In both animals and plants, many developmentally important regulatory genes have complementary microRNAs (miRNAs), which suggests that these miRNAs constitute a class of developmental signalling molecules. Leaves of higher plants exhibit a varying degree of asymmetry along the adaxial/abaxial (upper/lower) axis. This asymmetry is specified through the polarized expression of class III homeodomain/leucine zipper (HD-ZIPIII) genes. In Arabidopsis, three such genes, PHABULOSA (PHB), PHAVOLUTA (PHV) and REVOLUTA (REV), are expressed throughout the incipient leaf, but become adaxially localized after primordium emergence. Downregulation of the HD-ZIPIII genes allows expression of the KANADI and YABBY genes, which specify abaxial fate. PHB, PHV and REV transcripts contain a complementary site for miRNA165 and miRNA166, which can direct their cleavage in vitro. Here we show that miRNA166 constitutes a highly conserved polarizing signal whose expression pattern spatially defines the expression domain of the maize hd-zipIII family member rolled leaf1 (rld1). Moreover, the progressively expanding expression pattern of miRNA166 during leaf development and its accumulation in phloem suggests that miRNA166 may form a movable signal that emanates from a signalling centre below the incipient leaf.

Small RNAs are important regulators of gene expression. In maize, adaxial/abaxial (dorsoventral) leaf polarity is established by an abaxial gradient of microRNA166 (miR166), which spatially restricts the expression domain of class III homeodomain leucine zipper (HD-ZIPIII) transcription factors that specify adaxial/upper fate. Here, we show that leafbladeless1 encodes a key component in the trans-acting small interfering RNA (ta-siRNA) biogenesis pathway that acts on the adaxial side of developing leaves and demarcates the domains of hd-zipIII and miR166 accumulation. Our findings indicate that tasiR-ARF, a ta-siRNA, and miR166 establish opposing domains along the adaxial-abaxial axis, thus revealing a novel mechanism of pattern formation.Supplemental material is available at www.genesdev.org.Received January 8, 2007; revised version accepted February 20, 2007. In both animals and plants, many developmentally important regulatory genes are predicted targets of microRNAs (miRNAs), which suggests that such small RNAs constitute a class of developmental determinants (Alvarez-Garcia and Miska 2005;Jones-Rhoades et al. 2006). Patterning and outgrowth of lateral organs in plants depend on the specification of adaxial/abaxial (dorsoventral) polarity in the incipient primordium. This asymmetry is established through the polarized expression of class III homeodomain leucine zipper (HD-ZIPIII) transcription factors that specify adaxial/upper cell fate (McConnell et al. 2001;Emery et al. 2003;Juarez et al. 2004a). The adaxial-specific expression of hd-zipIII family members is delineated by the expression pattern of a 21-nucleotide (nt) miRNA, miR166, which directs the cleavage of hd-zipIII transcripts (Juarez et al. 2004a;Kidner and Martienssen 2004). In maize, miR166 accumulates most abundantly immediately below the incipient leaf, but a gradient of miR166 extends into the abaxial side of the initiating organ that establishes organ polarity (Juarez et al. 2004a).Specification of adaxial/abaxial organ polarity in maize also requires the activity of leafbladeless1 (lbl1). Recessive mutations in lbl1 lead to a variable abaxialization of leaves (Timmermans et al. 1998). The weak lbl1-ref allele causes a partial loss of adaxial identity revealed as patches of abaxial cells on the upper leaf surface, whereas leaves of the severe ragged seedling1 allele (lbl1-rgd1) are often radially symmetric and completely abaxialized (Fig. 1A). Expression of the hd-zipIII family member rld1 is reduced in lbl1 mutants. Conversely, increased levels of hd-zipIII expression in Rld1-O mutants, which carry a miR166-insensitive allele of rld1, can fully suppress the vegetative defects of lbl1 (Juarez et al. 2004b). lbl1 thus contributes to organ polarity by regulating the accumulation of rld1 transcripts on the adaxial side of the developing leaf.Here, we show that lbl1 encodes a homolog of SUPPRESSOR-OF-GENE-SILENCING3 (SGS3), which is specifically required for the biogenesis of trans-acting small interfering RNAs (ta-siRNAs) (Peragine et al. 2...

Dorsoventral (adaxial/abaxial) polarity of the maize leaf is established in the meristem and is maintained throughout organ development to coordinate proper outgrowth and patterning of the leaf. rolled leaf1(rld1) and leafbladeless1 (lbl1) are required for the specification of the adaxial/upper leaf surface. rld1 encodes a class III homeodomain-leucine zipper (HD-ZIPIII) protein whose adaxial expression is spatially defined by miRNA166-directed transcript cleavage on the abaxial side. The semi-dominant Rld1-Original (Rld1-O)mutation, which results from a single nucleotide substitution in the miRNA166 complementary site, leads to persistent expression of mutant transcripts on the abaxial site. This causes the adaxialization or partial reversal of leaf polarity. By contrast, recessive mutations in lbl1 cause the formation of abaxialized leaves. The lbl1 and Rld1-Omutations mutually suppress each other, indicating that these two genes act in the same genetic pathway. Adaxial and meristematic expression of rld1is reduced in lbl1 mutants, indicating that lbl1 acts upstream of rld1 to specify adaxial fate during primordium development. However, rld1 expression in the vasculature of lbl1 is normal, suggesting that the specification of adaxial/abaxial polarity during vascular and primordia development is governed by separate but overlapping pathways. We also show that members of the maize yabbygene family are expressed on the adaxial side of incipient and developing leaf primordia. This expression pattern is unlike that observed in Arabidopsis, where YABBY expression is correlated with abaxial cell fate. The yabby expression patterns in lbl1 and Rld1-O mutants suggest that the yabby genes act downstream in the same pathway as lbl1 and rld1. Moreover, our observations suggest that maize yabby genes may direct lateral organ outgrowth rather than determine cell fate. We propose that a single genetic pathway involving lbl1, rld1 and the yabby genes integrates positional information within the SAM, and leads to adaxial/abaxial patterning and mediolateral outgrowth of the leaf.

Nature © Macmillan Publishers Ltd 1998 8 letters to nature 878 NATURE | VOL 394 | 27 AUGUST 1998reflection does not vary significantly along-axis 9 , so we use a waveform inversion method that assumes that the layers are horizontally stratified. The inversion scheme is implemented in intercept time-slowness domain 23 . This transformation also allows a clear representation of the AMC and converted arrival (P melt S) to be seen (Fig. 3) without interference of slow-phase-velocity events such as the sea-floor reflection. The waveform inversion method determines the P-and S-wave velocities of the crust by improving the fit between observed data and synthetically calculated data. It is an automated method, and therefore reduces human bias and also provides error estimates on the final solution 6,23 . The large-scale initial P-wave velocity was taken from Tolstoy et al. 9 . The initial S-wave velocity was obtained from the P-wave velocity assuming a Poisson's ratio of 0.26 (ref. 24). We used a P-wave attenuation coefficient of 16 for the first 200 m, and 40-90 for the crust below 24 . The S-wave attenuation coefficient was half of the P-wave attenuation coefficient. The model consists of a stack of 8-m-thick layers, for which the P-wave and S-wave velocities, density and attenuation are defined. This sampling interval was based on the minimum thickness that can be resolved from the observed data, which has a frequency bandwidth of 5-30 Hz. As data from all the slownesses were inverted simultaneously, the inverted results contain a model which is consistent with the data from all slownesses, and is therefore less likely to be influenced by incoherent noise due to two-and three-dimensional effects.The turning rays above the sill and the P-wave reflection arrivals from the AMC constrain the P-wave and S-wave velocities above the AMC. Further constraint on S-wave velocity structure above the AMC comes from the arrival time of the P melt S phase. For this phase, the P-to-S conversion occurs at the solid-fluid interface of the AMC and at the sea bed such that the wave travels one leg between the sea bed and AMC as a P-wave, and the other as an S-wave. The waveforms of the AMC reflection and the P melt S phase constrain P-wave and S-wave velocities within the AMC and its neighbourhood. However, as with all seismic techniques, we have to guard against non-uniqueness of the final solution. To gain confidence in our results, we altered individual features within various candidate models while keeping the rest of the model fixed, and re-ran the inversion; the models that gave the best fit with the data are presented here. Stacking. To enhance the partial-stack images, we firstly applied a frequencywavenumber filter to the CMP gathers to remove sea-floor contamination of the AMC events, then performed a normal-moveout correction of 2.10 and 1.85 km s −1 for the P-wave and S-wave image, respectively. The partial stacks were obtained by stacking data between 2 and 3 km offsets, and all other processing parameters were as for the ϳ...

Leaf senescence is an active process involving remobilization of nutrients from senescing leaves to other parts of the plant. Whereas senescence is accompanied by a decline in leaf cytokinin content, supplemental cytokinin delays senescence. Plants that overexpress isopentenyl transferase (ipt), a cytokinin-producing gene, or knotted1 (kn1), a homeobox gene, have many phenotypes in common. Many of these phenotypes are characteristic of altered cytokinin physiology. The effect of kn1 on leaf senescence was tested by driving its expression using the promoter of the senescence-associated gene SAG12. SAG:kn1 tobacco plants showed a marked delay in leaf senescence but otherwise developed normally. The delay in senescence was revealed by an increase in chlorophyll content in SAG:kn1 leaves relative to leaves of the control plants and by a decrease in the number of dead leaves. Senescence was also delayed in detached leaves of SAG:kn1 plants. Delayed senescence was accompanied by increased leaf cytokinin content in older leaves expressing kn1. These experiments extend the current understanding of kn1 function and suggest that in addition to mediating meristem maintenance, kn1 is capable of regulating the onset of senescence in leaves.

Wounds in Drosophila and mouse embryos induce similar genetic pathways to repair epidermal barriers. However, the transcription factors that transduce wound signals to repair epidermal barriers are largely unknown. We characterize the transcriptional regulatory enhancers of 4 genes— Ddc , ple , msn , and kkv —that are rapidly activated in epidermal cells surrounding wounds in late Drosophila embryos and early larvae. These epidermal wound enhancers all contain evolutionarily conserved sequences matching binding sites for JUN/FOS and GRH transcription factors, but vary widely in trans - and cis -requirements for these inputs and their binding sites. We propose that the combination of GRH and FOS is part of an ancient wound–response pathway still used in vertebrates and invertebrates, but that other mechanisms have evolved that result in similar transcriptional output. A common, but largely untested assumption of bioinformatic analyses of gene regulatory networks is that transcription units activated in the same spatial and temporal patterns will require the same cis -regulatory codes. Our results indicate that this is an overly simplistic view.

The epidermis is the largest organ of the body for most animals, and the first line of defense against invading pathogens. A breach in the epidermal cell layer triggers a variety of localized responses that in favorable circumstances result in the repair of the wound. Many cellular and genetic responses must be limited to epidermal cells that are close to wounds, but how this is regulated is still poorly understood. The order and hierarchy of epidermal wound signaling factors are also still obscure. The Drosophila embryonic epidermis provides an excellent system to study genes that regulate wound healing processes. We have developed a variety of fluorescent reporters that provide a visible readout of wound-dependent transcriptional activation near epidermal wound sites. A large screen for mutants that alter the activity of these wound reporters has identified seven new genes required to activate or delimit wound-induced transcriptional responses to a narrow zone of cells surrounding wound sites. Among the genes required to delimit the spread of wound responses are Drosophila Flotillin-2 and Src42A, both of which are transcriptionally activated around wound sites. Flotillin-2 and constitutively active Src42A are also sufficient, when overexpressed at high levels, to inhibit wound-induced transcription in epidermal cells. One gene required to activate epidermal wound reporters encodes Dual oxidase, an enzyme that produces hydrogen peroxide. We also find that four biochemical treatments (a serine protease, a Src kinase inhibitor, methyl-ß-cyclodextrin, and hydrogen peroxide) are sufficient to globally activate epidermal wound response genes in Drosophila embryos. We explore the epistatic relationships among the factors that induce or delimit the spread of epidermal wound signals. Our results define new genetic functions that interact to instruct only a limited number of cells around puncture wounds to mount a transcriptional response, mediating local repair and regeneration.

After injury to the animal epidermis, a variety of genes are transcriptionally activated in nearby cells to regenerate the missing cells and facilitate barrier repair. The range and types of diffusible wound signals that are produced by damaged epidermis and function to activate repair genes during epidermal regeneration remains a subject of very active study in many animals. In Drosophila embryos, we have discovered that serine protease function is locally activated around wound sites, and is also required for localized activation of epidermal repair genes. The serine protease trypsin is sufficient to induce a striking global epidermal wound response without inflicting cell death or compromising the integrity of the epithelial barrier. We developed a trypsin wounding treatment as an amplification tool to more fully understand the changes in the Drosophila transcriptome that occur after epidermal injury. By comparing our array results with similar results on mammalian skin wounding we can see which evolutionarily conserved pathways are activated after epidermal wounding in very diverse animals. Our innovative serine protease-mediated wounding protocol allowed us to identify 8 additional genes that are activated in epidermal cells in the immediate vicinity of puncture wounds, and the functions of many of these genes suggest novel genetic pathways that may control epidermal wound repair. Additionally, our data augments the evidence that clean puncture wounding can mount a powerful innate immune transcriptional response, with different innate immune genes being activated in an interesting variety of ways. These include puncture-induced activation only in epidermal cells in the immediate vicinity of wounds, or in all epidermal cells, or specifically in the fat body, or in multiple tissues.