sub alleles was aimed at obtaining a robust set of SUB-responsive genes for the different pairwise comparisons and the list of SUB-responsive genes represents the overlap of misexpressed genes in sub-1 and sub-3 mutants . Since we had observed tissue-specific differences in the phenotypes of slm mutants, which suggested tissue-specific sub-functions for the corresponding genes, we tried to capture these differences also in the sampling for transcriptome analysis. To this end, we sampled inflorescence apices plus flowers up to stage 9 (apex data set), representing mostly proliferative tissues, while stage 10–12 flowers were collected to represent maturation stages (flower data set). The samples thus covered SUB-related aspects such as the control of cell shape/division plane in L2 cells of floral stage 3 meristems, and the regulation of petal, carpel and ovule development.
Since traditional analysis tools to identify differential gene expression in whole genome transcriptome data are not well suited for comparisons of multiple samples we developed a meta analysis tool based on Z-score statistics (see Materials and Methods). It offers a variety of benefits for the simultaneous analysis of multiple pairwise comparisons, such as normalization for the overall biological effect observed in the individual experiments, increased sensitivity, and the possibility of querying data using Bolean logic. Using this tool we identified 89 and 193 significantly misexpressed genes, respectively, in the apex and flower samples of sub plants (Table 4). The other mutants were characterised by higher numbers of misexpressed genes (Table 4) indicating that DOQ, QKY, and ZET affect more processes than SUB. More importantly, we systematically analyzed the overlap of misexpressed genes from individual slm mutants compared to wild type. The results of all pairwise comparisons are given in Table 5. For example, a pairwise comparison between sub and doq-1 in the apex sample revealed a 42 gene overlap. This corresponded to 47% of genes misexpressed in sub and about 17% of genes aberrantly expressed in doq-1. Comparable values were observed for sub versus qky-8 and sub vs zet-2 comparisons. The overlap with SUB-dependent genes was even more pronounced in the flower data set and in the sub-doq-1 comparison we found 118 common genes, representing 61% and 17% of the genes misexpressed in sub and doq-1, respectively. The sub qky-8 and sub zet-2 comparisons revealed higher values with 76% and 81% of SUB-responsive genes being found to overlap. Thus, a significant number of SUB-responsive genes are also sensitive to DOQ, QKY and ZET function. Similar pairwise comparisons were made between doq-1, qky-8 and zet-2 (Table 6). Again there was considerable overlap of misexpressed genes between different mutants. Most strikingly, in stage 10–12 flowers 67% and 71% of the QKY-responsive genes are also misregulated in doq-1 and zet-2, respectively. This finding implies that the activity of more than two-thirds of QKY-responsive genes also depends on DOQ and ZET function in stage 10–12 flowers. Taken together the microarray data corroborate the morphological analysis of the single mutants and are compatible with the notion that the different SLM genes share overlapping functions.
Table 4. Genes misexpressed in sub and slm mutants.doi:10.1371/journal.pgen.1000355.t004
Table 5. Comparison of genes co-misexpressed in sub and slm mutants.doi:10.1371/journal.pgen.1000355.t005
Table 6. Comparisons of genes co-misexpressed in slm mutants.doi:10.1371/journal.pgen.1000355.t006
To further study the complexity of the functional relationship among SLM genes we carried out additional pairwise comparisons, examined the identity of genes present in the overlap lists, and identified transcripts which are misregulated in all four mutants (i.e., a sub, doq-1, qky-8 and zet-2 “quadruple” overlay) (Figures 7,8; Tables 5,6; Datasets S1, S2). While doing so we uncovered striking trends in the transcriptome data. For example, we found that more than 30% of the misregulated genes found in any sub/slm overlap identified in the apex samples are indeed shared by all mutants (Figure 7A). More significantly, 13 of the 14 transcripts identified as common SLM responsive in the apex were consistently elevated in expression in all mutants, suggesting that a core function of the SLM genes is to repress this set of genes in proliferating tissue (Figure 7B). Strikingly, this trend was reversed in differentiating tissue represented by the floral samples. Here the misregulated genes shared by all mutants made up more than 65% of transcripts identified in pairwise comparisons (Figure 7A) and 88 of the 93 genes were reduced in expression (Figure 7B). In addition to supporting the notion of functional overlap between the SLM genes, this analysis also demonstrated that diversification of SLM function observed in different tissues is based on a dramatic switch in the underlying molecular mechanism. While signaling through SLM proteins in proliferating tissue mainly causes repression of target gene transcription, SLM mediated signals lead to a transcriptional activation of downstream genes in differentiating cells, suggesting that SUB and other SLM proteins interact with a radically different regulatory environment. This notion is supported by another striking observation: while we found strong similarities between different genotypes within one tissue we observed very little overlap in misregulated genes between the same genotype in the two tissues analysed (Figure 8). For example, only four misregulated transcripts were shared in apex and flower samples of sub mutants. Similar observations were made for the other single and pairwise comparisons. In addition, the core target gene sets of apex and flower, derived from quadruple comparisons, did not share any transcripts.
Figure 7. Comparison of gene expression changes between sub and slm mutants.
(A) Venn diagrams showing the overlap of genes detected as significantly misexpressed (z-score<−2 or >2) in sub, doq-1, qky-8 and zet-2 mutants. Analyses were performed using data generated from apex (left) and late-stage flower (right) probe sets. Gene lists that compared sub to each independent slm mutant were initially generated (numbers in boxes) and then overlapped for a final comparison (quadruple overlap set). (B) Line graphs depicting the expression profiles of genes present in the quadruple overlap set across sub and slm mutants in apex (left) and flower (right) tissues. Note how the up- or downregulation of individual genes coincides across all mutant genotypes.doi:10.1371/journal.pgen.1000355.g007
Figure 8. Comparison of genes misexpressed in apex and flower overlap sets.
(A) Venn diagram depicting the overlap of genes misexpressed in apex and flower quadruple overlap sets. (B) Box diagram detailing the overlap of genes misregulated in apex and flower gene sets for individual sub and slm comparisons.doi:10.1371/journal.pgen.1000355.g008
To investigate the nature of SLM-dependent processes we assessed the known or predicted functions of the apex and flower core sets of SLM-responsive genes by searching The Arabidopsis Information Resource (TAIR), the literature and making use of the AtGenExpress expression atlas  with the help of the AtGenExpress Visualisation Tool (AVT, http://jsp.weigelworld.org/expviz/) (Tables 7, 8). In both core sets we observed a strong representation of genes that are involved in processes, such as cell wall biosynthesis and function, hormone signaling and abiotic and biotic stress responses. For example, two genes in the apex core set encode the inositol oxygenase family enzymes MIOX2 and MIOX4 required for biosynthesis of uridine-diphospho-glucuronic acid (UDP-GlcA), a precursor for various cell-wall matrix polysaccharides . In addition, GERMIN-LIKE PROTEIN 6 (GLP6) is a member of a gene family encoding putative extracellular GERMIN-LIKE proteins  some of which play a role in biotic stress responses and cell wall biology. The flower core set of SLM-responsive genes also included several genes coding for cell wall proteins, such as ECS1 , At2g05540 (Glycine-rich protein), At5g03350 and At5g18470 (lectin family proteins), and At2g43570 and At4g01700 (chitinases). In another example that relates to hormone signaling, five of the 13 upregulated apex core set genes were inducible by jasmonates (JA) while the flower core set was characterized by a large group of genes that relate to different aspects of signaling mediated by salicylic acid (SA) (see Discussion). SLM-responsive genes whose expression is sensitive to JA encoded for example JAZ1, a central negative regulator of JA signaling ,, ERD5 a mitochondrial proline dehydrogenase (ProDH) , AtLEA5/SAG21 ,, and AtTTPG, a class III trehalose-6-phosphate phosphatase involved in the biosynthesis of trehalose ,. All of these genes are believed to contribute to the plant's response to dehydration, cold and oxidative stress , , –. We also observed SLM-dependency of genes involved in the homeostasis of the auxin indole-3-acetic acid (IAA) and the production of the Arabidopsis phytoalexin camalexin and indole glucosinolates. Glucosinolate metabolites play important roles in plant nutrition, growth regulation and the interactions of plants with pathogens and insect herbivores ,. In the apex core set we identified IAA-LEUCINE RESISTANT-LIKE GENE 6 (ILL6), which belongs to a gene family encoding IAA amino acid conjugate hydrolases involved in the release of the active auxin from corresponding inactive amino acid conjugates ,. CYP83B1 and CYP79B2 encode cytochrome P450 monooxygenases that regulate the pool of indole-3-acetaldoxime (IAOx) – a key branching point of primary and secondary metabolic pathways and a precursor of indole glucosinolates, camalexin , and IAA (reviewed in ,). In addition, the flower core set contained two downregulated cytochrome P450 genes, CYP71A13 and CYP71B15, which promote distinct steps in camalexin biosynthesis –.
Table 7. List of genes co-misexpressed in slm mutant apex tissues.doi:10.1371/journal.pgen.1000355.t007
Table 8. List of genes co-misexpressed in slm mutant flowers.doi:10.1371/journal.pgen.1000355.t008
Molecular Identification of QKY
As a first step to molecularly define the SLM pathway, the QKY gene was identified by map-based cloning (Figure 9 and Materials and Methods). In all three EMS-induced qky alleles we could identify single point mutations in At1g74720 (Figure 9A). Further, two distinct T-DNA insertions in this locus result in qky mutants phenocopies (not shown). Combined this demonstrates that At1g74720 is QKY. Sequence analysis predicted that QKY contains no introns and encodes a transmembrane protein of 1081 amino acids with a calculated molecular mass of 121.4 kDa harboring four C2 domains (Figure 9B). In addition, two transmembrane domains are embedded in a plant-specific phosphoribosyltransferase C-terminal region (PRT_C, PFAM identifier PF08372), which according to the PFAM database occurs characteristically at the carboxy-terminus of phosphoribosyltransferases and phosphoribosyltransferase-like proteins . Interestingly, this domain often appears together with C2-domains, as in QKY. A related domain topology, albeit with only three C2 domains, is found in several proteins present in humans, Drosophila melanogaster and Caenorhabditis elegans (C. elegans), collectively referred to as multiple C2 domain and transmembrane region proteins (MCTPs) . The role of animal MCTPs is not well defined despite the fact that in C. elegans a single MCTP gene is present and essential for embryo viability . QKY represents the first described plant MCTP-class gene. In accordance with proposed transmembrane topography for MCTPs the QKY C2 domains most likely have an intracellular localisation as QKY lacks a distinguishable N-terminal signal peptide and all known C2 domains are cytoplasmic. C2 domains are autonomously folding modules and form Ca2+-dependent phospholipid complexes, although some exceptions are known that do not bind Ca2+ or phospholipids or both –. Other animal proteins with multiple C2 domains, but only one transmembrane region, include the synaptotagmins , the extended synaptotagmins , and the ferlins . Synaptotagmins and ferlins are known to play a role in membrane trafficking. Although there is general resemblance in domain topology, very little primary sequence conservation is observed between QKY and the animal MCTPs, synaptotagmins, and ferlins.
Figure 9. Molecular characterization of QKY.
(A) Molecular organization of the QKY locus on chromosome 1. The horizontal black line highlights genomic DNA. The red arrow indicates the orientation and extent of the QKY open reading frame. The small black boxes mark the 5′ and 3′ untranslated regions, respectively. The orange arrow and box denote the end of the preceding and the start of the next annotated open reading frames, respectively. The positions of the various point mutations or T-DNA insertion sites are marked. (B) Predicted domain topology of the QKY protein. The sequence and numbering of the C2 domains, and the altered C-termini of the mutant protein variants, are given. The italic capital letters represent the de novo residues present in the predicted mutant QKY proteins from the two T-DNA lines. Stars mark the premature stops in the predicted mutant proteins. The green box denotes the plant phosphoribosyltransferase C-terminal region (PRT_C). Abbreviations: aa, amino acids; N, amino terminus; C, carboxy terminus; TM, transmembrane domain.doi:10.1371/journal.pgen.1000355.g009
The qky-7, qky-8, and qky-9 alleles likely cause a complete loss of QKY function. All three mutations are predicted to introduce stop codons leading to shorter proteins with variable numbers of C2 domains but always lacking the two transmembrane domains (Figure 9). Thus, the three mutants likely contain truncated QKY variants that are not properly located to the membrane. Membrane localisation, however, seems essential for QKY function as all three mutants show identical phenotypes, regardless of the number of C2 domains still present in the predicted truncated proteins. In summary, the results suggest that QKY is a membrane-bound Ca2+-signaling factor.
DOQ, QKY, and ZET Contribute to SUB-Dependent Organ Development
In this work we set out to genetically identify additional components of the SUB-dependent mechanism regulating ovule development and floral morphogenesis. To this end we identified second-site mutations that resulted in a sub-like phenotype, reasoning that such mutations should reside in genes that either act directly in the SUB signaling pathway, or, that SUB and these genes affect parallel pathways that converge on a common molecular or developmental target. A defined set of criteria was applied in the identification of candidate mutants. At the developmental level, candidates were chosen that showed a sub-like ovule phenotype and twisting of petals, carpels, siliques and the stem. In addition, they were analysed for two types of sub-like cellular defects: an increased frequency of periclinal cell divisions in the L1/L2 layers of stage 3 floral meristems and a defect in root hair patterning. Mutations in three genes, DOQ, QKY, and ZET, satisfied all criteria and were considered as sub-like mutants.
Although slm mutants were phenotypically very similar there were some noteworthy distinctions. For example, doq-1 was generally characterised by a weak ovule phenotype and mild silique/stem twisting. At the same time, doq-1 showed the highest frequency of periclinal cell divisions in the L1/L2 layers of stage 3 floral meristems. Furthermore, rosette leaves of doq-1 plants had elongated petioles and narrow leaf blades, features not observed in other slm mutants. These findings indicate that DOQ is less important for ovule, silique and stem development compared to other SLM genes, but plays a more prominent role in the regulation of cell shape/division plane in the young floral meristem and leaf development. In contrast ZET does not seem to be involved in leaf development as zet mutants exhibited apparently normal leaves. This clearly distinguishes ZET from the other SLM genes, which all seem to be important for leaf development. Plants with a defect in ZET activity also showed more defects in perianth development when compared to other slm mutants. Cell separation and disintegration observed in stage 3 floral meristems of zet mutants indicated that ZET may function in cell adhesion and viability, roles that seem to be unique among SLM genes.
Since the phenotypes of sub and the other slm mutants overlap we assessed the genetic relationship among SLM genes by testing whether the expression of SUB is sensitive to SLM activity. This was apparently not the case, at least not in a relevant manner, since SUB expression was unaltered in the respective other slm mutants in our transcriptome analysis and showed only minor downregulation in slm flowers when assayed by qRT-PCR. In addition, a functional 35S::SUB transgene was unable to rescue the phenotype of doq, qky and zet mutants, including the floral defects. To further investigate the genetic interactions between SLM genes we performed a comprehensive double mutant analysis. The pleiotropic and variable phenotypes of slm single and double mutants complicated this task. Although an exaggerated phenotype was often observed in double mutants it was sometimes difficult to decide whether this represented an additive or synergistic phenotype. In addition, the various effects were often tissue dependent and thus precluded conclusions at the whole-plant level. One exception was qky-8 zet-2 where zet-2 appears to be largely epistatic to qky-8. This result indicates that ZET either acts upstream of QKY in a genetic pathway, or that ZET generally acts prior to QKY. Focusing on ovule development simplified our analysis, since a stringent set of morphological criteria could be applied. In sub-1 zet-2, sub-1 qky-8, and doq-1 zet-2 double mutants a mass of cells with integument-like structures often developed in place of true ovules. Furthermore, all double mutants, with the exception of qky-8 zet-2, often showed aberrant inner integuments, in contrast to slm single mutants that only displayed defects in the outer integument. Taken together, these synergistic effects indicate that the SLM genes contribute to similar aspects of ovule development and show that QKY and ZET function is still present in ovules of sub plants. In addition, the analysis supports the notion that DOQ may play a more peripheral role in ovule development in comparison to SUB, QKY, and ZET. The exact genetic relationship between SLM genes during this process remains to be elucidated.
The phenotypic analysis of slm single mutants suggested that SLM genes variably contribute to several common functions but it also showed that they have distinct roles, both of which depend on tissue context. To characterise the molecular mechanisms that are under control of SLM genes and to narrow down the overlap of SLM gene function, we used transcript profiling of mRNA isolated from two different tissues. Our Z-score based meta-analysis revealed 89 and 193 largely non-overlapping SUB-responsive genes in the apex and flower data sets, respectively (Table 4). That we consistently observed larger numbers of affected transcripts in the flower data set might reflect the relatively higher importance of SLM genes for later stages of flower development, but it could also be explained by differences in the nature of the samples. While the flower sample only contained tissue from two well-defined floral stages, the apex sample contained the shoot apical meristem as well as flowers from stage 1 to stage 9 and thus was much more diverse. Consequently, the sensitivity of the array to detect changes that are limited to only a few developmental stages was reduced. When comparing the number of misregulated transcripts across the various genotypes, we found that fewer genes were affected in sub mutants than in the other slm mutants. This observation suggests that DOQ, QKY and ZET play broader roles than SUB. The result might have been expected for ZET, as zet mutants show the most severely affected flowers of all slm mutants, while sub, qky and doq mutant flowers, however, are more alike and thus their morphology is less congruent with a more dramatic change at the molecular level. Meta-analysis revealed a consistent overlap in misregulated genes between sub and the other slm mutants within a given tissue and even allowed us to identify a set of core targets shared by all four SLM genes. Interestingly, these transcripts were consistently up- or downregulated in all genotypes, while the direction of the change was dependent on tissue context. These results demonstrate not only that SLM genes affect common processes with a high tissue specificity, but also that the molecular mechanisms employed by SLM genes in the different tissues are distinct. In addition, the identification of common targets that show consistent behaviour across all genotypes underlines the sensitivity and specificity of our meta-analysis.
QKY Encodes a Putative Transmembrane Protein Involved in Ca2+-Mediated Signaling
Sequence analysis of QKY suggests that QKY is the first described plant representative of the previously described MCTP family  and that it might act as a membrane-bound protein involved in a process regulated by Ca2+ and phospholipids. While the function of animal MCTPs is unknown, human MCTP2 is a membrane protein located to intracellular vesicular structures . Interestingly, the C2 domains of human MCTPs were found to bind Ca2+ with high affinity but lacked any phospholipid binding capacity . Other membrane-bound proteins with multiple C2 domains are the synaptotagmins and ferlins ,. Synaptotagmins contain two C2 domains and an amino-terminal transmembrane domain while most ferlins carry between four and six C2 domains and a carboxy-terminal transmembrane domain. Members of these two protein families function during regulated exocytosis, a process in which specific vesicles are signaled to fuse with the plasma membrane. Processes that rely on regulated exocytosis include neurotransmitter release at synapses and plasma membrane repair, a basic cellular process that mends physical injuries inflicted upon the plasma membrane –. During the repair process lysosomes  or specialised vesicles, such as the enlargosomes ,, fuse with the plasma membrane providing new membrane material and facilitating resealing.
The synaptotagmins Syts 1 and 2 are required for Ca2+-regulated synaptic vesicle exocytosis during neurotransmitter release into the synaptic cleft ,, while Syt VII promotes lysosomal exocytosis during plasma membrane repair in fibroblasts . The C. elegans ferlin FER-1 is localized to the membranes of membranous organelles (MOs) and promotes the fusion of MOs with the plasma membrane during the development of crawling spermatozoa ,. Mutations in dysferlin result in progressive muscular dystrophies – and dysferlin appears to be required for Ca2+-dependent sarcolemma resealing during membrane repair in skeletal muscle fibres ,.
Given the predicted domain topology of QKY one could speculate that QKY, and possibly other SLM genes, could participate in the control of vesicle trafficking. In principle, the predicted membrane localisation of QKY also allows for the possibility that the SUB and QKY proteins interact directly. A role of QKY in vesicle trafficking would help to explain the non-cell-autonomy of SUB function in inter-cell-layer signaling . Thus, in one possible scenario SUB could directly or indirectly regulate QKY which in turn might affect vesicular transport of factors mediating the non-cell-autonomy of SUB-dependent signaling. Further work needs to address this exciting hypothesis.
Many SLM-Responsive Genes Relate to Cell Wall Biology, Hormone, and Stress Responses
To obtain additional insights concerning the molecular processes influenced by SLM function we investigated the known and predicted functions of SLM-responsive genes. The core sets of direct or indirect SLM targets included many genes encoding proteins relating to cell wall biology and a notable enrichment of genes that are inducible by hormones, such as JA and SA, and presumed to play a role in the plant's response to various stresses. For example, a group of genes misregulated in slm mutants encode proteins with a known or predicted role in the biosynthesis and/or function of the cell wall, such as MIOX2/4, GLP6, and putative glycine-rich proteins, lectin family proteins, and chitinases. Defects in SLM-activity also resulted in a deregulation of the basal expression of JA-inducible genes with a presumed role in abiotic stress responses to cold or dehydration. The altered expression of genes responsible for camalexin and glucosinolate metabolism, secondary metabolites involved in the defense against pathogens, indicates that SLM-activity appears to play a role in biotic stress responses as well.
The notion that SLM activity somehow relates to stress is reinforced by the interesting finding that a major group of SLM-responsive genes in flowers relate to various aspects of salicylic acid (SA)-dependent signaling. SA is synthesized upon exposure of plants to abiotic stresses, such as ozone or UV-C light, and plays an important role in pathogen defense (for reviews see for example –). Three of the SLM-responsive genes are known to be required for the pathogen-induced production of SA: ICS1/SID2/EDS16 encodes an isochorismate synthase that synthesizes SA ,, EDS5/SID1 encodes a MATE family transporter potentially involved in the transport of SA intermediates , and PBS3/GDG1 encodes an adenylate-forming enzyme required for SA accumulation ,. Other SLM-dependent genes are involved in further aspects of SA signaling. EDS1 encodes a protein with a lipase-like domain , contributes to rapid SA accumulation in response to many SA-inducing stimuli, and is part of a central regulatory node of SA signaling. NPR1/NIM1 constitutes another, though SLM-independent, central regulator of SA signaling. At low SA levels NPR1 is present in the cytoplasm as a homo-multimeric complex. Increased SA levels result in the dissociation of NPR1 oligomers into monomers, which enter the nucleus where they interact with specific TGA transcription factors. These NPR1-TGA complexes regulate expression of defense genes, such as PATHOGENESIS-RELATED (PR) genes and WRKY-type transcription factor genes. Interestingly, we found NIMIN1 and GRX480, both of which act in the NPR1 pathway, to be downregulated in slm mutants. NIMIN1 encodes an interactor of NPR1 that counteracts NPR1 activity . GRX480 codes for a glutaredoxin that interacts with TGA2  and is possibly involved in cross-talk between SA and JA signaling. Another important class of regulators implied in plant immunity, the regulation of PR genes and SA-mediated signaling are WRKY transcription factors. Interestingly, we found five WRKY genes to be downregulated in slm mutants and three of them are known direct targets of NPR1: WRKY38, WRKY53, and WRKY70 . WRKY53 is a positive modulator of systemic acquired resistance (SAR)  while WRKY70 is involved in the cross-talk between SA and JA responses by acting as a negative regulator of JA-inducible genes ,. In addition, WRKY70 is required for resistance against several bacteria, fungi and an oomycete ,,,. Another SLM-responsive WRKY gene, WRKY33, is a positive regulator of resistance against the necrotrophic fungi Botrytis cinerea and Alternaria brassicicola . Fitting in the overall picture was the observation that expression of five PR genes was reduced in slm mutants, including PR1, PR2, and PR5 whose activities often undergo coordinated changes in response to various stimuli including SA . In addition, PR1 is likely a direct target of a NPR1/TGA complex ,.
Many of the SLM-responsive genes involved in SA-signaling and pathogen defense are normally transcribed at basal levels while their expression is induced several fold in the case of pathogen attack and increased levels of SA. Our transcriptome analysis of slm mutants suggests that SLM function is required for basal steady-state expression of SA-responsive genes. Thus, SLM function could be involved in priming components of the cellular defense machinery, thereby enabling floral cells to respond faster to invading pathogens. In this scenario SLM genes would contribute to basal resistance.
Another scenario is based on the hypothesis that QKY may play a role in membrane traffic. Since membrane repair or cell wall synthesis require extensive vesicle trafficking , SLM-dependent processes could be involved in the regulation of the composition of the plasma membrane or the cell wall and alterations in the two cellular compartments of slm mutants could be interpreted as wounding stress. For example, it was suggested that the SLM- and SA-responsive WALL-ASSOCIATED RECEPTOR KINASE1 (WAK1) gene transmits changes in the cell wall to the interior of the cell. In addition, WAK1 protects plant cells against negative aspects of the pathogen response . The RLK THESEUS1 (THE1) has been proposed to act as a sensor of cell wall integrity . It was shown that in the absence of proper cellulose synthesis THE1 modulates the activity of genes affecting pathogen defense, cell wall cross-linking and stress responses. Although THE1 and the SLM genes do not share common transcriptional targets, they affect a functionally related spectrum of genes. Thus, it is conceivable that one aspect of SLM function relates to the surveillance of cell wall integrity or the repair of plasma membranes.
The two proposed models are not mutually exclusive and it will be an exciting challenge to further dissect the biological function of SLM genes in development, cell biology, and stress response.
Materials and Methods
Plant Work and Genetics
Arabidopsis thaliana (L.) Heynh. var. Columbia (Col-0) and var. Landsberg (erecta mutant) (Ler) were used as wild-type strains. The sub-1 mutant was described previously . Plants were grown in a greenhouse under Philips SON-T Plus 400 Watt fluorescent bulbs on a long day cycle (16 hrs light). Dry seeds were sown on soil (Patzer Einheitserde, extra-gesiebt, Typ T, Patzer GmbH & Co. KG, Sinntal-Jossa, Germany) overlying perlite, stratified for 4 days at 4°C and then placed in the greenhouse. Plant trays were covered for 7–8 days to increase humidity and support equal germination. Ler seeds mutagenized with ethylmethane sulfonate (EMS) were obtained from Lehle Seeds (Round Rock, TX, USA). 60'000 M2 plants, corresponding to about 7'500 M1 plants, were screened for plants exhibiting a sub-like phenotype. All sub-like mutants described in this paper were outcrossed three times to Ler prior to further analysis. Two qky T-DNA insertion mutants (line SALK_140123 and SALK_043901)  were obtained from the ABRC (http://www.arabidopsis.org). The GL2::GUS line  in Ler was crossed into slm mutants for analysis of root hair specification.
Recombinant DNA Work
For DNA and RNA work standard molecular biology techniques were used . PCR-fragments used for cloning were obtained using either PfuUltra high-fidelity DNA polymerase (Stratagene) or TaKaRa PrimeSTAR HS DNA polymerase (Lonza, Basel, Switzerland). All PCR-based constructs were sequenced. Information regarding all primers used in this study is given in supporting Table S1.
Generation of the 35S::SUB Construct
In order to generate a translational fusion between SUB and the small c-myc tag SUB cDNA was amplified by PCR from plasmid H2H6T7  using primers SUB-CmycF and SUB-Cmyc-R. A 3xmyc-tag was amplified from a pFastBac-HT A plasmid (Invitrogen) containing a 2xmyc tag using primers cmyc-F and cmyc-R. Both PCR fragments were gel purified and an overlap PCR was set up using primers SUB-Cmyc-F and cmyc-R. The overlap PCR product was cloned into vector PCRII TOPO (Invitrogen) and was designated PCRII TOPO SUB:3xmyc. SUB:3xmyc was then cloned into pCAMBIA2300 (www.cambia.org). To this end pCAMBIA2300 was digested with BamHI and PmlI to remove the 35S::GUS cassette. PCRII TOPO SUB:3xmyc was digested with BamHI and BsrBI and the resulting SUB:3xmyc fragment was cloned into BamHI/PmlI-digested pCAMBIA2300 resulting in plasmid SUB:3xmyc pCAMBIA2300. To obtain the 35S promoter  the vector pART-7  was digested with NotI, blunt-ended by T4 DNA polymerase treatment and the 35S:NOS cassette was isolated by gel purification and digested with XbaI, resulting in a 35S promoter fragment with a 5′ blunt and 3′ sticky end. SUB:3xmyc pCAMBIA2300 was blunt-ended after SpeI digestion at the 5′ end and redigested with BlnI to create an XbaI-compatible 3′ end. Ligation resulted in the final 35S::SUB:myc construct (abbreviated 35S::SUB). The plasmid was verified by sequencing.
Generation of 35S::SUB slm Plants
Transformation of sub-1, doq-1, qky-8 and zet-2 mutants with the 35S::SUB construct was performed using the floral dip method  and Agrobacterium tumefaciens strain GV3101 . Transgenic T1 plants were selected on 50 µg/ml Kanamycin plates and subsequently transferred to soil for phenotypic inspection. About 41 percent of the independent 35S::SUB sub-1 T1 plants scored showed a wild-type phenotype (51/126 scored). This indicates that the c-myc tag at the carboxy-terminus of SUB may result in a reduction of SUB functionality. To assay transgene expression RNA was isolated from inflorescences using the NucleoSpin RNA II kit (Macherey-Nagel, Düren, Germany). First-strand cDNA was synthesized from 2 µg of total RNA using Moloney Murine Leukemia Virus (M-MuLV) reverse transcriptase (New England Biolabs, Frankfurt, Germany). Semi-quantitative PCR was performed using Taq DNA polymerase (New England Biolabs, Frankfurt, Germany) and a transgene-specific primer pair (sRT-SUBcmyc_F, sRT-SUBcmyc_R). Between 24 and 32 thermal cycles were tested. The GAPC gene was used as positive control .
Map-Based Cloning of QKY
To map the QKY locus at high resolution, an F2-mapping population was generated. F1 plants from qky/qky (Ler) and QKY/QKY (Col) crosses were allowed to self-pollinate, and the F2 progeny were screened for qky individuals based on twisted inflorescence morphology. DNA was isolated and used for PCR-based amplification of molecular markers. Indel polymorphism data was derived from the Monsanto Ler sequence database at TAIR . Primer sequences for indel and CAPS markers are shown in supporting Table S1. Marker amplifications from 598 mutant individuals restricted the qky map interval to 103 kb, as defined by markers F25A4(BglII) and 27.99(RsaI). Candidate genes were analysed via T-DNA insertion mutant analysis and/or sequence determination revealing that At1g74720 carried mutations in various qky alleles. To confirm qky allelic mutations, nucleotide sequences were obtained from both strands of PCR-amplified fragments. In all three EMS-induced qky alleles transitions result in stop codons. The qky-7 allele carries a G to A transition at position 2229 (genomic DNA, relative to the start ATG triplet (+1)) resulting in a shorter predicted protein (W743*). The qky-8 allele carries another G to A transition at position 2706 (W902*), and in qky-9 a C to T transition was found at position 649 (Q217*). We also determined the genomic integration sites of two T-DNA insertions in At1g74720: SALK_140123 is located at position 2576 and SALK_043901 at position 3056. Both lines exhibit a qky phenotype. SALK_140123 is predicted to carry a shorter QKY protein of 878 residues with the 29 last residues (RSHKGSHVMTPADDAGQAVLRLELTEPQR*) being encoded by the T-DNA. SALK_043901 results in a predicted shorter protein of 1022 residues with residues 1019–1022 (LFVV*) being encoded by the T-DNA.
Near full-length QKY cDNA sequence was assembled from sequences derived from four publicly available RIKEN RAFL cDNA clones . The cDNA clones partially overlapped (RAFL16-35-G10, RAFL22-02-B15, RAFL22-66-H20, RAFL22-96-F19) with one clone (RAFL22-96-F19) containing the 3′ poly(A) stretch. Additional 5′ RACE experiments  did not result in more extended 5′ cDNA sequences and comparisons of the available QKY genomic and cDNA sequences did not reveal introns.
Quantitative Real-Time PCR
Tissue was harvested as described for the whole-genome transcriptome analysis. RNA was extracted as outlined above. First-strand cDNA was synthesized from 1.5 µg of total RNA via reverse transcription, using the Transcriptor High Fidelity cDNA Synthesis Kit (Roche Diagnostics GmbH, Mannheim, Germany). Quantitative real-time PCR was performed on a Roche LightCycler using the SYBR Green I detection kit according to the manufacturer's recommendations (Roche). Amplification of UBC21/At5g25760 served as a normalization control . Primer sequences are summarized in supporting Table S1. Using the comparative Ct method, all gene expression levels were calculated relative to UBC21.
Microscopy and Art Work
Preparation and analysis of propidium iodide-stained samples for confocal laser scanning microscopy, scanning electron microscopy, and histochemical localisation of ß-glucuronidase (GUS) activity in whole-mount tissue was done essentially as described , , –. Confocal laser scanning microscopy was performed with an Olympus FV1000 setup using an inverted IX81 stand and FluoView software (FV10-ASW version 01.04.00.09) (Olympus Europa GmbH, Hamburg, Germany). After excitation at 488 nm with a multi-line argon laser, propidium iodide fluorescence (580–630 nm slit width) and autofluorescence (500–530 nm slit width) was detected. One-way scan images (scan rate 12.5 s/pixel, 512×512 pixels, Kahlman frame, average of four scans) were obtained using an Olympus 40× objective (UApo/340 40×/1.35 Oil Iris). Confocal Z-stack imaging of floral meristems was performed using 1.5 m sections. Plants or various plant organs were analysed under an Olympus SZX12 stereo microscope. Images were obtained using a ColorView III digital camera (Olympus) and Cell∧P software (version 2.4 build 1131, Olympus), saved as TIFF files, and adjusted for color and contrast using Adobe Photoshop CS3 (Adobe, San Jose, CA, USA) software.
Whole-Genome Transcriptome Analysis and Global Expression Profiling
Tissue was harvested from 21-day plants grown in continuous light at 23°C. Wild type and mutants were in the Ler background. The inflorescence apex plus flowers up to stage 9 were kept separate from later stage 10–12 flowers (the oldest 3–4 closed flower buds). The tissue was immediately frozen in liquid nitrogen and stored at −80°C. RNA was extracted with the Plant RNeasy Mini kit (Qiagen). 1 µg of total RNA was used for probe synthesis with the MessageAmp II-Biotin Enhanced kit (Ambion) according to the manufacturer's instructions. Affymetrix ATH1 GeneChips were hybridised, washed, and stained as described . The gcRMA implementation in the Bioconductor and R software packages was used for background correction, quantile normalization and expression estimate computation , (www.bioconductor.org) . Expression estimates were transformed to linear scale and the fold change (FC) was calculated for each gene in each mutant-wildtype comparison by dividing the expression estimate of the mutant condition by the value of an according control. Comparisons were done on a single replicate basis for all possible pair wise comparisons. A Z-score for each fold change was calculated by dividing the difference of log2 transformed FC and the mean of the FC population by the standard deviation of the FC population (). Genes were considered as differentially expressed if they displayed a Z-score>1 in a single comparison and an average Z-score>2 over all comparisons or, if they displayed a Z-score<−1 in a single comparison and an average Z-score over all comparisons <−2.
The QKY cDNA sequence was deposited at GenBank under the accession number FJ209045. The expression profile data have been deposited to the EMBL-EBI ArrayExpress repository under the accession E-MEXP-1592.
This list details the 14 genes that show significant changes in expression level in all Ler versus sub-like mutant (slm) comparisons. This list thus comprises the apex core target gene list.
(0.03 MB XLS)
This list details the 93 genes that show significant changes in expression level in all Ler versus sub-like mutant (slm) comparisons. This list thus comprises the flower core target gene list.
(0.07 MB XLS)
List of primers
(0.08 MB DOC)
We thank David Chevalier for initial assistance with the mutant screen, Erich Glawischnig for helpful discussions, Martin Hülskamp for the GL2::GUS line, Martin F. Yanofsky for sharing results prior to publication, and Silke Robatzek for comments on the manuscript. We also thank Silvia Blank and Sarah Hintermayer for technical assistance and members of the Schneitz and Lohmann labs for stimulating discussions.
Conceived and designed the experiments: LF MB RKY JUL KS. Performed the experiments: LF MB PV RKY WB SUA SJ. Analyzed the data: LF PV WB KS. Wrote the paper: LF JUL KS.
- 1. Steeves TA, Sussex IM (1989) Patterns in Plant Development. Cambridge: Cambridge University Press.
- 2. Satina S, Blakeslee AF, Avery AG (1940) Demonstration of the three germ layers in the shoot apex of Datura by means of induced polyploidy in periclinal chimeras. Am J Bot 27: 895–905.
- 3. Satina S (1945) Periclinal chimeras in Datura in relation to the development and structure of the ovule. Am J Bot 32: 72–81.
- 4. Satina S, Blakeslee AF (1941) Periclinal chimeras in Datura stramonium in relation to development of leaf and flower. Am J Bot 28: 862–871.
- 5. Jenik PD, Irish VF (2000) Regulation of cell proliferation patterns by homeotic genes during Arabidopsis floral development. Development 127: 1267–1276.
- 6. Szymkowiak EJ, Sussex IM (1996) What chimeras can tell us about plant development. Annu Rev Plant Physiol Plant Mol Biol 47: 351–376.
- 7. Tilney-Bassett RAE (1986) Plant chimeras. London: E. Arnold.
- 8. Gallagher KL, Benfey PN (2005) Not just another hole in the wall: understanding intercellular protein trafficking. Genes Dev 19: 189–195.
- 9. Ingram GC, Waites R (2006) Keeping it together: co-ordinating plant growth. Curr Opin Plant Biol 9: 12–20.
- 10. Ingram GC (2004) Between the sheets: inter-cell-layer communication in plant development. Philos Trans R Soc Lond B Biol Sci 359: 891–906.
- 11. Williams L, Fletcher JC (2005) Stem cell regulation in the Arabidopsis shoot apical meristem. Curr Opin Plant Biol 8: 582–586.
- 12. Savaldi-Goldstein S, Peto C, Chory J (2007) The epidermis both drives and restricts plant shoot growth. Nature 446: 199–202.
- 13. Reinhardt D, Frenz M, Mandel T, Kuhlemeier C (2003) Microsurgical and laser ablation analysis of interactions between the zones and layers of the tomato shoot apical meristem. Development 130: 4073–4083.
- 14. Conti L, Bradley D (2007) TERMINAL FLOWER1 is a mobile signal controlling Arabidopsis architecture. Plant Cell 19: 767–778.
- 15. Jackson D (2002) Double labeling of KNOTTED1 mRNA and protein reveals multiple potential sites of protein trafficking in the shoot apex. Plant Physiol 129: 1423–1429.
- 16. Lucas WJ, Bouché-Pillon S, Jackson DP, Nguyen L, Baker L, et al. (1995) Selective trafficking of KNOTTED1 homeodomain protein and its mRNA through plasmodesmata. Science 270: 1980–1983.
- 17. Perbal M-C, Haughn G, Saedler H, Schwarz-Sommer Z (1996) Non-cell-autonomous function of the Antirrhinum floral homeotic proteins DEFICIENS and GLOBOSA is exerted by their polar cell-to-cell trafficking. Development 122: 3433–3441.
- 18. Cui H, Levesque MP, Vernoux T, Jung JW, Paquette AJ, et al. (2007) An evolutionarily conserved mechanism delimiting SHR movement defines a single layer of endodermis in plants. Science 316: 421–425.
- 19. Nakajima K, Sena G, Nawy T, Benfey PN (2001) Intercellular movement of the putative transcription factor SHR in root patterning. Nature 413: 307–311.
- 20. Becraft PW, Stinard PS, McCarty DR (1996) CRINCLY4: a TNFR-like receptor kinase involved in maize epidermal differentiation. Science 273: 1406–1409.
- 21. Becraft PW, Kang SH, Suh SG (2001) The maize CRINKLY4 receptor kinase controls a cell-autonomous differentiation response. Plant Physiol 127: 486–496.
- 22. Gifford ML, Dean S, Ingram GC (2003) The Arabidopsis ACR4 gene plays a role in cell layer organisation during ovule integument and sepal margin development. Development 130: 4249–4258.
- 23. Tanaka H, Watanabe M, Sasabe M, Hiroe T, Tanaka T, et al. (2007) Novel receptor-like kinase ALE2 controls shoot development by specifying epidermis in Arabidopsis. Development 134: 1643–1652.
- 24. Watanabe M, Tanaka H, Watanabe D, Machida C, Machida Y (2004) The ACR4 receptor-like kinase is required for surface formation of epidermis-related tissues in Arabidopsis thaliana. Plant J 39: 298–308.
- 25. Chevalier D, Batoux M, Fulton L, Pfister K, Yadav RK, et al. (2005) STRUBBELIG defines a receptor kinase-mediated signaling pathway regulating organ development in Arabidopsis. Proc Natl Acad Sci USA 102: 9074–9079.
- 26. Kwak SH, Schiefelbein J (2007) The role of the SCRAMBLED receptor-like kinase in patterning the Arabidopsis root epidermis. Dev Biol 302: 118–131.
- 27. Kwak SH, Shen R, Schiefelbein J (2005) Positional signaling mediated by a receptor-like kinase in Arabidopsis. Science 307: 1111–1113.
- 28. Yadav RK, Fulton L, Batoux M, Schneitz K (2008) The Arabidopsis receptor-like kinase STRUBBELIG mediates inter-cell-layer signaling during floral development. Dev Biol 323: 261–270.
- 29. Eyüboglu B, Pfister K, Haberer G, Chevalier D, Fuchs A, et al. (2007) Molecular characterisation of the STRUBBELIG-RECEPTOR FAMILY of genes encoding putative leucine-rich repeat receptor-like kinases in Arabidopsis thaliana. BMC Plant Biol 7: 16.
- 30. Shiu S-H, Bleecker AB (2001) Receptor-like kinases from Arabidopsis form a monophyletic gene family related to animal receptor kinases. Proc Natl Acad Sci USA 98: 10763–10768.
- 31. Castells E, Casacuberta JM (2007) Signalling through kinase-defective domains: the prevalence of atypical receptor-like kinases in plants. J Exp Bot.
- 32. Kroiher M, Miller MA, Steele RE (2001) Deceiving appearances: signaling by “dead” and “fractured” receptor protein-tyrosine kinases. BioEssays 23: 69–76.
- 33. Masucci JD, Rerie WG, Foreman DR, Zhang M, Galway ME, et al. (1996) The homeobox gene GLABRA 2 is required for position-dependent cell differentiation in the root epidermis of Arabidopsis thaliana. Development 122: 1253–1260.
- 34. Schmid M, Davison TS, Henz SR, Pape UJ, Demar M, et al. (2005) A gene expression map of Arabidopsis thaliana development. Nat Genet 37: 501–506.
- 35. Kanter U, Usadel B, Guerineau F, Li Y, Pauly M, et al. (2005) The inositol oxygenase gene family of Arabidopsis is involved in the biosynthesis of nucleotide sugar precursors for cell-wall matrix polysaccharides. Planta 221: 243–254.
- 36. Carter C, Graham RA, Thornburg RW (1998) Arabidopsis thaliana contains a large family of germin-like proteins: characterization of cDNA and genomic sequences encoding 12 unique family members. Plant Mol Biol 38: 929–943.
- 37. Aufsatz W, Amry D, Grimm C (1998) The ECS1 gene of Arabidopsis encodes a plant cell wall-associated protein and is potentially linked to a locus influencing resistance to Xanthomonas campestris. Plant Mol Biol 38: 965–976.
- 38. Chini A, Fonseca S, Fernandez G, Adie B, Chico JM, et al. (2007) The JAZ family of repressors is the missing link in jasmonate signalling. Nature 448: 666–671.
- 39. Thines B, Katsir L, Melotto M, Niu Y, Mandaokar A, et al. (2007) JAZ repressor proteins are targets of the SCF(COI1) complex during jasmonate signalling. Nature 448: 661–665.
- 40. Kiyosue T, Yoshiba Y, Yamaguchi-Shinozaki K, Shinozaki K (1996) A nuclear gene encoding mitochondrial proline dehydrogenase, an enzyme involved in proline metabolism, is upregulated by proline but downregulated by dehydration in Arabidopsis. Plant Cell 8: 1323–1335.
- 41. Mowla SB, Cuypers A, Driscoll SP, Kiddle G, Thomson J, et al. (2006) Yeast complementation reveals a role for an Arabidopsis thaliana late embryogenesis abundant (LEA)-like protein in oxidative stress tolerance. Plant J 48: 743–756.
- 42. Weaver LM, Gan S, Quirino B, Amasino RM (1998) A comparison of the expression patterns of several senescence-associated genes in response to stress and hormone treatment. Plant Mol Biol 37: 455–469.
- 43. Leyman B, Van Dijck P, Thevelein JM (2001) An unexpected plethora of trehalose biosynthesis genes in Arabidopsis thaliana. Trends Plant Sci 6: 510–513.
- 44. Ramon M, Rolland F (2007) Plant development: introducing trehalose metabolism. Trends Plant Sci 12: 185–188.
- 45. Chary SN, Hicks GR, Choi YG, Carter D, Raikhel NV (2008) Trehalose-6-phosphate synthase/phosphatase regulates cell shape and plant architecture in Arabidopsis. Plant Physiol 146: 97–107.
- 46. Goyal K, Walton LJ, Tunnacliffe A (2005) LEA proteins prevent protein aggregation due to water stress. Biochem J 388: 151–157.
- 47. Grennan AK (2007) The role of trehalose biosynthesis in plants. Plant Physiol 144: 3–5.
- 48. Nakashima K, Satoh R, Kiyosue T, Yamaguchi-Shinozaki K, Shinozaki K (1998) A gene encoding proline dehydrogenase is not only induced by proline and hypoosmolarity, but is also developmentally regulated in the reproductive organs of Arabidopsis. Plant Physiol 118: 1233–1241.
- 49. Peng Z, Lu Q, Verma DPSReciprocal regulation of Δ1-pyrroline-5-carboxylate synthetase and proline dehydrogenase genes controls proline levels during and after osmotic stress in plants. Mol Gen Genet 253: 334–341.
- 50. Verbruggen N, Hua XJ, May M, Van Montagu M (1996) Environmental and developmental signals modulate proline homeostasis: evidence for a negative transcriptional regulator. Proc Natl Acad Sci U S A 93: 8787–8791.
- 51. Grubb CD, Abel S (2006) Glucosinolate metabolism and its control. Trends Plant Sci 11: 89–100.
- 52. Yan X, Chen S (2007) Regulation of plant glucosinolate metabolism. Planta 226: 1343–1352.
- 53. Bartel B, Fink GR (1995) ILR1, an amidohydrolase that releases active indole-3-acetic acid from conjugates. Science 268: 1745–1748.
- 54. LeClere S, Tellez R, Rampey RA, Matsuda SP, Bartel B (2002) Characterization of a family of IAA-amino acid conjugate hydrolases from Arabidopsis. J Biol Chem 277: 20446–20452.
- 55. Barlier I, Kowalczyk M, Marchant A, Ljung K, Bhalerao R, et al. (2000) The SUR2 gene of Arabidopsis thaliana encodes the cytochrome P450 CYP83B1, a modulator of auxin homeostasis. Proc Natl Acad Sci U S A 97: 14819–14824.
- 56. Bak S, Tax FE, Feldmann KA, Galbraith DW, Feyereisen R (2001) CYP83B1, a cytochrome P450 at the metabolic branch point in auxin and indole glucosinolate biosynthesis in Arabidopsis. Plant Cell 13: 101–111.
- 57. Hull AK, Vij R, Celenza JL (2000) Arabidopsis cytochrome P450s that catalyze the first step of tryptophan-dependent indole-3-acetic acid biosynthesis. Proc Natl Acad Sci U S A 97: 2379–2384.
- 58. Mikkelsen MD, Hansen CH, Wittstock U, Halkier BA (2000) Cytochrome P450 CYP79B2 from Arabidopsis catalyzes the conversion of tryptophan to indole-3-acetaldoxime, a precursor of indole glucosinolates and indole-3-acetic acid. J Biol Chem 275: 33712–33717.
- 59. Glawischnig E, Hansen BG, Olsen CE, Halkier BA (2004) Camalexin is synthesized from indole-3-acetaldoxime, a key branching point between primary and secondary metabolism in Arabidopsis. Proc Natl Acad Sci U S A 101: 8245–8250.
- 60. Glazebrook J, Ausubel FM (1994) Isolation of phytoalexin-deficient mutants of Arabidopsis thaliana and characterization of their interactions with bacterial pathogens. Proc Natl Acad Sci U S A 91: 8955–8959.
- 61. Nafisi M, Goregaoker S, Botanga CJ, Glawischnig E, Olsen CE, et al. (2007) Arabidopsis cytochrome P450 monooxygenase 71A13 catalyzes the conversion of indole-3-acetaldoxime in camalexin synthesis. Plant Cell 19: 2039–2052.
- 62. Schuhegger R, Nafisi M, Mansourova M, Petersen BL, Olsen CE, et al. (2006) CYP71B15 (PAD3) catalyzes the final step in camalexin biosynthesis. Plant Physiol 141: 1248–1254.
- 63. Zhou N, Tootle TL, Glazebrook J (1999) Arabidopsis PAD3, a gene required for camalexin biosynthesis, encodes a putative cytochrome P450 monooxygenase. Plant Cell 11: 2419–2428.
- 64. Finn RD, Tate J, Mistry J, Coggill PC, Sammut SJ, et al. (2008) The Pfam protein families database. Nucleic Acids Res 36: D281–8.
- 65. Shin OH, Han W, Wang Y, Südhof TC (2005) Evolutionarily conserved multiple C2 domain proteins with two transmembrane regions (MCTPs) and unusual Ca2+ binding properties. J Biol Chem 280: 1641–1651.
- 66. Maeda I, Kohara Y, Yamamoto M, Sugimoto A (2001) Large-scale analysis of gene function in Caenorhabditis elegans by high-throughput RNAi. Curr Biol 11: 171–176.
- 67. Bai J, Chapman ER (2004) The C2 domains of synaptotagmin–partners in exocytosis. Trends Biochem Sci 29: 143–151.
- 68. Rizo J, Südhof TC (1998) C2-domains, structure and function of a universal Ca2+-binding domain. J Biol Chem 273: 15879–15882.
- 69. Sondermann H, Kuriyan J (2005) C2 can do it, too. Cell 121: 158–160.
- 70. Südhof TC (2002) Synaptotagmins: why so many? J Biol Chem 277: 7629–7632.
- 71. Min SW, Chang WP, Südhof TC (2007) E-Syts, a family of membranous Ca2+-sensor proteins with multiple C2 domains. Proc Natl Acad Sci U S A 104: 3823–3828.
- 72. Bansal D, Campbell KP (2004) Dysferlin and the plasma membrane repair in muscular dystrophy. Trends Cell Biol 14: 206–213.
- 73. Andrews NW (2002) Lysosomes and the plasma membrane: trypanosomes reveal a secret relationship. J Cell Biol 158: 389–394.
- 74. McNeil PL, Steinhardt RA (2003) Plasma membrane disruption: repair, prevention, adaptation. Annu Rev Cell Dev Biol 19: 697–731.
- 75. Chieregatti E, Meldolesi J (2005) Regulated exocytosis: new organelles for non-secretory purposes. Nat Rev Mol Cell Biol 6: 181–187.
- 76. Reddy A, Caler EV, Andrews NW (2001) Plasma membrane repair is mediated by Ca(2+)-regulated exocytosis of lysosomes. Cell 106: 157–169.
- 77. Borgonovo B, Cocucci E, Racchetti G, Podini P, Bachi A, et al. (2002) Regulated exocytosis: a novel, widely expressed system. Nat Cell Biol 4: 955–962.
- 78. Cocucci E, Racchetti G, Podini P, Meldolesi J (2007) Enlargeosome traffic: exocytosis triggered by various signals is followed by endocytosis, membrane shedding or both. Traffic 8: 742–757.
- 79. Geppert M, Goda Y, Hammer RE, Li C, Rosahl TW, et al. (1994) Synaptotagmin I: a major Ca2+ sensor for transmitter release at a central synapse. Cell 79: 717–727.
- 80. Fernández-Chacón R, Königstorfer A, Gerber SH, García J, Matos MF, et al. (2001) Synaptotagmin I functions as a calcium regulator of release probability. Nature 410: 41–49.
- 81. Achanzar WE, Ward S (1997) A nematode gene required for sperm vesicle fusion. J Cell Sci 110: 1073–1081.
- 82. Washington NL, Ward S (2006) FER-1 regulates Ca2+ -mediated membrane fusion during C. elegans spermatogenesis. J Cell Sci 119: 2552–2562.
- 83. Bashir R, Britton S, Strachan T, Keers S, Vafiadaki E, et al. (1998) A gene related to Caenorhabditis elegans spermatogenesis factor fer-1 is mutated in limb-girdle muscular dystrophy type 2B. Nat Genet 20: 37–42.
- 84. Bansal D, Miyake K, Vogel SS, Groh S, Chen CC, et al. (2003) Defective membrane repair in dysferlin-deficient muscular dystrophy. Nature 423: 168–172.
- 85. Liu J, Aoki M, Illa I, Wu C, Fardeau M, et al. (1998) Dysferlin, a novel skeletal muscle gene, is mutated in Miyoshi myopathy and limb girdle muscular dystrophy. Nat Genet 20: 31–36.
- 86. Lennon NJ, Kho A, Bacskai BJ, Perlmutter SL, Hyman BT, et al. (2003) Dysferlin interacts with annexins A1 and A2 and mediates sarcolemmal wound-healing. J Biol Chem 278: 50466–50473.
- 87. Dong X (2005) NPR1, all things considered. Curr Opin Plant Biol 7: 547–552.
- 88. Eulgem T, Somssich IE (2007) Networks of WRKY transcription factors in defense signaling. Curr Opin Plant Biol 10: 366–371.
- 89. Glazebrook J (2005) Contrasting mechanisms of defense against biotrophic and necrotrophic pathogens. Annu Rev Phytopathol 43: 205–227.
- 90. Jones JD, Dangl JL (2006) The plant immune system. Nature 444: 323–329.
- 91. Wiermer M, Feys BJ, Parker JE (2005) Plant immunity: the EDS1 regulatory node. Curr Opin Plant Biol 8: 383–389.
- 92. Strawn MA, Marr SK, Inoue K, Inada N, Zubieta C, et al. (2007) Arabidopsis isochorismate synthase functional in pathogen-induced salicylate biosynthesis exhibits properties consistent with a role in diverse stress responses. J Biol Chem 282: 5919–5933.
- 93. Wildermuth MC, Dewdney J, Wu G, Ausubel FM (2001) Isochorismate synthase is required to synthesize salicylic acid for plant defence. Nature 414: 562–565.
- 94. Nawrath C, Heck S, Parinthawong N, Metraux JP (2002) EDS5, an essential component of salicylic acid-dependent signaling for disease resistance in Arabidopsis, is a member of the MATE transporter family. Plant Cell 14: 275–286.
- 95. Jagadeeswaran G, Raina S, Acharya BR, Maqbool SB, Mosher SL, et al. (2007) Arabidopsis GH3-LIKE DEFENSE GENE 1 is required for accumulation of salicylic acid, activation of defense responses and resistance to Pseudomonas syringae. Plant J 51: 234–246.
- 96. Nobuta K, Okrent RA, Stoutemyer M, Rodibaugh N, Kempema L, et al. (2007) The GH3 acyl adenylase family member PBS3 regulates salicylic acid-dependent defense responses in Arabidopsis. Plant Physiol 144: 1144–1156.
- 97. Falk A, Feys BJ, Frost LN, Jones JD, Daniels MJ, et al. (1999) EDS1, an essential component of R gene-mediated disease resistance in Arabidopsis has homology to eukaryotic lipases. Proc Natl Acad Sci U S A 96: 3292–3297.
- 98. Weigel RR, Pfitzner UM, Gatz C (2005) Interaction of NIMIN1 with NPR1 modulates PR gene expression in Arabidopsis. Plant Cell 17: 1279–1291.
- 99. Ndamukong I, Abdallat AA, Thurow C, Fode B, Zander M, et al. (2007) SA-inducible Arabidopsis glutaredoxin interacts with TGA factors and suppresses JA-responsive PDF1.2 transcription. Plant J 50: 128–139.
- 100. Wang D, Amornsiripanitch N, Dong X (2006) A genomic approach to identify regulatory nodes in the transcriptional network of systemic acquired resistance in plants. PLoS Pathog 2: e123.
- 101. Li J, Brader G, Kariola T, Tapio Palva E (2006) WRKY70 modulates the selection of signaling pathways in plant defense. Plant J 46: 477–491.
- 102. Li J, Brader G, Tapio Palva E (2004) The WRKY70 transcription factor: a node of convergence for jasmonate-mediated and salicylate-mediated signals in plant defense. Plant Cell 16: 319–331.
- 103. AbuQamar S, Chen X, Dhawan R, Bluhm B, Salmeron J, et al. (2006) Expression profiling and mutant analysis reveals complex regulatory networks involved in Arabidopsis response to Botrytis infection. Plant J 48: 28–44.
- 104. Knoth C, Ringler J, Dangl JL, Eulgem T (2007) Arabidopsis WRKY70 is required for full RPP4-mediated disease resistance and basal defense against Hyaloperonospora parasitica. Mol Plant Microbe Interact 20: 120–128.
- 105. Zheng Z, Qamar SA, Chen Z, Mengiste T (2006) Arabidopsis WRKY33 transcription factor is required for resistance to necrotrophic fungal pathogens. Plant J 48: 592–605.
- 106. Uknes S, Mauch-Mani B, Moyer M, Potter S, Williams S, et al. (1992) Acquired resistance in Arabidopsis. Plant Cell 4: 645–656.
- 107. Johnson C, Boden E, Arias J (2003) Salicylic acid and NPR1 induce the recruitment of trans-activating TGA factors to a defense gene promoter in Arabidopsis. Plant Cell 15: 1846–1858.
- 108. Zhang Y, Tessaro MJ, Lassner M, Li X (2003) Knockout analysis of Arabidopsis transcription factors TGA2, TGA5, and TGA6 reveals their redundant and essential roles in systemic acquired resistance. Plant Cell 15: 2647–2653.
- 109. Lerouxel O, Cavalier DM, Liepman AH, Keegstra K (2006) Biosynthesis of plant cell wall polysaccharides - a complex process. Curr Opin Plant Biol 9: 621–630.
- 110. He Z-H, He D, Kohorn BD (1998) Requirement for the induced expression of a cell wall associated receptor kinase for survival during the pathogen response. Plant J 14: 55–63.
- 111. Hématy K, Sado PE, Van Tuinen A, Rochange S, Desnos T, et al. (2007) A receptor-like kinase mediates the response of Arabidopsis cells to the inhibition of cellulose synthesis. Curr Biol 17: 922–931.
- 112. Alonso JM, Stepanova AN, Leisse TJ, Kim CJ, Chen H, et al. (2003) Genome-wide insertional mutagenesis of Arabidopsis thaliana. Science 301: 653–657.
- 113. Sambrook J, Fritsch EF, Maniatis T (1989) Molecular Cloning. Plainview, NY: Cold Spring Harbor Laboratory Press.
- 114. Odell JT, Nagy F, Chua NH (1985) Identification of DNA sequences required for activity of the cauliflower mosaic virus 35S promoter. Nature 313: 810–812.
- 115. Gleave AP (1992) A versatile binary vector system with a T-DNA organisational structure conducive to efficient integration of cloned DNA into the plant genome. Plant Molecular Biology 20: 1203–1207.
- 116. Clough SJ, Bent AF (1998) Floral dip: a simplified method for Agrobacterium-mediated transformation of Arabidopsis thaliana. Plant J 16: 735–743.
- 117. Koncz C, Schell J (1986) The promoter of TL-DNA gene 5 controls the tissue-specific expression of chimaeric genes carried by a novel Agrobacterium binary vector. Mol Gen Genet 204: 383–396.
- 118. Shih MC, Heinrich P, Goodmann HM (1991) Cloning and chromosomal mapping of nuclear genes encoding chloroplast and cytosolic glyceraldehyde-3-phosphate dehydrogenase from Arabidopsis thaliana. Gene 104: 133–138.
- 119. Jander G, Norris SR, Rounsley SD, Bush DF, Levin IM, et al. (2002) Arabidopsis map-based cloning in the post-genome era. Plant Physiol 129: 440–450.
- 120. Seki M, Narusaka M, Kamiya A, Ishida J, Satou M, et al. (2002) Functional annotation of a full-length Arabidopsis cDNA collection. Science 296: 141–145.
- 121. Frohman MA, Dush MK, Martin GR (1988) Rapid production of full-length cDNAs from rare transcripts-amplification using a single gene-specific oligonucleotide primer. Proc Natl Acad Sci USA 85: 8998–9002.
- 122. Krogh A, Larsson B, von Heijne G, Sonnhammer EL (2001) Predicting transmembrane protein topology with a hidden Markov model: application to complete genomes. J Mol Biol 305: 567–580.
- 123. Czechowski T, Stitt M, Altmann T, Udvardi MK, Scheible WR (2005) Genome-wide identification and testing of superior reference genes for transcript normalization in Arabidopsis. Plant Physiol 139: 5–17.
- 124. Clark SE, Running MP, Meyerowitz EM (1993) Clavata1, a regulator of meristem and flower development in Arabidopsis. Development 119: 397–418.
- 125. Jefferson RA, Burgess SM, Hirsch D (1986) ß-glucuronidase from Escherichia coli as a gene fusion marker. Proc Natl Acad Sci USA 83: 8447–8451.
- 126. Schneitz K, Hülskamp M, Pruitt RE (1995) Wild-type ovule development in Arabidopsis thaliana: a light microscope study of cleared whole-mount tissue. Plant J 7: 731–749.
- 127. Schneitz K, Hülskamp M, Kopczak SD, Pruitt RE (1997) Dissection of sexual organ ontogenesis: a genetic analysis of ovule development in Arabidopsis thaliana. Development 124: 1367–1376.
- 128. Sieburth LE, Meyerowitz EM (1997) Molecular dissection of the AGAMOUS control region shows that cis elements for spatial regulation are located intragenically. Plant Cell 9: 355–365.
- 129. Schmid M, Uhlenhaut NH, Godard F, Demar M, Bressan R, et al. (2003) Dissection of floral induction pathways using global expression analysis. Development 130: 6001–6012.
- 130. Gentleman RC, Carey VJ, Bates DM, Bolstad B, Dettling M, et al. (2004) Bioconductor: open software development for computational biology and bioinformatics. Genome Biol 5: R80.
- 131. Wu Z, Irizarry RA, Gentleman RC, Murillo FM, Spencer F (2004) A model based background adjustment for oligonucleotide expression arrays. Johns Hopkins University, Dept of Biostatistics Working Papers Working Paper 1:.
- 132. Team RDC (2007) R: a language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing.