Chapter One

CHAPTER ONE ____________________________________________________________________ INTRODUCTION

1.1

Background

According to Avé (1977), the sago palm was one of the first plants used by man in South-east Asia and Oceania. Traditionally, the trunk of the sago palm was used to obtain starch as a staple food for humans (Flach, 1984). The renewable raw material, starch of sago palm is tasteless and is usually flavored with other foodstuffs.

The sago palm, a crop that grows well in swampy areas has great potential for production of starch in Malaysia. It has been claimed that the crop can yield up to 37 tonnes of starch/ha/year, one of the highest for starch-producing crop in the world (Jong, 1995). The profitability of sago palm cultivation includes: it requires no fertilization, it has few natural pests or diseases and it can be grown in areas where it is impossible for other crops to thrive (Abd-Aziz, 2002). Furthermore, by growing the sago palm instead of other crops that involve the drainage of the peat swamps, the benefits of such swamps including preventing floods and droughts and maintaining a source of clean ground water can be retained (Abd-Aziz, 2002).

A considerable amount of research has been done on the morphology and physiology of flower development in the sago palm including by Tomlinson (1971), Utami

1

(1986), Beccari (1918) and Tomlinson (1990) as cited by Flach (1997). However, little work in isolating genes involved in the floral development process has been conducted. The identification of the genes involved in the floral development process will provide knowledge on molecular mechanism underlying the floral development process and thus provide useful tools for crops improvement. This study was carried out in the effort to provide a conceptual understanding of the molecular basis of flower induction and development of the starch producing tree, sago palm.

1.1.1

Research Objectives

The main objectives of this study were as follows:

(i)

Screen for LEAFY and CONSTANS genes via RT-PCR.

(ii)

Characterize the differentially expressed genes in meristem, inflorescence and leaf tissues of sago palm.

(iii)

Isolate and characterize flower-predominant chitinase gene.

2

1.2

Literature review

1.2.1

Sago palm

The sago palm (Metroxylon sagu) is belongs to the Lepidocaryoid subfamily of the Arecaceae (Palmae). It is once-flowering (hapaxanthic) and tillering or suckering perennial. The flowers of sago palm are borne spirally in pairs on the tertiary axis. Of each pairs of flower, one is male and the other complete but only functionally female (Flach, 1984). This plant found mainly in Papua New Guinea, Indonesia, Malaysia, Thailand, Philippines and Pacific Islands.

There are a number of characteristics of this crop that makes it a remarkable plant. According to Stanton (1993), the advantages of the crops are that it is economically acceptable, relatively sustainable, environmentally friendly, uniquely versatile, vigorous and promotes socially stable agroforestry systems. The innumerable uses of the sago palm are presented in figure 1.1.

Sago starch has a multitude of uses. In Sarawak, sago starch is widely used to produce sago pearls and "tabaloi", a local biscuit delicacy. It is also used in production of bread flour (Dendy et al., 1970; Clarke et al., 1980), the production of high fructose syrup (Ito et al., 1979), the manufacture of noodles, monosodium glutamate industry and the glue industry involved in plywood manufacture (Doelle, 1998).

3

Figure 1.1:

The many uses of the sago palm (Baay, 1983 taken from Flach, 1984).

4

Sago palm starch is almost pure carbohydrate. It consists of 27% amylose and 73% amylopectin (Ito et al., 1979). The granular size of sago palm starch ranges from about 80µm to 5µm, with an average of about 30µm. About 90% of the starch has a particle size of between 20 and 40µm. Only the granular size of potato starch is the same, in all other starch it is smaller (Griffin, 1977).

Sim (1977) points out that sago starch has several advantages over other starches such as it produces sizing pastes of lower viscosity at a given concentration than such pastes from maize and potato. In addition, it was easy to handle because of the sago pastes are less inclined to gelate under cooling than maize pastes. The sago pastes also show low retrogradation in which their stability in viscosity is high when kept for long periods at near boiling point that provided they are boiled for two hours before use.

The main impediment to the development of sago palm as a regular industrial crop is its long period of immaturity. Therefore, it takes a long time before the harvesting process in order to processing the sago starch. The trunk is judged to be ready for processing in order to get starch by the stage of flowering. The reason is the trunk is supposed to have reached its maximum starch content when the young fruits are developing (Flach, 1984).

Apart from that, the maturation period of sago is important because sago planters’ have used the flowering stage as an indicator for logging to occur. Probably at

5

flowering, the starch in the lowest part of the trunk has shifted to the top (Kraalingen, 1983). In the lowest part of the trunk, the number of vascular bundles increases, and the bundles also become harder. This may explain why the traditional processor waits until most of the starch from the lowest part of the trunk has been shifted in preparation of flowering to the upper part before he harvests the trunk (Flach, 1984).

Though there are no exact measurements of the length of the growing cycle of the sago palm from seed to next generation of seeds. Reports on the length of the life cycle in the literature are ranging from 8 to 17 years (Flach, 1984). Figure 1.2 show the relationships between palm age and starch accumulation.

Figure 1.2:

The relationships between palm age and starch accumulation. Adapted from Flach (1984). The sago palm,p.14.

6

1.2.2

Floral development

Flower is a part of the plant that allows sexual processes. The transition to flowering marks the end of vegetative growth (during which shoots and leaves are produced) and the beginning of reproductive development. In many plant, such as those with single flowers, flower transition also signifies the end of indeterminate growth (Howell, 1998). There is an enormous literature on the control of flowering, however, a unified physiological model has not emerged (McDaniel, 1996).

Studies from the model plant Arabidopsis thaliana which was the first plant to have its genome sequenced (The Arabidopsis Genome Initiative, 2000) showed that flowering is regulated by three classes of genes acting in consecutive order: the flowering time genes, the floral meristem identity genes and the floral organ identity genes (Piñeiro & Coupland, 1998).

The flowering time genes including CONSTANS (CO), TERMINAL FLOWER 1 (TFL1), and GIGANTEA (GI) are those that display major effects on the duration of vegetative development (Tsaftaris et al., 2004). The genetic analysis of more than 80 different flowering-time genes that have been discovered in Arabidopsis thaliana (Blazquez et al., 2001) has led to identification of four major pathways controlling flowering time which are the photoperiod, the vernalisation, the autonomous and the gibberellin pathway (Araki, 2001; Mouradov et al., 2002; Simpson & Dean, 2002; Bastow & Dean, 2003; Komeda, 2004; Parcy, 2005). The photoperiod and the vernalisation pathways mediate flowering in response to environmental factors such

7

as day length and low temperature, whereas the autonomous and the gibberellin pathways mainly act independently of these external signals (Mouradov et al., 2002).

Floral meristem identity genes are involved in switching the fate of meristems from vegetative to floral phase The best characterized of these genes are APETALA1 (AP1), APETALA2 (AP2), CAULIFLOWER (CAL) and LEAFY (LFY) (Tsaftaris et al., 2004).

Floral organ identity genes (homeotic genes), which fall into three classes: A, B and C class genes, determine the fate of floral organ primordia. These genes have been best characterized in Arabidopsis, snapdragon, and petunia (Zachgo et al., 1995). In all three species, flowers are composed of four concentric rings (whorls) of organs, with sepals in the first, outermost whorl, followed by petals, stamens, and carpels in whorls 2, 3, and 4, respectively. The ABC model explains how floral organ identity genes act combinatorially to specify each of the four organ identities (Bowman et al., 1991; Coen & Meyerowitz, 1991). The class A genes lead to the formation of sepals, class A and B genes together lead to the formation of petals, the class B and C genes specify the formation of stamens, and the class C genes are required for the formation of carpels (Irish, 1999; Ma, 2000; Theissen, 2001). Table 1.1 shows floral homeotic genes in Arabidopsis.

Table 1.1:

Floral Homeotic Genes in Arabidopsis. Adapted from Howell (1998). 8

Arabidopsis thaliana

Class

Genes

A

APETALA 1 APETALA 2

B

APETALA 3 PISTILLATA

C

AGAMOUS

Although there are numerous genes involved in the control of flower formation, this study was focus on CO and LFY gene which is the example of the flowering time gene and floral meristem identity gene respectively.

The Arabidopsis CO gene is a key regulator of the long day (LD)-dependent flowering pathway. CO encodes a zinc finger protein that recognized as a genetic component of the LD-dependent flowering pathway (Putterill et al., 1995). The CO mutant flowers later than the wild type under LD, but shows similar flowering time to the wild type under short day (SD). Consistently, the CO gene shows higher expression under LD than SD during the day and over expression of CO causes early flowering even under SD (Putterill et al., 1995; Suarez-Lopez et al., 2001).

LFY gene is necessary for transition to reproductive growth and the concomitant formation of flowers. Loss of LFY function leads to leaves and shoots in place of flowers, while constitutive expression of LFY results in precocious floral development (Weigel et al., 1992; Weigel & Nilsson, 1995). LFY is also required for the transcription of representatives of all three classes of ABC genes (Weigel & 9

Meyerowitz, 1993). It encodes a nuclear-localized product that can bind to DNA and so could act directly to regulate transcription of the floral homeotic genes (Parcy et al., 1998).

1.2.3

Chitinase gene

Chitinases are the protein (enzyme) that catalyses the hydrolysis of the β-1, 4linkages of N-acetyl-D-glucosamine polymer of chitin; a major component of the exoskeleton of insects, of crustacean shells and of the cell wall of many fungi (Nishizawa et al., 1993; Bishop et al., 2000; El-Sayed et al., 2000; Passarinho & de Vries, 2002). They rapidly hydrolyzed swollen and regenerated chitin, slowly hydrolyzed glycol chitin, however unable to hydrolyze N, N diacetyl chitobiose (Hirano et al., 1988; Leah et al., 1991). Chitinases are present in many species of higher plants, although higher plants themselves contain no chitin, chitosan or chitin like substrate (Boller et al., 1983; Hirano et al., 1988).

Chitinase genes have been classified in families 18 and 19 based on its amino acid sequence similarity of the catalytic domains in which chitinase genes of family 18 are present in bacteria, fungi, yeast, viruses, plants and animals whereas chitinase genes of family 19 are almost exclusively present in plants (Henrissat, 1991; Passarinho & de Vries, 2002). The differences in the sequence and structure suggesting that chitinase genes of family 18 and 19 are arisen from a different ancestor (Hamel et al., 1997). In addition, the chitinase genes of both families are differed in several of their biochemical properties (Hart et al., 1995; Iseli et al., 1996;

10

Ohno et al., 1996; Brameld et al., 1998; Brameld & Goddard, 1998; Garcia-Casado et al., 1998).

Plant chitinases are organized in classes numbered from I to V according to their sequences and primary structures (Collinge et al., 1993; Neuhaus et al., 1996). Chitinase genes of classes I, II and IV belong to the family 19 whereas chitinase genes of classes III and V belong to family 18 (Passarinho & de Vries, 2002). Chitinase genes of class I have a highly conserved N-terminal cysteine-rich region which is separated from the catalytic domain that often followed by a C-terminal extension by a short proline-rich variable hinge region (Neuhaus et al., 1991; Iseli et al., 1993). N-terminal cysteine-rich region is involved in chitin-binding whereas Cterminal extension is involved in vacuolar targeting (Neuhaus et al., 1991; Iseli et al., 1993). Chitinase genes of class II have a catalytic domain with a high sequence and structural similarity to that of chitinase genes of class I, however, lack of both the Nterminal cysteine-rich region and the C-terminal extension (Passarinho & de Vries, 2002). Chitinase genes of class IV have a very similar main structure with chitinase genes of class I, but they are significantly smaller due to four deletions distributed along the chitin-binding domain and the catalytic region (Passarinho & de Vries, 2002). According to Graham & Sticklen (1994), chitinase genes of class III are more similar to fungal and bacterial chitinase genes than to chitinase genes of other plant. Chitinase genes of class V have a C-terminal extension and may contain a chitinbinding domain as well (Heitz et al., 1994; Ponstein et al., 1994). In addition,

11

chitinase genes of both class III and class V display an additional lysozymal activity (Majeau et al., 1990; Heitz et al., 1994).

Chitinases are often classified into pathogenesis-related proteins since their expressions, which are constitutively at low levels, are dramatically induced in response to fungal, bacteria or viral infections (Graham & Sticklen, 1994; Neuhaus, 1999; van Loon, 1999). The expression of chitinase genes also can be induced in response to general sources of stress such as wounding, salicylic acid and ethylene or elicitors such as fungal and plant cell wall components (Graham & Sticklen, 1994; Zhou, 1999; Leon et al., 2001). Pathogen related proteins, chitinases play role in plant defense by damaging the structures of the parasites (Schlumbaum et al., 1986; Roberts et al., 1988; Bishop et al., 2000; Odjakova & Hadjiivanova, 2001). In addition, they can work indirectly by releasing oligosaccharides that can act as elicitors to activate other plant defense responses (Shibuya & Minami, 2001).

Several studies revealed that some chitinase genes are expressed at higher levels in healthy floral organ such as flower flower-predominant chitinase genes of potato (Wemmer et al., 1994) and tomato (Harikrishna et al., 1996). The expression of chitinase genes in flowers also have been detected in Arabidopsis thaliana (Samac et al., 1990; Passarinho et al., 2001), petunia (Leung, 1992), parsley (Ponath et al., 2000), rice (Takakura et al., 2000) and tobacco (Lotan et al., 1989; Trudel & Asselin, 1989; Neale et al., 1990). Moreover, the expression of chitinase genes are found in roots of Arabidopsis thaliana (Samac &Shah,1991), rice (Lamb et al., 1991) and

12

tobacco (Memelink et al.,1990; Neale et al., 1990); and in embryogenic cultures of carrot (van Hengel et al., 1998) and spruce (Egertsdotter, 1996; Dong & Dunstan, 1997) as well. In other plants such as barley (Leah et al., 1994), carrot (van Hengel et al., 1998), pea (Petruzzelli et al., 1999) and soybean (Yeboah et al., 1998) the presence of chitinase genes are detected in seeds.

The developmentally-regulated expression of some chitinase genes suggests that these genes which are conventionally considered as pathogen-related proteins could also play a role in normal developmental processes in healthy plants (de Jong et al., 1992; de Jong et al., 1993; Kragh et al., 1996; Baldan et al., 1997; Patil & Widholm, 1997; Neuhaus, 1999; Ovtsyna et al., 2000; Gomez et al., 2002).

1.2.4

Differential display reverse transcription polymerase chain reaction

(DDRT-PCR) Differential display reverse transcription polymerase chain reaction (DDRT-PCR) is a powerful technique that widely employed for identification of differentially expressed genes in various cells or under different conditions (Liang & Pardee, 1992; Bauer et al., 1993; Liang et al., 1993). The technique which is first described by Liang & Pardee (1992) to compare messages that differ between normal and tumorigenic cells has allowed the identification of a number of differentially expressed genes in fundamental biological processes such as cell and tissue differentiation,

cellular

responses

to

various

stimuli,

embryogenesis

and

13

organogenesis and long-term plasticity of the nervous system (Guimaraes et al., 1995; Alves et al., 1998; Wen, 2000).

Subtractive hybridization is the only available technique that can be used to identify differentially expressed genes prior to invention of DDRT-PCR (Alves et al., 1998). The usefulness of the subtractive hybridization is limited because large amounts of mRNA are required in order to perform the technique (Alves et al., 1998; Sturtevant, 2000). The technique is also time consuming, tedious and difficult to perform (Alves et al., 1998). The major advantage of DDRT-PCR as compared to subtractive hybridization is the technique can be performed with relatively small amounts on RNA (Alves et al., 1998; Sturtevant, 2000; Kim et al., 2004). In addition, the technique is claimed to be simple, rapid and sensitive (Alves et al., 1998; Sturtevant, 2000; Chen et al., 2004; Kim et al., 2004).

The DDRT-PCR consists of two stages which are the generation of cDNA pools using total RNA isolated from different cell populations as a template and an anchored oligo(dt) as a primer in a process called reverse transcription and the amplification of the resulting cDNA via PCR using the same anchored oligo(dt) primer in combination with an upstream arbitrary decamer (Alves et al., 1998; Sturtevant, 2000; Wen, 2000). Separation of the DDRT-PCR amplicon from two or more samples on adjacent lanes of sequencing gel has allowed the detection of differentially expressed amplicon from those samples. The bands corresponding to the differentially expressed amplicon are excised, cloned and further analysed.

14

As the DDRT-PCR is simple, rapid and sensitive, it has become a popular technology in gene expression work. In 1998, Mathews & Heinz reported the use of the technique for identification of over 50 genes in soybean roots in response to cyst nematode infection. The technique also employed in the study of developmentally regulated genes under stressful conditions in plants (Knaap & Kende, 1995; Momiyama et al., 1995; Sharma& Davis, 1995; Tsengh et al., 1995; Wilkinson et al., 1995; Tieman & Handa, 1996; Alves & Vantoai, 1997). However, the technique does have its limitations such as false positive (Sun et al., 1994; Debouck, 1995; Wan et al., 1996) and poor reproducibility of the result (Liang & Pardee, 1995). In addition, it is unclear how well low abundance mRNAs are represented (Bertioli et al., 1995). A number of modifications to overcome the problems have been introduced, including the use of the improved gel resolution systems (Averboukh et al., 1996), the use of longer primers in combination with two-step PCR cycle (Zhao et al., 1995) and the use of cytoplasmic RNA to avoid unprocessed mRNA (Bauer et al., 1993; Sompayrac et al., 1995; Luce & Burrows, 1998).

15