Plant Biotechnology Research Group, School of Biological Sciences and Biotechnology, Western Australian State Agricultural Biotechnology Centre, Murdoch University, Perth WA 6150, www.sabc.murdoch.edu.au Email: firstname.lastname@example.org
Plant parasitic nematodes are major pathogens of a wide range of crops. Giant cells induced by root-knot nematodes are highly specialised cells which function as transfer cells and provide nutrients to support the growth and reproduction of the nematode. Using a differential display approach, 81 differentially displayed bands were detected between the cytoplasm of giant cells induced in tomato roots by Meloidogyne javanica and control tissues. Of these, 73 were up-regulated and 8 were down-regulated. Sixteen were further analysed by real-time quantitative RT-PCR. The most highly up-regulated transcript increased 56 fold in giant cells, and the greatest down-regulation was 11 fold. A time course of expression of selected transcripts using RT-PCR from giant cell enriched tissue showed similar changes. Sequenced transcripts showed significant similarity to mitogen-activated protein kinase, S-adenosylmethionine decarboxylase, cysteine synthase, cytochrome c reductase subunit, and ribosomal proteins. The observed gene expression patterns reflect the high metabolic rate in mature giant cells rather than processes of giant cell induction. This work has been extended to analysis using Affymetrix GeneChip microarrays. A comparison of transcripts between giant cell enriched and control tissues revealed a total of 2,448 genes with more than 2-fold changes in expression (about 10% of the 24,000 genes on the chips). Of these genes, 744 were up-regulated in nematode feeding cells and 1,704 were down-regulated. These genes have been classified into functional groups, and the results show substantial changes in gene expression in giant cells that is consistent with the function of giant cells in supporting the development of the nematode parasites.
Root-knot nematodes are major crop pathogens which induce ‘giant’ feeding cells by re-programming gene expression in host cells: understanding these changes will lead to development of novel synthetic resistance to these pathogens.
Root-knot nematode; Meloidogyne; giant cells; gene expression; differential display; microarray
Root-knot nematodes (Meloidogyne spp.) infect more than 2,000 plant species and significantly reduce agricultural production (Sasser and Freckman, 1987). After hatching from eggs, second-stage juveniles invade roots of host plants, and migrate intercellularly to differentiating vascular regions. The nematodes then become sedentary and induce the formation of giant cells which act as the nutrient source for their development and reproduction (Jones, 1981; Williamson and Hussey, 1996). Between 2-12 giant cells are induced from pro-vascular cells in the differentiating vascular cylinder (Jones, 1981). They become multinucleate by repeated mitosis without cytokinesis and fill with metabolically active cytoplasm (Jones and Payne, 1978). There is an increase in the activities of many enzymes in giant cells, as well as in the rate of synthesis of DNA, ribosomal RNA and protein, compared to that in the surrounding cells, which reflects the increased metabolic activity inside giant cells (Jones, 1981; Favery et al., 1998). Formation and maintenance of giant cells requires continuous stimulus from nematode esophageal gland secretions (Williamson and Hussey, 1996). The modification of normal root cells into feeding structures includes complex morphological and physiological changes, and a new pattern of gene expression. The application of two approaches to study changes in gene expression in giant cells is presented here: differential display (DD) with real-time quantitative RT-PCR, and microarray technology.
Tomato (Lycopersicon esculentum) cv. Grosse Lisse (Yates, Smithfield, Australia) was cultured in vitro as host for root-knot nematode infections as described by Hutangura et al. (1999). Plants were grown at a constant temperature of 25°C with a light regime of 16 h light and 8 h dark.
Extraction of giant cell cytoplasmic contents was carried out with a modified pressure probe system described by Wang et al. (2001). The extracted cytoplasmic sample was expelled into 1μl of mRNA isolation buffer (100mM Tris-HCl pH 8.0, 500mM LiCl, 10mM EDTA pH 8.0, 1% LiDS, 5mM dithiothreitol.
mRNA from in vitro cultured healthy tomato roots or giant cells cytoplasm was isolated with Oligo(dT)25 Dynabeads (Dynal, Carlton South, Australia) following the method of Wang et al. (2003). For reliable DD analysis of giant cell cytoplasm, semi-quantitative RT-PCR was carried out to normalise the template amount for DD (Wang et al 2003). Differences in template amounts were estimated and different dilutions of mRNA from healthy root tissue were made. Expression levels of the actin gene in the dilutions and the giant cell extracts were also compared using semi-quantitative RT-PCR. The concentrations of mRNAs from healthy root tissue and giant cell extracts were then normalised for DD.
Differential display analysis was carried out using modified degenerate two-base anchor primers (ET12VN, where E is an EcoRI site as 5’-CGGAATTCGG-3’) and 18-21 mer elongated arbitrary primers that have over 50% GC content and dG or dC at the 3’ base (Zhao et al., 1995) as described by Wang et al (2003).
For selected cDNA fragments identified by direct sequencing, forward and reverse primers were specific to the nucleic acid sequences of the fragments, and real-time quantitative PCR analysis of the expression of various genes was then carried out using SYBR Green with an ABI PRISM 7700 Sequence Detector (Applied BioSystems) (see Wang et al. 2003). Quantification of the transcript level of the 16 cDNA fragments was normalised to the expression of the actin gene in giant cell cytoplasm and healthy root extracts.
Giant cell enriched and equivalent control tissues were harvested from infected Arabidopsis thaliana roots. Total RNA was isolated from 2,000 dissected galls and 500 mg control tissue using Trizol: the quality and quantity of RNA was checked using an Agilent Bioanalyser. About 5 μg intact RNA from each sample was used to synthesise first-strand cDNA via reverse transcription with a T7-(T)24 primer. The cDNA was used to generate biotin-labelled cRNA by in vitro transcription using the Enzo BioArray RNA transcript labelling method. Yield and the quality of the labelled cRNA was checked by spectrometer and electrophoresis. The cRNA was fragmented to short sequences: 15 μg of cRNA was used for each sample to hybridise to the Arabidopsis ATH1 genome array chip. Detection of the labelled cRNA with streptavidin-phycoerythrin and was achieved by scanning the chip. Duplicated samples for both infected and control tissues were set up to increase the accuracy of the result. Data generated was analysed using Affy Microarray Suite 5.0 software.
DDRT-PCR was done using 44 different primer combinations, and 81 differentially displayed bands (180bp - 800bp) with intensity differences between healthy root tissue and giant cell cytoplasmic extracts were identified. These 81 DD bands were: 73 up-regulated and 8 down-regulated. The reproducibility of the DD analysis was confirmed by repeating experiments with a number of primers and different mRNA samples as templates. Direct sequencing was undertaken to screen differentially displayed bands and 27 unique sequences were obtained. Details of these bands are presented in Table 1. For these bands the deduced amino acid sequences were compared to known protein sequences in GenBank using Blastp to identify possible homologues: 9 fragments showed significant identity to known sequences. For the other cDNA fragments, comparison of nucleotide sequences with the EST database in GenBank was carried out. The results indicate that all the DD bands, except two had at least one significant match with known EST sequences (Table 1). The deduced amino acid sequences of the matched longer ESTs were then compared with known protein sequences in the database
Table 1. Characteristics of sequenced DD bands.
a: Numbers in brackets for each gene show the levels of expression in giant cells cf control; tissue. b: homology at amino acid level. * no significant identity to known protein: homology determined by the longer EST sequence.
Sixteen fragments were analysed by real-time quantitative RT-PCR to confirm their differential expression. The differences in expression for the 16 candidate fragments are summarised in Table 2.
Table 2. Differences in expression level of genes between giant cell cytoplasmic contents and healthy root tissue, determined by the quantification of template amounts by real-time quantitative RT-PCR.
CtH and CtG are the mean Ct value from triplicate sample from healthy root tissue and giant cell cytoplasmic contents.
This work revealed 744 genes up-regulated more than 2-fold in feeding sites, and 1,704 down-regulated genes. The 744 up-regulated genes have been classified into different functional groups using Affymetrix annotation and other literature resources (Table 3). Further data analysis is being carried out to search for common motifs in promoter regions of the up-regulated genes to identify nematode responsive elements.
Table 3 Summary of functional classes of up-regulated genes and 3 representative genes from each class
a The number in parentheses following each gene represent degree of up-regulation compared to control root tissue.
Of the 744 up-regulated genes, 41% encode hypothetical, putative, or unknown proteins. The functional classification is: general cellular metabolism (17.7%), transcription regulation (9.8%), plant defence (7.4%), signal transduction (4.8%), cell structure and maintenance (4.4%), transport activity (4.2%). However, since these data were generated in one experiment, expression of selected transcripts is being checked by quantitative RT-PCR and in situ RT-PCR.
Figure 1. Functional categories of up-regulated genes identified by microarrays.
These results reflect the fact that giant cells have very active cytoplasm related to their function. Many genes identified in previous work are found in this data. In addition, several hundred genes have been identified here that have not been identified before in plant-nematode interactions.
Favery B, Lecomte, P, Gil N, Bechtold N, Bouchez, D, Dalmasso A. and Abad P. (1998) RPE, a plant gene involved in early developmental steps of nematode feeding cells. EMBO J. 17, 6799-6811.
Hutangura P, Mathesius U, Jones MGK and Rolfe BG (1999). Auxin induction is a trigger for root gall formation caused by root-knot nematodes in white clover and is associated with the activation of the flavonoid pathway. Aust. Plant Pathol. 26, 221-231.
Jones MGK (1981). Host cell responses to endoparasitic nematode attack: structure and function of giant cells and syncytia. Annal. Appl. Biol. 97, 353-372.
Jones MGK and Payne HL (1978). Early stage of nematode-induced giant-cell formation in roots of Impatiens balsamina. J. Nematol. 10, 70-84.
Sasser JN and Freckman DW (1987). A world perspective on nematology: The role of society. In Vistas in Nematology (Veech, J.A. and Dickerson, D.W., eds) Hyattsville: Society of Nematologists, pp. 7-14.
Wang Z, Potter RH and Jones MGK (2001). A novel approach to extract and analyse cytoplasmic contents from individual giant cells in tomato roots induced by Meloidogyne javanica. Inter. J. Nematol. 11, 219-225.
Wang Z, Potter RH and Jones MGK (2003). Differential display analysis of gene expression in the cytoplasm of giant cells induced in tomato roots by Meloidogyne javanica. Mol Plant Pathol 4, 361-371.
Zhao S, Ooi SL and Pardee AB (1995) New primer strategy improves precision of differential display. BioTechniques 18:842-850.