Pollen tubes are highly polarized tip-growing cells that depend on cytosolic pH gradients for signaling and growth. Autoinhibited plasma membrane proton (H + ) ATPases (AHAs) have been proposed to energize pollen tube growth and underlie cell polarity, however, mechanistic evidence for this is lacking. Here we report that the combined loss of AHA6, AHA8 , and AHA9 in Arabidopsis thaliana delays pollen germination and causes pollen tube growth defects, leading to drastically reduced fertility. Pollen tubes of aha mutants had reduced extracellular proton (H + ) and anion fluxes, reduced cytosolic pH, reduced tip-to-shank proton gradients, and defects in actin organization. Furthermore, mutant pollen tubes had less negative membrane potentials, substantiating a mechanistic role for AHAs in pollen tube growth through plasma membrane hyperpolarization. Our findings define AHAs as energy transducers that sustain the ionic circuit defining the spatial and temporal profiles of cytosolic pH, thereby controlling downstream pH-dependent mechanisms essential for pollen tube elongation, and thus plant fertility.

Plasma membrane H+-ATPase pumps build up the electrochemical H+ gradients that energize most other transport processes into and out of plant cells through channel proteins and secondary active carriers. In Arabidopsis thaliana, the AUTOINHIBITED PLASMA MEMBRANE H+-ATPases AHA1, AHA2 and AHA7 are predominant in root epidermal cells. In contrast to other H+-ATPases, we find that AHA7 is autoinhibited by a sequence present in the extracellular loop between transmembrane segments 7 and 8. Autoinhibition of pump activity was regulated by extracellular pH, suggesting negative feedback regulation of AHA7 during establishment of an H+ gradient. Due to genetic redundancy, it has proven difficult to test the role of AHA2 and AHA7, and mutant phenotypes have previously only been observed under nutrient stress conditions. Here, we investigated root and root hair growth under normal conditions in single and double mutants of AHA2 and AHA7. We find that AHA2 drives root cell expansion during growth but that, unexpectedly, restriction of root hair elongation is dependent on AHA2 and AHA7, with each having different roles in this process.

Flowering plants undergo a life cycle in which the diploid sporophyte constitutes the predominant generation, and the haploid male and female gametophytes are reduced to only a few but highly specialized cells (McCormick, 1993McCormick S. Male gametophyte development.Plant Cell. 1993; 5: 1265-1275Crossref PubMed Scopus (459) Google Scholar). To achieve expression restricted to certain cell types, genes are not only actively expressed in these cells, they might also be repressed elsewhere. Here, we report that deleting an 88-base-pair (bp) promoter sequence of the pollen-specific gene ACA7 redirects its expression to the sporophyte. Our study thus identified a bifunctional cis-regulatory module (CRM) that functions as a repressor in a plant sporophyte and as an activator in the male gametophyte. Early mutation studies of pollen-specific promoters did not identify DNA sequences mediating gene repression (Twell et al., 1991Twell D. Yamaguchi J. Wing R.A. Ushiba J. McCormick S. Promoter analysis of genes that are coordinately expressed during pollen development reveals pollen-specific enhancer sequences and shared regulatory elements.Genes Dev. 1991; 5: 496-507Crossref PubMed Scopus (250) Google Scholar, Eyal et al., 1995Eyal Y. Curie C. McCormick S. Pollen specificity elements reside in 30 bp of the proximal promoters of two pollen-expressed genes.Plant Cell. 1995; 7: 373-384Crossref PubMed Scopus (96) Google Scholar). Later, the LAT59 gene of tomato was found to possess a sequence in its 5′ untranslated region that acts as a strong inhibitor of gene expression, probably by forming a stem-loop structure that causes proximal-promoter pausing of RNA polymerase (Curie and McCormick, 1997Curie C. McCormick S. A strong inhibitor of gene expression in the 5' untranslated region of the pollen-specific LAT59 gene of tomato.Plant Cell. 1997; 9: 2025-2036PubMed Google Scholar). However, the repressive function was exhibited in the sporophyte as well as in pollen (Curie and McCormick, 1997Curie C. McCormick S. A strong inhibitor of gene expression in the 5' untranslated region of the pollen-specific LAT59 gene of tomato.Plant Cell. 1997; 9: 2025-2036PubMed Google Scholar). Today, the only experimental evidence for a DNA motif conferring strict gene repression in every cell except the sperm cell in pollen is the 10-bp region GGCTGAATTT in the promoter of LILY GENERATIVE CELL- SPECIFIC 1 (LGC1) in lily, which is recognized by the silencing transcription factor GERMLINE RESTRICTIVE SILENCING FACTOR (GRSF) in non-male germline cells (Haerizadeh et al., 2006Haerizadeh F. Singh M. Bhalla P. Transcriptional repression distinguishes somatic from germ cell lineages in a plant.Science. 2006; 313: 496-499Crossref PubMed Scopus (41) Google Scholar). When the GUS reporter gene was expressed under control of the CaMV 35S promoter fused to the repressive GRSF binding motif, gene expression was abolished in transgenic Arabidopsis thaliana plants, suggesting the presence of a pathway to repress gene expression in the sporophyte (Haerizadeh et al., 2006Haerizadeh F. Singh M. Bhalla P. Transcriptional repression distinguishes somatic from germ cell lineages in a plant.Science. 2006; 313: 496-499Crossref PubMed Scopus (41) Google Scholar). However, mutational studies of the pollen-specific DUO POLLEN1 (DUO1) gene, which has a conserved GRSF binding motif in its promoter, found that DUO1 expression is not regulated by this sequence; this implies that there must be other mechanisms for sporophytic repression of pollen-specific genes (Brownfield et al., 2009Brownfield L. Hafidh S. Borg M. Sidorova A. Mori T. Twell D. A plant germline-specific integrator of sperm specification and cell cycle progression.PLoS Genet. 2009; 5: e1000430Crossref PubMed Scopus (129) Google Scholar). Expression of ACA7, which encodes for a plasma membrane Ca2+ P-type ATPase in A. thaliana, is restricted to pollen and its protein product is essential for proper pollen development (Lucca and León, 2012Lucca N. León G. Arabidopsis ACA7, encoding a putative auto-regulated Ca2+-ATPase, is required for normal pollen development.Plant Cell Rep. 2012; 31: 651-659Crossref PubMed Scopus (35) Google Scholar). To identify cis-regulatory elements involved in regulating the pollen-specific expression of ACA7, we created six constructs with ACA7 promoter sequences of different lengths fused to the beta-glucuronidase (GUS) gene (Figure 1A , Supplemental Table 3 for primer sequences). These constructs were stably transformed into A. thaliana and GUS expression was analyzed by X-GlcA assay. The two constructs with the longest ACA7 promoters (1504 bp and 1361 bp) gave rise to GUS activity in pollen (Figure 1B), which is in agreement with previous studies (Lucca and León, 2012Lucca N. León G. Arabidopsis ACA7, encoding a putative auto-regulated Ca2+-ATPase, is required for normal pollen development.Plant Cell Rep. 2012; 31: 651-659Crossref PubMed Scopus (35) Google Scholar) and microarray data (Supplemental Figure 1). Additionally, GUS was expressed in the tips of leaves, possibly hydathodes, which was not observed for endogenous ACA7, as revealed by reverse transcription PCR (Figure 1B and 1C). The pollen-specific pattern of GUS expression, however, was only observed in about 40% of individually transformed plants; a rate similar to that of a gene expressed in microspores and pollen (Honys et al., 2006Honys D. Oh S.A. Reňák D. Donders M. Šolcová B. Johnson J.A. Boudová R. Twell D. Identification of microspore-active promoters that allow targeted manipulation of gene expression at early stages of microgametogenesis in Arabidopsis.BMC Plant Biol. 2006; 6: 31Crossref PubMed Scopus (43) Google Scholar). The remaining 60% showed expression in the vasculature of leaves and sepals, but not in pollen. With the third longest promoter truncation (1273 bp), surprisingly, 100% of individually transformed plants showed GUS activity in the vasculature of leaves and sepals, but not in pollen. This expression profile was also found in 100% of individually transformed plants with a much shorter promoter sequence of 429 bp. Truncating the promoter further (to 218 bp or 101 bp) resulted in a lack of GUS activity in all plant parts (Figure 1B). GUS expression was tested in T1 and T2 generations, and all results were identical across generations. The results from the GUS expression analysis hinted at a regulatory element for pollen-specific expression located in the ACA7 promoter between −1361 bp and −1273 bp from the start codon. The activity of the element may be influenced by genetic or epigenetic clues conferred by the transgene's insertion sites (Matzke and Matzke, 1998Matzke A.J. Matzke M.A. Position effects and epigenetic silencing of plant transgenes.Curr. Opin. Plant Biol. 1998; 1: 142-148Crossref PubMed Scopus (308) Google Scholar, Day et al., 2000Day C.D. Lee E. Kobayashi T. Holappa L.D. Albert H. Ow D.W. Transgene integration into the same chromosome location can produce alleles that express at a predictable level, or alleles that are differentially silenced.Genes Dev. 2000; 14: 2869-2880Crossref PubMed Scopus (197) Google Scholar), which would explain why, in independent transgenic lines including this element, stable expression was either in the gametophyte or in the sporophyte. In the absence of this element, expression was exclusively in the sporophyte depending on a promoter sequence further downstream (located between −429 bp and −218 bp from the start codon). Thus, the factor(s) that bind to the element required for pollen-specific expression probably establish repression by interacting with a sequence (or a factor binding to that sequence) several hundred base pairs apart. Taken together, our results indicate the presence of an upstream CRM with a dual function: activating gene expression in the male gametophyte and repressing it in the sporophyte. Having identified an 88-bp CRM in the ACA7 promoter, we searched the DNA sequence for the presence of known transcription factor binding sites. Using stringent search criteria, a DNA motif at position −1301 to −1291 was identified. The motif's sequence is GAATATTCCT and is recognized by KANADI4/ABERRANT TESTA SHAPE (KAN4, AT5G42630), which belongs to the GARP family of transcription factors. Another member of this family, KANADI1 (KAN1, AT5G16560), recognizes a part of the same motif (GAATAT) at position −1301 to −1295 (and its palindromic sequence ATATTC at position −1299 to −1293). Both KAN1 and KAN4 are transcriptional repressors and are involved in determining the polarity of leaves (Huang et al., 2014Huang T.B. Harrar Y. Lin C.F. Reinhart B. Newell N.R. Talavera-Rauh F. Hokin S.A. Barton M.K. Kerstetter R.A. Arabidopsis KANADI1 acts as a transcriptional repressor by interacting with a specific cis-element and regulates auxin biosynthesis, transport, and signaling in opposition to HD-ZIPIII factors.Plant Cell. 2014; 26: 246-262Crossref PubMed Scopus (88) Google Scholar). Noteworthy, the pollen-specific expression of the tobacco NTP303 gene is driven by a cis-regulatory element with the sequence AAATGA (Weterings et al., 1995Weterings K. Schrauwen J. Wullems G. Twell D. Functional dissection of the promoter of the pollen-specific gene NTP303 reveals a novel pollen-specific, and conserved cis-regulatory element.Plant J. 1995; 8: 55-63Crossref PubMed Scopus (63) Google Scholar), which is present in an extension of the motif we found (AAATGAATATTCCT). We tested if the identified DNA motif would be overrepresented in promoters of pollen-specific genes, compared with genes being highly expressed in the sporophyte, which we identified previously (Hoffmann and Palmgren, 2013Hoffmann R.D. Palmgren M.G. Epigenetic repression of male gametophyte-specific genes in the Arabidopsis sporophyte.Mol. Plant. 2013; 6: 1176-1186Abstract Full Text Full Text PDF PubMed Scopus (12) Google Scholar). Using two different in silico tools to identify overrepresented DNA motifs, we did not find any differences in the abundance of these motifs within the different gene sets (Figure 1C, Supplemental Tables 1 and 2 show additional promoter lengths). We also analyzed the minimal ACA7 promoter sequence necessary to drive GUS expression in the sporophyte (between 429 bp and 218 bp upstream of the start codon) for known transcription factor binding sites. No motif was found under the same stringent parameters as we used for analyzing the CRM (data not shown). To further examine the biological relevance of the potential KAN1/4 binding site, we analyzed DNase-I cleavage data of genomic DNA purified from leaves and flowers (Zhang et al., 2012Zhang W. Zhang T. Wu Y. Jiang J. Genome-wide identification of regulatory DNA elements and protein-binding footprints using signatures of open chromatin in Arabidopsis.Plant Cell. 2012; 24: 2719-2731Crossref PubMed Scopus (162) Google Scholar). The absence of DNase-I cleavage at a site indicates that proteins are binding to it, making the site inaccessible for DNase-I. We found that the DNA at the motif was inaccessible in leaves (Figure 1D). In flowers, however, DNase-I could cleave DNA at the motif, suggesting it was not (or less abundantly) bound by proteins. This finding indicates that the CRM is bound by proteins in leaves. In summary, we have shown that an 88-bp DNA sequence forms a CRM that represses gene expression in the sporophyte, while activating it in pollen. This regulatory function must be driven by interaction of the CRM with other elements, of which one or more are located closer to the transcription start site. Additionally, because the activation in pollen and repression in leaves was found only in about 40% of individually transformed lines, the CRM's functionality may depend on genetic or epigenetic clues present at the transgene's insertion sites. Our finding of a bifunctional CRM is important for many aspects of gene regulation and the transcriptional changes underlying gametophyte development. Further experiments might identify the proteins involved in this process of gene regulation. This work was supported by the Danish National Research Foundation through the PUMPkin Center of Excellence.

BACKGROUND:Whole-genome duplications in the ancestors of many diverse species provided the genetic material for evolutionary novelty. Several models explain the retention of paralogous genes. However, how these models are reflected in the evolution of coding and non-coding sequences of paralogous genes is unknown.RESULTS:Here, we analyzed the coding and non-coding sequences of paralogous genes in Arabidopsis thaliana and compared these sequences with those of orthologous genes in Arabidopsis lyrata. Paralogs with lower expression than their duplicate had more nonsynonymous substitutions, were more likely to fractionate, and exhibited less similar expression patterns with their orthologs in the other species. Also, lower-expressed genes had greater tissue specificity. Orthologous conserved non-coding sequences in the promoters, introns, and 3' untranslated regions were less abundant at lower-expressed genes compared to their higher-expressed paralogs. A gene ontology (GO) term enrichment analysis showed that paralogs with similar expression levels were enriched in GO terms related to ribosomes, whereas paralogs with different expression levels were enriched in terms associated with stress responses.CONCLUSIONS:Loss of conserved non-coding sequences in one gene of a paralogous gene pair correlates with reduced expression levels that are more tissue specific. Together with increased mutation rates in the coding sequences, this suggests that similar forces of purifying selection act on coding and non-coding sequences. We propose that coding and non-coding sequences evolve concurrently following gene duplication.

Insufficient intake of zinc and iron from a cereal-based diet is one of the causes of 'hidden hunger' (micronutrient deficiency), which affects some two billion people(1,2). Identifying a limiting factor in the molecular mechanism of zinc loading into seeds is an important step towards determining the genetic basis for variation of grain micronutrient content and developing breeding strategies to improve this trait(3). Nutrients are translocated to developing seeds at a rate that is regulated by transport processes in source leaves, in the phloem vascular pathway, and at seed sinks. Nutrients are released from a symplasmic maternal seed domain into the seed apoplasm surrounding the endosperm and embryo by poorly understood membrane transport processes(4-6). Plants are unique among eukaryotes in having specific P1B-ATPase pumps for the cellular export of zinc(7). In Arabidopsis, we show that two zinc transporting P1B-ATPases actively export zinc from the mother plant to the filial tissues. Mutant plants that lack both zinc pumps accumulate zinc in the seed coat and consequently have vastly reduced amounts of zinc inside the seed. Blockage of zinc transport was observed at both high and low external zinc supplies. The phenotype was determined by the mother plant and is thus due to a lack of zinc pump activity in the seed coat and not in the filial tissues. The finding that P1B-ATPases are one of the limiting factors controlling the amount of zinc inside a seed is an important step towards combating nutritional zinc deficiency worldwide.

Tissue formation, the identity of cells, and the functions they fulfill, are results of gene regulation. The male gametophyte of plants, pollen, is outstanding in this respect as several hundred genes expressed in pollen are not expressed in the sporophyte. How pollen-specific genes are down-regulated in the sporophyte has yet to be established. In this study, we have performed a bioinformatics analysis of publicly available genome-wide epigenetics data of several sporophytic tissues. By combining this analysis with DNase I footprinting data, we assessed means by which the repression of pollen-specific genes in the Arabidopsis sporophyte is conferred. Our findings show that, in seedlings, the majority of pollen-specific genes are associated with histone-3 marked by mono- or trimethylation of Lys-27 (H3K27me1/H3K27me3), both of which are repressive markers for gene expression in the sporophyte. Analysis of DNase footprint profiles of pollen-specific genes in the sporophyte displayed closed chromatin proximal to the start codon. We describe a model of two-staged gene regulation in which a lack of nucleosome-free regions in promoters and histone modifications in open reading frames repress pollen-specific genes in the sporophyte.