«by CHIEH-TING WANG (Under the Direction of Jeffrey F. D. Dean) ABSTRACT Laccase and related laccase-like multicopper oxidases (LMCOs) have been ...»
Salk collection (http://signal.salk.edu/cgi-bin/tdnaexpress) (Alonso et al. 2003), and the chromosomal flanking sequence showed that the T-DNA was inserted into the last intron of the At2tg30210 gene (Figure 3.1). To confirm the insertion, unique PCR products were generated using gene-specific primers from the 3’ and 5’ ends of the At2g30210 gene and the T-DNA left border primer (Figure 3.2A). The T-DNA right border primer together with At2g30210 primers failed to yield any product. This result suggested that the T-DNA was inserted as an inverted repeat within the At2g30210 gene. Similar insertions in which the T-DNA insert has two left borders, one at each end of the insertion site (LB-RB-RB-LB), have previously been reported in the Salk populations (Alonso et al. 2003). Three plants homozygous for the insertion were identified in a screen of 44 individuals using pairs of gene-specific primers that spanned the insertion region. RT-PCR analysis showed that two out of the three homozygous lines no longer produced properly edited transcript as indicated by loss of the ampliner band. The third line showed a weak, but detectable, ampliner band, suggesting that the T-DNA can be spliced out at low frequency (Figure 3.2B). Surprisingly, we were unable to identify kanamycin-resistant seedlings from any of the lines when we tested for co-segregation of the insertion with kanamycin resistance (data not shown). This
database indicating that some lines show silencing of the phosphotransferase (NPTII) gene after several generations of growth.
While work with the Salk T-DNA line was proceeding, two additional potential insertions (GT3416 and GT7855, Figure 3.1) were identified among the gene-trap lines generated using Ac/Ds transposons at Cold Spring Harbor Laboratory (http://genetrap.cshl.org/). In that system, the Ds gene-trap element carries as a reporter a β-glucuronidase (GUS) gene having a multiple splice acceptor site, as well as the NPTII gene, as a selectable marker. The gene-trap line, GT7855, has a Ds insertion in the second exon of the At2g30210 gene, 135 bp downstream of the translation start codon, as annotated by chromosomal flanking sequence generated using a TAIL (thermal asymmetric InterLaced) PCR procedure (Martienssen and Springer 1998). The GUS gene is inserted with transcription in the same direction as At2g30210 transcription. The insertion was confirmed from a unique PCR product that was generated using a gene-specific primer and a Ds transposon element primer. We used PCR to identify homozygous seedlings in which no ampliner could be detected using a pair of gene-specific primers, while an amplification product was detected from wild–type genomic DNA (Figure 3.2C). Unfortunately, we were unable to detect any GUS
the reporter runs in the same direction as the LMCO. However, the insertion line was still a strong knockout mutation as no properly edited At2g30210 transcripts were detected by RT-PCT using primers that spanned the insertion region (Figure 3.2D).
The second putative gene-trap insertion line, GT3416, with a reported Ds insertion in the second exon of the At2g30210 gene, 127 bp downstream of the translation start codon, was also annotated from chromosomal flanking sequence as having the GUS reporter in the same orientation as the At2g30210 gene.
Kanamycin-resistant seedlings from this line were transferred to soil and allowed to set flowers. Screening for homozygous individuals using PCR showed that all kanamycin-resistant plants were heterozygous (data not shown), and no significant phenotype was identified. We attempted this screening process three times, but the results were consistent. Whole mount GUS staining showed that the highest activity was observed in young roots, flowers and the abscission zone of siliques, with lesser activity in the pistil and immature seed (Figure 3.3). Very weak activities were also observed in anthers and nodes, but no activity was found in leaves or inflorescences.
Although the GT3416 line had a functional reporter gene with which to map gene expression in various tissues, further analyses failed to confirm that the insertion site was actually
to perform TAIL-PCR also yielded negative results (data not shown). Further analyses showed that the kanamycin resistance for all progeny from 4 lines failed to co-segregate with the insertion. These results indicated that the insertion may not be associated with the At2g0210 gene after all. Consequently, this line was dropped from further analyses.
Levels of At2g30210 mRNA transcript in homozygous Salk-031901 and GT7855 plants were compared with levels in wild-type plants using quantitative PCR (qPCR). These PCR analyses showed At2g30210 mRNA levels in homozygous GT7855 and Salk-031901 lines to be only somewhat reduced from wild-type levels, 71% and 56% of, respectively. Interestingly, whereas RT-PCR analyses of the Salk line clearly showed low levels of normal transcript being produced, PCR analysis of the 3’ end of the transcript suggested instability of the T-DNA modified transcripts (increased turnover), which was not seen in the Ds modified transcripts expressed by the GT7855 line (Figure 3.4).
T-DNA knockouts of At2g30210 show reduced growth and altered root morphology In the Salk-031901 insertion line, no significant phenotype was observed when seedlings were grown on soil or MS medium (data not shown). A weak phenotype in which the mutants did not develop as much as wild- type germinants was observed when
growth was severely repressed in both due to a lack of nutrients. To investigate whether stored nutrients deposited during seed development caused the phenotype, trace element (Fe, Cu, Mn, Zn, and Co) levels were analyzed in seeds from both mutant and wild-type plants. No significant differences were noted between wild-type and mutant seeds (data not shown).
For the GT7855 insertion line, the first phenotype observed was in root architecture. When plants were grown on MS medium supplemented with 1% sucrose, GT7855 produced slightly more lateral roots compared to wild-type plants (Figure 3.6A).
However, when grown in soil, the GT7855 line showed severe repression of early development, while the Salk mutant grew nearly as well as wild-type (Figure 3.6B). Since no phenotype was observed when Gt7855 seedlings were transferred to soil after initial growth on MS medium supplemented with sucrose, we manipulated the MS medium formulation to isolate the components affecting the mutant. It proved to be a lack of sucrose, among the major nutrient and minor metal elements, that caused the severe repression of early root development in GT7855 seedlings.
The growth of roots was severely repressed after germination on MS medium without sucrose (Figure 3.7). To test whether sucrose was capable of rescuing the mutant grown on soil, two-week-old mutants were watered with a 1 % sucrose solution, after which
showed that glucose was also capable of rescuing the mutant phenotype, but fructose was not as effective (Figure 3.8B). The results indicated that sucrose and/or glucose somehow compensate for loss of the At2g30210 gene product to bring about normal root growth during the early stages of Arabidopsis development.
GT7855 mutant is hypersensitive to salt stress Salt sensitivity is most evident at germination and during the early stages of Arabidopsis seedling development (Xiong and Zhu 2002). At 80 mM NaCl, both mutant and wild-type seedlings stopped growing on MS medium after initial radical emergence (data not shown). However, when the salt concentration was lowered to 40 mM NaCl, wild-type and mutant seedlings grew, albeit poorly and showing obvious symptoms of salt stress, but the GT7855 plants eventually died after 4 weeks, while wild-type and the Salk mutant continued to survive (Figure 3.9).
At2g30210 is expressed in the root endodermis Previous study showed that the At2g30210 gene is primarily expressed in root tissues (McCaig et al. 2005). Gene expression patterns at the cellular level can facilitate linking physiological function of a gene product to the differentiation state of certain cells or tissue types. Thus, previous studies suggested that LMCOs might be involved in lignification of differentiating xylem tissues in stems. To test whether
Arabidopsis roots, transgenic plants expressing the βglucuronidase (GUS) reporter gene under the control of the At2g30210 promoter were generated. Whole mount histochemical staining for GUS activity showed that promoter activity was detected in whole roots and hypocotyls during early developmental stages, but later only in the elongation region behind primary and lateral root tips (Figure 3.10). Higher magnification appeased to show that the At2g30210 gene was preferentially expressed in the developing endodermis of Arabidopsis roots (Figure 3.10E). This result is consistent with our previous RT-PCR data and matches well with measurements made using Affymetrix chips (Birnbaum et al. 2003), which showed that expression of this gene is strongly up-regulated in those root developmental zones where the endodermis is initially formed.
GUS activity was up-regulated during development of lateral roots similar to the manner seen in the primary root (Figure 3.11).
Some transgenic seedlings grown in liquid MS medium, but not on solid MS agar plates, showed expression of At2g3210 in the major veins of cotyledons, and very low levels of expression were observed in developing pollen sacs if stainign for GUS activitiy was extended to three days (Figure 3.10). However,
stem, or silique tissues (Figure 3.10 and data not shown).
At2g30210 protein is localized to the cell wall Ttwo different algorithms (Predotar V1.03 and TargetP V1.01) predicted the At2g30210 protein to possess a secretory signal peptide that targeted the gene product for secretion via the endomembrane system. Extracellular localization would be consistent with a putative function in cell wall formation as has been proposed for other plant LMCOs. However, other software algorithms predicted subcellular localization to mitochondria or the nucleus (http://www.bioinfo.tsinghua.edu.cn/SubLoc/). To determine the precise subcellular localization site for the At2g30210 gene product, we prepared a chimeric protein in which an enhanced yellow fluorescent protein (EYFP) was linked in-frame to the carboxyl-terminal end of the At2g30210 gene product. The resultant construct, driven with a 35S promoter, was transiently expressed in suspension-cultured tobacco cells (BY2), where the protein appeared to be localized to the cell periphery and nucleus (Figure 3.12A). In contrast, fluorescence corresponding to EYFP protein alone was localized in both cytoplasm and nucleus, consistent with the subcellular localization pattern of the control protein reported previously (David and PerrotRechenmann 2001). The same construct was subsequently used to stably transform Arabidopsis, and confocal laser microscopy of
fluorescence that appeared to support apoplastic localization of the chimeric protein. EYFP fluorescence was localized to the cell periphery, most likely extracellularly and possibly to the cell wall (Figure 3.12E). We were not able to detect deformation in the pattern when the tissues were subjected to plasmolytic conditions, suggesting that the protein is likely not localized to the cytoplasm or internal plasmalemma (data not shown). These results are consistent with previous studies in which LMCOs were localized to cell walls, where they may be involved in the lignin biosynthesis.
Discussion The results presented here suggest involvement of LMCOs in development of plant root endodermis. Strong phenoloxidase activity of the heterologously expressed enzyme (see Chapter 2) suggests involvement in the polymerization of phenolic materials. Promoter activity was strongest in the endodermis, a tissue whose specialized cell walls are characterized by development of Casparian bands that contain suberin and the phenolic polymer, lignin (Brundrett et al. 1988, Ma and Peterson 2003). Casparian bands are initially formed very close to the root tip, usually within 10 mm (Ma and Peterson 2003). GUS reporter activity driven by the At2g30210 promoter was initially
more developed endodermis cells, where suberin lamellae are formed. The At2g30210-YFP chimeric protein was targeted to the cell periphery, possibly the cell wall, where lignin and suberin are deposited. Finally, and most importantly, observations that the At2g30210 mutant phenotypes showed a sugar requirement during early development and an increased sensitivity to salt stress were consistent with altered function of endodermal cell walls. Endodermal walls are characterized by the Casparian strip, a specialized type of cell wall structure that functions to seal the perimeter of the endodermal walls and, thereby, create an apoplastic transport barrier to water, ions, and sugar between the root cortex and stele (Sattelmacher 2001). Changes in structure or composition of the Casparian strip are known to impact hydraulic conductivity and ion flow (Schreiber et al.
It is well known that growth of A. thaliana is strongly dependent on a supply of exogenous sugar (Gibson 2000). When exogenous sugar is limited, photosynthetic products generated above ground become the sole source of sugars, especially for early stage cotyledons. For mutants lacking the functional At2g30210 gene product, photosynthetic sugar may not be efficiently transported to sink tissues in the root tips because of diffusive leaks along the malformed stele. As a result, root
sufficient nutrients from the surrounding soil to continue growth. A similar phenotype was observed in an Arabidopsis mutant in which loss-of-function mutants of enzymes involved in sugar-loading into the phloem impaired transport and allocation of sugars to other parts of the plant (Deeken et al. 2002).