From these previous sections, it should be clear that the list of genes potentially relevant to the defence responses of trees is very large indeed. Here we focus on those pathways associated with inducing the emission of volatiles because of their emerging importance in plant-plant, and plant-insect signalling (Haukioja 2005).
Therefore, rather than attempt to discuss all the possible genes that may be related to the defensive strategies of trees, a brief introduction will be given into the biochemistry of volatile production by plants and the genes that control it, along with some discussion of the current methodologies for studying them. This section will then conclude with a brief discussion as to how these or similar approaches may be utilised for studying other aspects of the defence responses of trees and capturing the genes responsible for further study, including by the use of transformation approaches.
10.8.1 The Biochemistry and Genetics of Plant Volatile Emission
The plant volatiles emitted following herbivore feeding are usually different to those emitted following mechanical or other forms of damage (Dicke et al. 1999; Paré and Tumlinson 1999; Degenhardt et al. 2003; Dudareva et al. 2004; but contrary to Mithofer et al. 2005). Thus, plants seem to have a broad ability to recognize herbivore attack.
The process of herbivore feeding appears to release specific substances into the tissues of the damaged plant, which elicit direct and indirect defence responses (Turlings et al. 1995; Malone 1996). Plant responses to these stimuli are presumably regulated through a complex network of herbivore or pathogen specific receptors, as well as local and long distance wound signals, the precise combination of which serve to initiate pre-programmed but flexible defence responses (see Sect. 10.4), including insect specific volatile release.
The volátiles reported from herbivore-damaged plants belong to several different chemical classes. One large group are the terpenes (see below), whose structures are all based on the union of C5 isoprenoid units. Another frequently encountered class of volatiles are the so called "green leaf volatiles" (GLVs), which are released very rapidly after damage (Dudareva et al. 2004; D'Auria and Gershenzon 2005).
These six-carbon alcohols, aldehydes, and acetates (Loughrin et al. 1994; Turlings et al. 1998) are products of the lipoxygenase pathway, which begins with the oxidation of linolenic acid, just as in jasmonic acid biosynthesis, although many details of these biosynthetic pathways remain to be elucidated (D'Auria and Gershenzon 2005). Other induced compounds such as methyl salicylate and indole emanate from the shikimic acid/tryp-tophan pathway (Paré and Tumlinson 1997). It also now appears that in association with the plant hormone ethylene, GLVs may be the responsible agents for plant-plant communication (see Sect. 10.3.3), rather than terpenoids (Ruther and Kleier 2005) - a finding that is bound to increase the interest in characterizing the genes associated with GLV production. These genes seem to exist as large gene families, which raises further questions as to what their ecological or physiological functions may be (D'Auria and Gershenzon 2005).
The biosynthesis of terpenoids in higher plants meanwhile is much better understood, and proceeds through two parallel, but probably interconnected pathways (Fig. 10.1; Gershenzon and Kreis 1999; Dudareva et al. 2005), both of which generate the same C5 building blocks for isoprenoid formation, isopentenyl pyrophosphate (IPP) and its allylic isomer dimethylallyl diphosphate (DMAPP). One pathway occurs in the cytoplasm and converts acetyl CoA via mevalonate to IPP (the so-called MVA pathway), while the methyl erythritol (MEP) pathway occurs only in the plastids and uses pyruvate and glyceraldehyde-3-phosphate to generate IPP and DMAPP.
Once formed, IPP and DMAPP units can combine to produce geranyl diphosphate (GPP, C10), farnesyl diphosphate (FPP, C15) and geranylgeranyl diphosphate (GGPP, C20). These linear intermediates are then put through a wide range of cyclizations and rearrangements by individual terpene synthases (encoded by structurally conserved tps genes) and cytochrome P450s to produce the huge diversity of mono-(C10), sesqui-(C15), and di-terpene (C20) end products (Dudareva et al. 2004; Martin et al. 2004).
Many genes associated with the biosynthesis of terpenoids in plant species have been cloned and found to be up-regulated after injury and immediately prior to volatile emission begins, or are associated with other aspects of plant-insect signaling, such as scents from flowers (Pichersky and Gershenzon 2002; Degenhardt et al. 2003; Dudareva et al. 2005).
Despite the complexity of terpene biosynthesis, a large number of terpene biosynthetic genes have been isolated from several tree species (Bohlmann et al. 1997; Linden and Phisalaphong 2000; Martin et al. 2003; Arimura et al. 2004). The manipulation of terpene synthases for plant transformation has also been successfully reported on a number of occasions (Spencer et al. 1993; Degenhardt et al. 2003), making it reasonable to consider genetically engineering the terpene composition in any transformable species (Mahmoud and Croteau 2002), including for ecological studies as shown recently for ara-bidopsis (Kappers et al. 2005; Schnee et al. 2006). Thus, it is likely that the manipulation of tree species for altered terpenoid metabolism and ecological studies will soon follow.
The general approach proceeds as follows: (1) Genes putatively associated with terpene biosynthesis or other volatile emissions are isolated from a cDNA library made from a relevant plant tissue, either by a homology based cloning approach to known terpene biosynthetic genes (Bohlmann et al. 1997; Martin et al. 2004), or increasingly these days after reference to a relevant EST or genomic database. (2) These are then functionally expressed one at a time in a suitable microbial expression system and the activity profile of the encoded protein determine by in vitro assay (Miller et al. 2001; Faldt et al. 2003b; Martin et al. 2004). (3) Next, it must be decided whether the terpenes whose formation is regulated by this gene are relevant to the ecological interaction under study. (4) Transformation studies then follow, where one attempts to over or under express the gene(s) under study - probably in ara-bidopsis first as an easy to transform model species, then back into the native host where possible (as advocated by Schnee et al. 2006 with maize). (5) This finally enables ecological studies comparing transformants with wild-type controls. Potentially it should then be possible to study the variation of these genes at the population level, and so determine the selection pressures operating on them, but such studies have yet to be undertaken with forest trees.
Functional expression is necessary for many of the genes associated with the terpenoid pathway, especially terpene synthases (Fig. 10.1), and those associated with other secondary metabolic pathways also, as their precise function cannot be reliably ascertained from their DNA sequence identity alone (Martin et al. 2004). Homology based computer search programs such as Blast™ can give some indication as to a function for many terpene biosyn-thetic genes but not a definite characterisation. For instance, the very first identification of the important gene responsible for isoprene biosynthesis in poplars (iso) - a simple C5 terpenoid emitted mainly in response to abiotic stress (Logan et al. 2000) - was initially confused by it being listed in the EMBL database as a mere limonene synthase (a C10 monoterpene), based on sequence similarity (Miller et al. 2001).
It should be noted, however, that plant terpene formation can be manipulated by other means than just transformation; inhibitors are available that block the synthesis of terpenoids in the two different pathways, so allowing classical physiological approaches to experimentation (e.g. Dudareva et al.
Isopentenyl pyrophosphate (IPP)
IPP IPP - Mevalonate -Acetyl-CoA
Isoprene synthase* Isoprene (C5) ----- DMAPP
-Geranyl pyrophosphate (GPP)
ggpp synthase +3 x IPP
fpp synthase +2 x IPP
(FPP)-^ Sesquiterpenoids (C15)
Diterpenoids (C20) -Geranylgeranyl pyrophosphate
Polyterpenoids Farnesylated proteins
Triterpenoids (C30) Sterols
Fig. 10.1. A general scheme for the biosynthesis of terpenoids in plants. The cytosolic mevalonic acid (MAP) pathway for the biosynthesis of terpenoids is shown in blue, while the methylerythritol (MEP) pathway of plastids is shown in green. Selected terpenoid end products are shown in red. Abbreviations (in alphabetical order): DMAPP - dimethylallyl diphosphate; DOXP - 1-deoxy-d-xylulose 5-phosphate (or DXP); dxr - 1-deoxy-d-xylulose 5-phosphate reduc-toisomerase gene; dx - 1-deoxy-d-xylulose 5-phosphate synthase gene; FPP - farnesyl pyrophosphate; GPP - geranyl pyrophosphate; GGPP - geranylgeranyl pyrophosphate; IPP - isopentenyl pyrophosphate, or isopentenyl diphosphate (IDP); MEP - 2-C-methyl-d-erythritol 4-phosphate
2005). Furthermore, Kessler and Baldwin (2001) were able to achieve a ~90% reduction in the insect herbivores establishing on wild tobacco (N. attenuata) under field conditions, by painting on to plants selected terpene oils characteristic of insect damaged plants and believed to be attractive to insect predators.
Such approaches should not be viewed as being in competition to the use of biochemical, genetic and plant transformation techniques, however, but rather as complementary methodologies. When studying biological phenomena as complicated and diverse as plant defence and environmental responses seem to be, it is important to seek out the most appropriate techniques for investigating the scientific question at issue, rather than adopt one method and go looking for questions it can be used to answer.
In either case, a vital piece of equipment for any lab wishing to undertake such work is some type of volatile collecting system, which can later be analysed in detail by GC-MS or other techniques as appropriate (Millar and Haynes 1998; Dudareva et al. 2004; Tholl et al. 2006). Although portable GC equipment is now available such as the zNose™ (Electronic Sensor Technology, California), most workers still prefer to trap volatiles from headspace and then analyse them using high precision lab based equipment.
This approach enables changes in the volatile emission profiles of plants under treatment to be closely monitored, as well as tracking changes in the relative concentration of individual volatiles, provided they have been identified by authentic standards previously. The above approach is especially valuable for studying the potentially altered emission profiles of transgenic plants either in comparison to control plant lines, and/or when induced by various elicitors or herbivores, as recently performed with Arabidopsis by Kappers et al. (2005), and hopefully soon for field elms (Sect. 10.5) and maybe other tree species.
10.8.3 Further Approaches for Identifying Other Genes of Interest
The critical question when using transformation approaches to study plant defense responses, is which are the best genes to invest all the necessary effort and time in? Clearly, if one is already interested in a particular pathway or sub-set of genes these will be the ones to use, but this pre-supposes that such an interest has already been established. If one's work is driven by an interest in a scientific question and one wants to know which genes may be relevant to it, then other approaches are needed, since pre-selecting particular genes or pathways to study runs the risk of biasing the outlook of ones project towards what is already known before the work even starts.
Making cDNA libraries from experimental plant material is a good start, but these need to be subjected to some form of analysis to determine the gene expression profiles that they contain. If there are good genomic resources available for the tree species under study (e.g. Pinus taeda, Picea ssp., Populus spp.), then the cDNA samples can be subjected to micro-array analysis to identify which genes are activated in response to the treatment(s).
However, if this choice is not available, other approaches are needed: These can include a large scale sequencing effort of the cDNA libraries, which is effective but expensive. The construction of a subtractive cDNA library to reduce the sequencing effort runs the risk that differentially expressed families of genes with homologous coding regions may be eliminated, as discussed previously, or with some sort of differential display or cDNA-AFLP analysis, but executing these techniques effectively is tricky and prone to artefacts and still have a heavy sequencing requirement. Similar problems can plague the interpretation of micro-array results, but at least it is easier to repeat the experiments once such a micro-array is established.
Either way, it is still likely that too many 'genes of interest' will be highlighted by these approaches to incorporate them all, so further rounds of micro-array experiments or real-time PCR studies are recommended prior to closing off the list of which genes might be included in the transformation part of the program.
In particular, various compounds such as methyl jasmonate, ethylene, salicylic acid, as well as other elicitors specific to the insects or fungi of interest may be used, if known (Nürnberger and Scheel 2001; Pickett and Poppy 2001; Farmer et al. 2003; Belkhadir et al. 2004). This approach has the advantage that elicitors can be used to stimulate and study plant defense reactions in plants under more controlled conditions than is possible with ecological experiments involving field work with insects. This approach has already been used to study the defense responses of conifers (Martin et al. 2002; Hudgins et al. 2004) and elms (Sect. 10.5).
An extension of this strategy that may prove particularly useful for studying the defense responses and associated gene induction in trees, is to use seedlings or ex-vitro clonal propagules as a source of standardized material suitable for lab studies, e.g. with seedlings of P. abies (Kozlowski et al. 1999; Elfstrand et al. 2001; Fossdal et al. 2003; Pervieux et al. 2004), or even tissue cultures themselves including those of P. abies (Nagy et al. 2005).
Indeed, cell culture systems have been used to study the synthesis of plant secondary metabolites for many years (Hamill et al. 1986), including for studies into the biosynthesis of terpenoids (Spencer et al. 1993; Mahmoud and Croteau 2002; Ishida 2005). More recently this work has been extended to study plant defense responses at the molecular level after treatment with specific elicitors, e.g. for taxol production (Linden and Phisalaphong 2000), as well as other defense related metabolites with other plant cell culture systems (Cane et al. 2005; Zhao et al. 2005).
Indeed, investigations into the altered responses of transformed cell lines can also begin at the cell culture level including with conifers (Elfstrand et al. 2001; Levée and Séguin 2001), speeding the flow of results and hypothesis testing of a trees defense responses and also possibly reducing the need to regenerate every transgenic cell line.
Because of this, there is every reason to believe that such approaches will be fruitful avenue for defence related gene discovery and characterisation in trees. This may also be a promising approach for the preliminary characterisation of defence related genes, metabolites and signalling compounds or elicitors with the caveat that the defence responses of adult trees may not be fully functional or active in such cell culture systems, so the need to examine and test the full trees responses at some stage still remains.
Was this article helpful?