cDNA-AFLP-Based Numerical Comparison of Leaf and Root Organ cDNAs in Catharanthus roseus

Introduction

C atharanthus roseus is an alkaloid reservoir plant known to produce around 130 terpenoid indole alkaloids (TIAs; van der Heijden et al., 2004). Cellular differentiation plays an important role in defining the pattern of TIA biosynthesis in C. roseus (Shukla et al., 2010). Wide differences have been noted in the spectrum of the alkaloids when they are isolated from the plant's underground and aerial tissues (Shukla, 2005). The plant's roots accumulate ajmalicine and serpentine. These two are essential constituents of medicines used for controlling high blood pressure and other cardiovascular disorders. Moreover, C. roseus is principally known for two other alkaloids: vinblastine and vincristine. These two are leaf-specific bisindole alkaloids, which are economically vital as antineoplastic agents and are ingredients of most cancer chemotherapies (Neuss, 1980; Svoboda and Blake, 1975). Vinblastine and vincristine are produced in vivo by the condensation of vindoline and catharanthine. Vindoline, which is found only in the green parts of the plant and not in the roots or cell suspension cultures, is biosynthesized from tabersonine (which in turn is derived from the branch-point intermediate strictosidine through multiple steps) through the action of six enzymatic steps (De Luca et al., 1986). On the contrary, catharanthine (also derived from strictosidine) has been reported to be present in the root tissue as well as the aerial part of the plant (Deus-Neumann et al., 1987). Interestingly, it has also been shown in the related species C. pusillus that the leaves rather than the roots are the major accumulative tissues of vindoline and catharanthine (Za'rate et al., 2001).

The existence of different alkaloids and expressed genes within the aerial and underground tissues of C. roseus indicates the expression of distinct tissue-specific pathways. In turn, these pathways are responsible for the precise tissue-specific compartmentation of TIA biosynthesis (Laflamme et al., 2001). Additionally, there are several distinct functions associated with either of the tissue types, all of which result from the differential expression of structural and regulatory factors in these two tissues. Gene prospecting through comparative transcriptome analysis of the aerial and underground tissues may provide a better understanding of the tissue-specific differences. Earlier, an effort was made in this direction by our group using an expressed sequence tag (EST)-based approach to differentiate the root and leaf transcriptomes in C. roseus (Shukla et al., 2006). In the present study we describe a cDNA-AFLP approach to differentiate the root and leaf organ cDNAs in C. roseus only on a numerical basis.

Materials and Methods

Results and Discussion

In order to get a numerical estimate of the differences and similarities between the root and leaf tissue cDNAs of C. roseus, a cDNA-AFLP approach was used in the present study and the results obtained are described below.

TDF analysis of the selective amplification products (of the 64 multiplexed PCR reactions) revealed a comprehensive comparative picture of the root (R) and leaf (L) organ cDNAs (Table 1). Figure 1 (Venn diagram) illustrates the organ-specific distribution of the obtained TDFs. In this analysis, a total of 784 distinct TDFs could be detected in the root and leaf organ cDNAs. The analyzed leaf (L) subset yielded a larger number of TDFs than the root (R) subset (556 compared to 464), indicating a greater variety of expressing genes in the leaf (L). TDFs that were expressed either only in the root (228) or only in the leaf (320) may be considered organ-specific, and hence may be implicated in development and/or function of these distinct organs. Of the 236 TDFs that were detected in both of the organ types, 42 had nearly equal expression levels in both organs (L being equal to R with L/R or R/L expression ratio≤1.25). Common TDFs having higher expression levels in the leaf (L>R; 124) outnumbered those having higher expression levels in the root (L<R; 70). Size distribution of the detected TDFs was also determined (Table 2). Most of the detected TDFs were less than 150 bp in size (487/784; 62.12%). Approximately 25.76% (202/784) of the TDFs were in the range 150–300 bp, whereas only 12.12% (95/784) of the TDFs were more than 300 bp in size.

OPEN IN VIEWER

FIG. 1.

Comparative distribution of transcriptionally-defined fragments (TDFs) in the root and leaf transcriptomes of Catharanthus roseus.

Leaf-specific TDFs (320) outnumbered root-specific TDFs (228), indicating higher numbers of leaf-specific functions and the relative complexity of the leaf transcriptome vis-à-vis the root transcriptome. This is in accord with earlier reports in literature that have described the plant root as a simple organ that is comprised of a relatively few differentiated cell types (Schiefelbein and Benfey, 1991). The root apical meristem is easily accessible (unlike the shoot apical meristem), and lacks pigmentation and branching primordial. Root morphogenesis normally occurs in a reiterative and uniform fashion, without any major change in the organization of the root apex. Thus, all the stages of root development are apparent at all times, and there is nothing analogous to the vegetative-to-floral developmental transition that occurs in shoot meristems.

Even though the root possesses a smaller number of specifically-expressing transcripts, attempts have been made earlier in other plants to find genes that express specifically in the root (Choi et al., 1996; Conkling et al., 1990), and those that express preferentially in the root (Evans et al., 1988; Lin et al., 2005). In a previous microarray-based study of 1400 Arabidopsis cDNAs to compare expression patterns between roots and leaves, it was found that 30% of the genes had higher expression levels in roots compared to leaves, but few differences were detected between open flowers and flower buds or leaves (Ruan et al., 1998). It has also been demonstrated that a considerable number of the leaf mRNAs in Arabidopsis are differentially expressed in a day- or night-dependent manner (Schaffer et al., 2000). These observations indicate that the root and the leaf organs are transcriptomically different, and are consequently functionally distinct, with higher numbers of distinct functions associated with the leaf, possibly due to its exposure to aerial stress factors. The leaf has to provide the first line of defense against herbivory and pests, which makes a sophisticated arsenal of secondary metabolites mandatory. Functional regulation of this arsenal, as well as the extra leaf-specific primary metabolic functions (like photosynthesis and transpiration), are primarily responsible for the complexity of the leaf transcriptome compared to the root transcriptome.

It is evident that the numerical differences obtained in the leaf and root organ transcriptomes of C. roseus could not be directly ascribed only to the TIA diversity present in the two organs. However, the leaf-specific TDF subset could provide important clues related to the biosynthesis and accumulation of key leaf-specific TIAs upon downstream analysis. In this context it is relevant to note that despite functional divergence of the two organ-specific transcriptomes, there might exist homology at the genetic level. An example of this could be observed in the C. roseus enzymes—root-specific minovincinine-19-O-acetyltransferase (MAT) and leaf-specific deacetylvindoline-4-O-acetyltransferase (DAT), which have 63% nucleic acid identity and 78% deduced amino acid sequence identity, but different substrate specificities (Laflamme et al., 2001).

Although redundancy was not determined in the present cDNA-AFLP-based study, the number of leaf-specific TDFs (320) obtained was higher than the number of leaf-specific ESTs (55) obtained through a subtractive hybridization approach, in which leaf and root cDNA populations were used as “tester” and “driver,” respectively (Shukla et al., 2006). This demonstrates the higher efficiency of the cDNA-AFLP approach to identify differential transcripts. Also, next-generation sequencing methods may be adopted for comparative root and leaf deep transcriptome sequencing analysis, as was shown recently for Avena barbata (Swarbreck et al., 2011). In conclusion, the cDNA-AFLP analysis of root and leaf organs of C. roseus provides a direct and easy tool for discovering novel transcripts and genes expressing in an organ-specific manner. This study was carried out only to obtain a numerical estimate of the differences between the transcripts expressing in the leaf and root of C. roseus. To fully exploit the benefits of the open system cDNA-AFLP approach used in the present study vis-à-vis bisindole alkaloid biosynthesis, further characterization of the leaf-specific TDFs is required.