Journal of Experimental Botany Advance Access published March 1, 2015 Journal of Experimental Botany doi:10.1093/jxb/erv060

Research Paper

Arabidopsis C3H14 and C3H15 have overlapping roles in the regulation of secondary wall thickening and anther development Guohua Chai1, Yingzhen Kong2, Ming Zhu1, Li Yu1, Guang Qi1, Xianfeng Tang1, Zengguang Wang1, Yingping Cao1, Changjiang Yu1 and Gongke Zhou1,* Key Laboratory of Biofuels, Chinese Academy of Sciences, Shandong Provincial Key Laboratory of Energy Genetics, Qingdao Institute of BioEnergy and Bioprocess Technology, Chinese Academy of Sciences, Qingdao 266101, China 2  Key Llaboratory of Tobacco Gene Resource, Tobacco Research Institute, Chinese Academy of Agricultural Sciences, Qingdao 266101, China *  To whom correspondence should be addressed. E-mail: [email protected] Received 15 August 2014; Revised 4 January 2015; Accepted 21 January 2015

Abstract Plant tandem CCCH zinc finger (TZF) proteins play diverse roles in developmental and adaptive processes. Arabidopsis C3H14 has been shown to act as a potential regulator of secondary wall biosynthesis. However, there is lack of direct evidence to support its functions in Arabidopsis. It is demonstrated here that C3H14 and its homologue C3H15 redundantly regulate secondary wall formation and that they additionally function in anther development. Plants with double, but not single, T-DNA mutants for C3H14 or C3H15 have few pollen grains and thinner stem secondary walls than the wild type. Plants homozygous for c3h14 and heterozygous for c3h15 [c3h14 c3h15(±)] have slightly thinner secondary walls than plants heterozygous for c3h14 and homozygous for c3h15 [c3h14(±) c3h15], and c3h14(±) c3h15 have lower fertility. Overexpression of C3H14 or C3H15 led to increased secondary wall thickness in stems and the ectopic deposition of secondary walls in various tissues, but did not affect anther morphology. Transcript profiles from the C3H14/15 overexpression and c3h14 c3h15 plants revealed marked changes in the expression of many genes associated with cell wall metabolism and pollen formation. Subcellular localization and biochemical analyses suggest that C3H14/15 might function at both the transcriptional and post-transcriptional levels. Key words:  Anther development, Arabidopsis, C3H14, C3H15, gene regulation, secondary wall thickening.

Introduction Zinc finger proteins are a large family of proteins involved in numerous cellular functions including transcriptional regulation, RNA binding, regulation of apoptosis, and protein– protein interactions (Laity et  al., 2001). These proteins have been classified into nine subfamilies (C4, C6, C8, C2H2, C2HC, C2HC5, CCCH, C3HC4, and C4HC3; in this schema ‘C’ represents cysteine and ‘H’ represents histidine) based on their structure and function (Takatsuji, 1999). The CCCH proteins contain a tandem zinc finger domain characterized by three cysteines followed by one histidine, and have been found widely in eukaryotes, from yeast to mammals (D. Wang et al., 2008).

In animals, tandem CCCH zinc finger (TZF) proteins control a variety of cellular processes via regulation of gene expression at both the transcriptional and post-transcriptional level (Brown, 2005). The best known TZF protein is human tristetraprolin (hTTP/ZFP36), which often binds to class II AU-rich elements (AREs; AUUUA) in the 3′-untranslated region (UTR) of many target genes and recruits ARE mRNAs to cytoplasmic foci for silencing via RNA decay and translational repression (Lai et al., 1999). hTTP is known to shuttle between the cytoplasm and the nucleus via interaction with the nuclear exit signals (NESs), suggesting its potential

© The Author 2015. Published by Oxford University Press on behalf of the Society for Experimental Biology. All rights reserved. For permissions, please email: [email protected]

Downloaded from http://jxb.oxfordjournals.org/ at KTH Royal Institute of Technology on March 10, 2015

1 

Page 2 of 15 | Chai et al.

Materials and methods Gene expression analyses Total RNA isolation and first-strand cDNA synthesis were performed as described previously (Chai et  al., 2014). Quantitative

real-time PCR (qRT-PCR) assays were conducted on a LightCycler® 480 Detection System (Roche) with the Power 2×SYBR Real-time PCR pre-mixture (Bioteke). Relative gene expression with respect to ACTIN2 (At3g18780) was determined as described previously (Livak and Schmittgen, 2001). Semi-quantitative RT-PCR was conducted with gene-specific primers (Supplementary Table S5 available at JXB online) for 25–30 cycles. ACTIN2 was used as an internal control. Data presented represent the average of three biological replicates. Promoter fragments 1.4 kb and 0.88 kb upstream of the ATG codons of C3H14 and C3H15, respectively, were cloned into the pKGWFS7 vector (Invitrogen) to create the β-glucuronidase (GUS) reporter constructs. The resulting constructs were transformed into wild-type (Col-0) Arabidopsis using the floral-dip procedure, and transgenic plants were selected on plates containing 50 mg l–1 kanamycin. T2 transgenic plants were visualized by staining for GUS activity as described previously (Zhong et al., 2008). Generation of the C3H14/15 overexpression and c3h14 c3h15 plants The coding region of C3H14 or C3H15 was ligated downstream of the 35S promoter in the pH2GW7 vector (Invitrogen) to generate the overexpression constructs. The resulting constructs were introduced into wild-type Arabidopsis, and T0 transgenic plants were selected on hygromycin (25 mg l–1). T2 transgenic plants were used for examination of phenotypes. A T-DNA insertion line (c3h14, GABI_239E11) for C3H14 and two T-DNA insertion lines (c3h15-1, SALK_045897 and c3h15-2, SALK_065040) for C3H15 were ordered in the Salk institute T-DNA insertion library database. For each mutant, two pairs of primers (Supplementary Table S5 at JXB online) were used in PCR amplification to confirm the presence of T-DNA and to identify the homozygous mutant lines. The c3h14 c3h15 double mutants were generated by separately crossing c3h14 and c3h15-1 or c3h15-2 homozygotes. Microscopy and histochemistry Basal stems (5–6 mm) from 6- or 8-week-old Arabidopsis were fixed in 2.5% glutaraldehyde in phosphate buffer (pH 7.2), and embedded in resin (SPI-Chem.) according to a previous method (Chai et al., 2014). Stem sections (1 μm thick) were cut with a Leica EM UC6 microtome (Leica) and stained with 0.05% Toluidine blue O (TBO; Sigma) for light microscopy (Olympus DX51). For transmission electron microscopy, 70 nm thick sections were cut with a microtome, stained with TBO, and observed using a Hitachi H-7650 electron microscope (Hitachi). Wall thickness was measured in metaxylem vessels and in interfascicular fibres next to the endodermis. For each construct, at least six transgenic plants (those with the most severe phenotypes were used) were examined. For scanning electron microscopy analysis, Arabidopsis flower bud clusters were fixed in formalin, acetic acid, and ethanol solution for 24 h, dehydrated with a graded ethanol series, and dried at critical point in liquid CO2. Samples were coated with gold and then examined in an S-4160 field emission scanning electron microscope (Hitachi) at an accelerating voltage of 10 kV. For examination of cellulose in secondary walls, 1 μm thick sections were stained with 0.01% Calcofluor White and observed with a UV fluorescence microscope. For examination of xylan in secondary walls, 1 μm thick sections were probed with LM10 monoclonal antibody, which binds to 4-O-methylglucuronoxylan (McCartney et al., 2005), and detected with fluorescein isothiocyanate-conjugated secondary antibodies. The fluorescence-labelled xylan signals were visualized with a light microscope (Olympus DX51). Analysis of cell wall composition Inflorescence stems from 6- or 8-week-old Arabiodpsis were collected and were ground into a fine powder in liquid nitrogen.

Downloaded from http://jxb.oxfordjournals.org/ at KTH Royal Institute of Technology on March 10, 2015

secondary role in transcriptional regulation (Phillips et  al., 2002). TIS11d, another member of the TTP family, not only regulates target mRNA degradation in RNA processing by a similar mechanism (Ramos et al., 2004), but also activates transcription of a luciferase reporter gene in cell culture in transactivation assays (Murata et al., 2002). Plant TZF proteins participate in multiple developmental and adaptive processes including seed germination (Kim et al., 2008), embryo formation (Li and Thomas, 1998), floral reproductive organ identity (Cheng et al., 2003), plant architecture determination (L. Wang et al., 2008), leaf senescence (Kong et  al., 2006), and abiotic stress responses (Sun et  al., 2007; Lin et al., 2011; Deng et al., 2012). Compared with animals, relatively little is known about the regulatory mechanism of plant TZF proteins. Recently, two studies have suggested that Arabidopsis TZF1 may function in RNA regulation in the cytoplasm and in transcriptional regulation in the nucleus (Pomeranz et  al., 2010; Qu et  al., 2014). AtTZF1 has been shown to shuttle between cytoplasmic foci and the nucleus and to have both RNA- and DNA-binding capacities (Pomeranz et al., 2010). Moreover, AtTZF1 has been proven to trigger the decay of ARE-containing mRNAs in vivo (Qu et al., 2014). The Arabidopsis TZF protein C3H14 has gained attention for its potential role in secondary wall formation. For example, C3H14 is one of the direct targets of MYB46, which is itself a key transcriptional activator that regulates secondary wall formation (Kim et al., 2012). The expression of C3H14 is negatively regulated by WRKY12, which is the master switch that controls secondary wall thickening of pith cells (Wang et  al., 2010). Moreover, C3H14 has been shown to activate the expression of several secondary wall biosynthetic genes in Arabidopsis leaf protoplast (Ko et  al., 2009). As secondary wall thickening has important roles in various biological processes including the formation of tracheary elements (TEs) and fibres (Zhong et al., 2006, 2007; Ohashi-Ito et al., 2010), the dehiscence of anthers (Mitsuda et al., 2005), and the shattering of siliques pods (Mitsuda and Ohme-Takagi, 2008), it is important and necessary to elucidate the function of C3H14 in planta and to characterize the molecular mechanism underlying its involvement in secondary wall formation. In this study, direct evidence is provided that C3H14 and its homologue C3H15 have overlapping roles in the regulation of secondary wall formation and anther development. It appears that C3H14 contributes more to secondary wall thickening while C3H15 is more important in anther development. Microarray analyses of C3H14/15 overexpression and c3h14 c3h15 plants revealed that both genes regulate the expression of many genes involved in various biological processes, particularly those associated with cell wall metabolism and pollen development. The results also suggest that C3H14 and C3H15 might regulate secondary wall formation and anther development at both the transcriptional and post-transcriptional levels.

CCCHs regulate secondary wall and anther formation  |  Page 3 of 15

Microarray analysis Wild-type, 35S:C3H14, 35S:C3H15, and c3h14ch15 plants were grown side by side, and 4 cm basal inflorescence stems of the 13–17 cm height plants were sampled following a previously described method (Mitsuda et al., 2007). RNA isolation, probe labelling, chip hybridization, probe array scanning, and data pre-processing normalization were performed by Affymetrix custom service (CapitalBio, Beijing, China). The probes were labelled with an Affymetrix 3′ IVT Express kit. The GeneChip used contained 22 810 probe sets covering most identified cDNA and open reading frames of Arabidopsis. The Significance Analysis of Microarrays (SAM) software package (Tusher et  al., 2001) was used to evaluate three independent biological replicates for each sample. When all replicates were clustered together, further analysis was performed based on mean values. A  2-fold change in the gene expression levels between one sample versus another with a q-value ≤0.05 was set as the threshold for defining ‘altered’ gene expression. The microarray data set is available in the Gene Expression Omnibus (accession no. GSE53580). Protein targeting and transcriptional activation analysis The coding regions of C3H14 and C3H15 were separately fused inframe with a green fluorescent protein (GFP) in pH7WGF2 to generate the N-terminal fusion constructs. The AtTZF1 coding region was ligated into the modified pMDC163 vector, in which GUS is replaced by mCherry, to generate a control construct. 35S:GFP– C3H14 or 35S:GFP–C3H15 and mCherry-tagged AtTZF1 were co-infiltrated into tobacco leaves following a previously described method (Chai et al., 2014). GFP and mCherry signals were observed using an Olympus FluoView FV1000 confocal microscope. For transcriptional activation analysis, the full-length or partial coding sequences of C3H14 and C3H15 were separately fused inframe with the GAL4 DNA-binding domain in the pGBKT7 vector. The recombinant vectors and the pGBKT7 empty vector were transformed into yeast strain AH109, and the transformed strains were cultured on minimal medium without His (SD/-His) and SD/-Trp plates. The transactivation activity of each protein was evaluated according to their growth status and the activity of α-galactosidase. In vitro nucleic acid-binding assay The C3H14, C3H15, or SND1 (At1g32770, a positive control) coding region was separately cloned into the pGEX-4T-1 vector that

harbours a GST gene at the 5′ end of the multiple cloning sites (GE Healthcare). The recombinant proteins were purified from Escherichia coli with ProteinIso™ GST Resin according to the manufacturer’s protocol (TransGen Biotech). In vitro nucleic acid binding assays were performed as described previously (Deng et al., 2012). Briefly, 0.5 μg of purified protein was incubated with 20 μl of poly(rU), poly(rC) attached to agarose beads, and double-stranded and single-stranded calf thymus DNA attached to cellulose beads [note: poly(rA) and poly (rG) are out of production, Sigma] in RHPA binding buffer (10 mM TRIS, pH 7.4, 2.5 mM MgCl2, 0.5% Triton X-100, and 1 mg ml–1 heparin) with 0.1 M NaCl. Reactions were incubated at 4 °C for 10 min and then washed five times. Beads were then boiled in SDS loading buffer for 5 min and samples were separated on 12.5% SDS–polyacrylamide gels prior to western blot analysis.

Results Gene characteristics and expression patterns of C3H14 and C3H15 Arabidopsis C3H14 and C3H15 originate from a wholegenome duplication event, share 45% sequence identity over their full lengths, and 84% identity over their CCCH domains, and are therefore paralogues. Both proteins are typical tandem CCCH zinc finger proteins as they have two tandem CX8CX5CX3H motifs separated by 18 amino acids and conserved lead-in sequences (MM/TKTEL or RYKTEV) at their N-termini (Supplementary Fig. S1 at JXB online). In 8-week-old plants, both C3H14 and C3H15 were expressed predominantly in inflorescence stems, flowers, and siliques (Fig.  1A). They were highly expressed in the basal portion of stems, where cells are undergoing secondary wall thickening. C3H14 was expressed at higher levels than C3H15 during the early stages of stem development, and its expression declined to lower levels during later stages (Fig.  1A, B). C3H15 was expressed at a higher level than C3H14 in flowers and buds (Fig. 1A). These results suggest that the two genes may have overlapping roles in stem and flower development. Transgenic Arabidopsis plants expressing proC3H14:GUS or proC3H15:GUS were generated to investigate their developmental expression patterns. GUS staining was detected in the vascular cylinder in the mature region of roots (Fig. 1C, I) and leaves (Fig. 1E, K) and at the apex of lateral roots (Fig. 1D, J). Interestingly, GUS staining was discontinuous in the vasculature. This phenotype is similar to the tracheary element-specific expression patterns of proLBD30:GUS and proVND6:GUS (Soyano et al., 2008; Ohashi-Ito et al., 2010). GUS staining was also observed in all floral development stages from young buds through to open flowers. C3H14 expression was seen at low levels in stigmas and in the vascular tissues of the sepals (Fig. 1F). After fertilization, strong GUS signals were visible in the anther dehiscence zone and in pollen within floral tissues (Fig.  1G, H). In contrast, GUS staining for C3H15 was restricted to anthers in buds (Fig.  1L) and to the anther dehiscence zone and pollen in open flowers (Fig. 1M, N).

Downloaded from http://jxb.oxfordjournals.org/ at KTH Royal Institute of Technology on March 10, 2015

Alcohol-insoluble residues (AIRs) were prepared by treating the powder sequentially with 80% ethanol, 100% ethanol, and acetone. The resulting AIRs were dried under vacuum at 60 °C overnight and used for analysis of monosaccharide composition following the procedure described by Selvendran et al. (1979). Briefly, cell walls were hydrolysed for 2 h at 120 °C with 2 M trifluroacetic acid (TFA). The TFA-released materials were derivatized with 1-phenyl-3-methyl5-pyrazolone (PMP) and analysed on a Thermo ODS-2 C18 column (4.6 × 250 mm) connected to a Waters high-performance liquid chromatography (HPLC) system with a 2489 UV visible detector at a 245 nm wavelength. Elution solution A  was 0.1 M phosphate buffered (pH 7.0) and elution solution B was acetonitrile. The PMP derivative (10 μl) was injected, and eluted at 1 ml min–1 with elution solution A:elution solution B=82:18. To determine the cellulose content (Updegraff, 1969), TFA-resistant materials were treated with Updegraff reagent (acetic acid/nitric acid/water, 8:1:2 v/v/v) at 100 °C for 30 min, and the resulting pellets were then completely hydrolysed with 67% H2SO4 (v/v). The released glucose was measured using a glucose assay kit (Cayman Chemical) with a dehydration factor of 0.9. To determine the lignin content (Fukushima and Hatfield, 2001), 3 mg of AIR samples was solubilized by acetyl bromide solution, and 2 M sodium hydroxide and 0.5 M hydroxylamine hydrochloride were then added to stop the reaction. Absorbance at 280 nm was measured using a UV-visible spectrophotometer (VARIAN Cary 50).

Page 4 of 15 | Chai et al.

GUS activity in stem sections taken from positions corresponding to non-elongating internodes (basal regions) and to rapidly elongating internodes (upper regions) indicated that C3H14 and C3H15 have similar expression patterns. Intensive GUS staining was observed in interfascicular fibres and xylary fibres (Fig. 1P, R) that are undergoing secondary wall thickening (Ye, 2002). They were also expressed in developing vessels in the protoxylem, but not in the interfascicular fibres of elongating internodes (Fig. 1O, Q), where interfascicular fibre cells are undergoing rapid elongation but not secondary wall thickening (Ye, 2002). Together, the partially overlapping expression patterns of C3H14 and C3H15 observed in stems and anthers suggest that these two genes may be involved in secondary wall formation and flower development.

The c3h14 c3h15 double mutants have reductions in stem secondary wall thickening and defects in anther development To investigate the roles of C3H14 and C3H15 in secondary wall formation and flower development, stable T-DNA lines for C3H14 (c3h14, GABI_239E11) and C3H15 (c3h151, SALK_045897; c3h15-2, SALK_065040) (Fig.  2A) were obtained, and two double mutants were generated by separately crossing c3h14 and c3h15-1 or c3h15-2 homozygotes (Supplementary Fig. S2A, B at JXB online). The absence of expression of C3H14 and C3H15 transcripts was confirmed (Supplementary Fig. S2C). The c3h15-1 and c3h15-2 mutants had slightly longer inflorescence stems and roots than did the wild type, whereas the

Downloaded from http://jxb.oxfordjournals.org/ at KTH Royal Institute of Technology on March 10, 2015

Fig. 1.  Expression patterns of C3H14 and C3H15. (A and B) qRT-PCR showing the expression levels of C3H14 and C3H15 in various tissues of 8-weekold plants (A) and in different developmental stages in basal stems (B). ACTIN2 was used as an internal control. The expression levels of C3H14 and C3H15 in siliques (A) and in 5-week-old, basal stems (B) were set to 1. Error bars represent the SD of three biological replicates. Sl, siliques; Rt, root; Rl, rosette leaf; Yl, young leaf; Bs, basal stem; Us, upper stem; Fl, flower; Bd, bud. (C–R) Promoter assay showing expression patterns of C3H14 and C3H15. proC3H14:GUS expression in root (C and D), young leaf (E), and floral tissues (F–H), and proC3H15:GUS expression in root (I and J), young leaf (K), and floral tissues (L–N). The black box shows the proC3H15:GUS bud (L) at a similar developmental stage to the proC3H14:GUS bud (F). Stem tissues from proC3H14:GUS (O and P) and proC3H15:GUS (Q and R) lines were stained for GUS activity, and sections were cut from both upper (O and Q) and basal regions (P and R) of the stem. co, cortex; if, interfascicular fibre; xf, xylary fibre; ph, phloem; px, protoxylem. Scale bars=200 μm (C–G, I–M) or 100 μm (H, N–R). (This figure is available in colour at JXB online.)

CCCHs regulate secondary wall and anther formation  |  Page 5 of 15

c3h14 mutant was indistinguishable from wild-type plants (Table  1; Fig.  2B). The c3h14 c3h15-1 and c3h14 c3h15-2 plants had similar phenotypes. It appeared that the c3h14 c3h15-2 plants had a slightly more severe phenotype than the c3h14 c3h15-1 plants, and were therefore used in subsequent investigations. Compared with the wild type, three c3h14 c3h15 mutant combinations [c3h14 c3h15(±), c3h14(±) c3h15, and c3h14 c3h15] exhibited visible differences in stem length, rosette leaf size, root length, and fertility. The most striking alteration was observed in the c3h14 c3h15 plants (Table  1; Fig. 2B–N). It is noteworthy that very few pollen grains and slightly collapsed anther walls were observed in the c3h14(±) c3h15 and c3h14 c3h15 plants, but not in the c3h14 c3h15(±) plants. These results, combined with their different rates of sterility (Table 1), indicate that C3H15 contributes to fertility more than C3H14.

To investigate the roles of C3H14 and C3H15 during anther development, the anthers of open flowers were observed under UV illumination. A  net-like structure of lignified materials was found in the anther dehiscence zones of the c3h14(±) c3h15 and c3h14 c3h15 plants, similar to the wild type (Supplementary Fig. S3 at JXB online). This observation indicated that mutation in both C3H14 and C3H15 did not affect secondary wall thickening in anther dehiscence zones. Semi-thin microscopy analysis was performed on stage 7–14 anthers to examine further the morphological defects in the c3h14 c3h15 double mutant (Fig. 2O–X). In stage 7 c3h14 c3h15 anthers, microspore mother cells were observed to undergo meiotic cytokinesis, but the tetrads were found to be partially dissolved in the locules. It appeared that the tapetal layer of the double mutant was thinner than that of the wild type (Fig. 2O, T). At stage 9, the wild-type microspores

Downloaded from http://jxb.oxfordjournals.org/ at KTH Royal Institute of Technology on March 10, 2015

Fig. 2.  Phenotypes of C3H14 and C3H15 T-DNA knockout mutants. (A) Diagram of C3H14 and C3H15 showing the positions of the exon (solid box), intron (line), 5′-untranslated regions (line before the first box), and the T-DNA insertion sites. (B) Six-week-old wild-type, c3h14 (GABI_239E11), c3h15-1 (SALK_045897), c3h15-2 (SALK_065040), c3h14 c3h15(±), c3h14(±) c3h15, and c3h14 c3h15 plants. The c3h14 c3h15 plants are represented by the c3h14 c3h15-2 plants due to the similar phenotypes between c3h14 c3h15-1 and c3h14 c3h15-2. Arrows show the sterile siliques. (C–F) Top halves of the wild-type (C), c3h14 c3h15(±) (D), c3h14(±) c3h15 (E), and c3h14 c3h15 (F) flowers. Some of the petals and sepals have been removed. The arrows show the abnormal anthers in the c3h14(±) c3h15 (E) and c3h14 c3h15 (F) plants. Brown anthers were found in the c3h14 c3h15 plants (F). (G–N) Scanning electron microscopy images of the anthers of the wild type (G and K), c3h14 c3h15(±) (H and L), c3h14(±) c3h15 (I and M), and c3h14 c3h15 (J and N). No pollen grains were observed in c3h14 c3h15 plants (J and N), and only a few pollen grains were observed in c3h14(±) c3h15 plants (I and M). Arrows show the anthers with collapsed cell walls in c3h14(±) c3h15 (M) and c3h14 c3h15 (N) plants. (O–X) Transverse section of the anthers of the wild type (O–S) and c3h14 c3h15 double mutant (T–X) at stage 7 (O and T), 9 (P and U), 11 (Q and V), 13 (R and W), and 14 (S and X). In the c3h14 c3h15 anther, the tapetal cells and middle layer cells have delayed degeneration compared with the wild type, and few pollen grains are formed. E, epidermis; En, endothecium; T, tapetal layer; V, vascular region; Tds, tetrads; Msp, microspores; PG, pollen grains. Scale bars=1 mm (C–F), 100 μm (G–J), 10 μm (K–N), or 50 μm (O–X). (This figure is available in colour at JXB online.)

Page 6 of 15 | Chai et al. Table 1.  Measurements of growth parameters for the wild type, the C3H14 and C3H15 transgenic lines, and mutant plants Sample

Height (cm)

Stem diametera (mm)

Leaf blade lengthb (cm)

Leaf blade widthb (cm)

Root lengthc (cm)

Rate of sterility (%)

Wild type 35S:C3H14 35S:C3H15 c3h14 c3h15-1 c3h15-2 c3h14 c3h15(±)d c3h14(±) c3h15d c3h14 c3h15d

31.55 ± 3.16 25.73 ± 2.19* 23.20 ± 1.88* 30.98 ± 2.30 35.61 ± 3.62* 36.17 ± 3.64* 39.69 ± 2.63* 42.77 ± 3.34* 45.72 ± 2.03**

1.22 ± 0.16 1.28 ± 0.14 1.27 ± 0.18 1.25 ± 0.12 1.24 ± 0.15 1.25 ± 0.11 1.25 ± 0.20 1.26 ± 0.12 1.26 ± 0.17

3.73 ± 0.19 3.62 ± 0.29 3.67 ± 0. 23 3.75 ± 0.37 3.87 ± 0.25 3.86 ± 0.26 3.91 ± 0.14* 3.94 ± 0.18* 4.20 ± 0.18*

1.43 ± 0.21 1.21 ± 0.25 1.26 ± 0.11 1.39 ± 0.20 1.40 ± 0.24 1.40 ± 0.18 1.59 ± 0.09* 1.60 ± 0.14* 1.61 ± 0.10*

2.03 ± 0.15 1.59 ± 0.21* 1.63 ± 0.17* 1.98 ± 0.13 2.21 ± 0.16* 2.21 ± 0.17* 2.39 ± 0.15* 2.46 ± 0.24* 2.57 ± 0.21**

0.1 0.3 0.2 0.2 0.9 4.6 15.7 95.6 99.9

became vacuolated and the middle layer disappeared. In the double mutant, however, few microspores were observed and the middle layer was still present (Fig.  2P, U). At stage 11, the wild-type tapetum partially degenerated and round pollen grains were found in the locules, whereas the c3h14 c3h15 tapetal cells became vacuolated and greatly expanded and occupied the majority of the anther locules. The remaining pollen grains of the double mutant disintegrated into cell fragments and the middle layer was still clearly visible (Fig. 2Q, V). From stage 13 to 14, dehiscence occurred and pollen grains were released in the wild-type anthers. In contrast, in the c3h14 c3h15 anthers, the degeneration of tapetal and middle layer cells was delayed and the locules contained no pollen grains (Fig. 2R, S, W, X). These results indicate that C3H14 and C3H15 are required for pollen development. To determine whether C3H14 and C3H15 function redundantly in regulating secondary wall thickening in stems, transverse sections were taken from the basal stems of the wild type and of all of the single and double c3h14 c3h15 mutant plants (Fig. 3; Table 2). In c3h14 and c3h15 stems, no clear visible difference from the wild type was observed in the secondary wall thickness of fibres or vessels. (Fig. 3A1–D4; Table 2). In contrast, the c3h14 c3h15 double mutant stems exhibited a significant reduction in secondary wall thickness of fibres or vessels (Fig. 3G1–G4; Table 2). The observations that c3h14 c3h15(±) and c3h14 c3h15 plants had some deformed xylem vessels (Fig. 3E1, G1), and that the c3h14 c3h15(±) plants had slightly thinner (15%, P

Arabidopsis C3H14 and C3H15 have overlapping roles in the regulation of secondary wall thickening and anther development.

Plant tandem CCCH zinc finger (TZF) proteins play diverse roles in developmental and adaptive processes. Arabidopsis C3H14 has been shown to act as a ...
3MB Sizes 0 Downloads 7 Views