Comprehensive Summaries of Uppsala Dissertations from the Faculty of Science and Technology 680
_____________________________ _____________________________
Differentiation and Pathogenicity
within the Saprolegniaceae
Studies on Physiology and Gene Expression Patterns in Saprolegnia parasitica and Aphanomyces astaci
BY
Dissertation for the Degree of Doctor of Philosophy in Physiological Mycology presented at Uppsala University in 2002
Abstract
Andersson, M. G. 2001. Differentiation and Pathogenicity within the Saprolegniaceae. Studies on Physiology and Gene Expression Patterns in Saprolegnia parasitica and Aphanomyces astaci. Acta Universitatis Upsaliensis. Comprehensive Summaries of Uppsala Dissertations from the Faculty of Science and Technology 680, 41 pp. Uppsala. ISBN 91-554-5203-5.
Saprolegnia parasitica and Aphanomyces astaci are parasitic water moulds belonging to the Oomycetes. Despite their importance as parasites they are very little studied at the molecular level and the work described in this thesis was aimed at increasing the molecular knowledge of these organisms by cloning and characterising genes of potential importance for reproduction and pathogenicity.
Stage-specific transcripts from Saprolegnia parasitica were isolated by differential display RT-PCR. One of the markers, puf1 encodes a putative mRNA binding protein which may be involved in post-transcriptional regulation of gene expression. S. parasitica puf1 is expressed exclusively in spore cysts that have not been determined for germination or repeated zoospore emergence indicating that the cyst stage has two phases, of about equal duration, which are physiologically and transcriptionally distinct. A similar expression pattern is observed in Aphanomyces spp. with different regulation of spore development and in the transcript is detected in both primary and secondary cysts.
A putative chitinase AaChi1, was cloned from the crayfish plague fungus, Aphanomyces astaci. Analysis of chitinase activity and AaChi1 expression showed that chitinase in A. astaci is constitutively expressed in growing and sporulating mycelia, but absent in zoospores, a pattern which reflects the infectious life cycle of A. astaci. This expression pattern is conserved between the four known genotypes of A. astaci, in contrast to saprophytic and fish-pathogenic Aphanomyces spp.
Genetic and physiological analysis were conducted on five strains of Aphanomyces, isolated from suspected outbreaks of crayfish plague in Spain and Italy. The strains are not virulent against freshwater crayfish, and RAPD PCR and ITS sequence analysis show that they are unrelated to the crayfish plague fungus, A. astaci.
M. Gunnar Andersson, Department of Comparative Physiology, Evolutionary Biology Centre, Uppsala University, Norbyvägen 18A, SE-752 36, Uppsala, Sweden
© 2001 M. Gunnar Andersson ISSN 1104-232X
ISBN 91-554-5203-5
Preface
The thesis is based on the following papers, which will be referred to by their Roman numerals:
I. Andersson, M. G. and Cerenius, L. (2002). Pumilio homologue from
Saprolegnia parasitica specifically expressed in undifferentiated spore
cysts. Eukaryotic Cell, in press.
II. Andersson, M. G. and Cerenius, L. (2002). Comparison of pufI expression in Aphanomyces spp. with different regulation of germination. (in manuscript).
III. Andersson, M. G. and Cerenius, L. (2002). Analysis of chitinase expression in the crayfish plague fungus, Aphanomyces astaci. (Submitted)
IV. Royo, F., Andersson, G., Bangyeekhun, E., Cerenius, L., Múzquiz, J. L. and Söderhäll, K. (2002). Physiological and Genetic Characterisation of some Aphanomyces Strains Isolated from Freshwater Crayfish. (in manuscript)
The manuscript I was reproduced with the permission from the American Society for Microbiology.
Contents
Abstract 2 Preface 4 Table of contents 5 Abbreviations 6 Introduction 7 Background 8 Parasitic fungi 8 Oomycetes 9 Zoospores 10 Mechanisms of infection 12Host response to fungi 15
Molecular mechanisms of parasitic adaptations 16
Results and discussion 18
Identification of stage specific transcripts (I). 18 A putative RNA binding protein (II) 21
puf1 is specific for undifferentiated cysts (I,II) 24
puf1 in species with other regulation of germination (II) 24 Chitinase in the crayfish plague fungus, A. astaci (III,IV). 25 New Aphanomyces strains from dead crayfish (IV). 28
Conclusions 29
Acknowledgements 31
Abbreviations
AaChi1 A. astaci chitinase 1
CSP Conserved sequence flanking Puf-repeats ddRT-PCR Differential display RT-PCR
Ef-1 Elongation factor 1 EST Expressed sequence tag
FBF fem-3 binding factor
IpiO In planta induced "O"
ITS Internal transcribed spacer LPS Lipopolysaccharide
mst Mycelium specific transcript NAG N-acetylglucoseamine
NRE Nanos responsive element PCR Polymerase chain reaction PkaC Protein kinase C
PME Point mutation element ProPO Prophenoloxidase
Puf-family/protein Pumilio and FBF family/protein.
Puf1 Saprolegnia parasitica Puf like protein 1.
RAPD Random amplification of polymorphic DNA RT-PCR Reverse trancription PCR.
SAPS Secreted aspartyl proteases SAR Systemic aquired response
sst Spore specific transcript
Introduction
Saprolegnia parasitica and Aphanomyces astaci are pathogenic water molds,
belonging to the oomycete family Saprolegniaceae. Despite their importance as pathogens on plants and animals Oomycetes have been little studied at the molecular level and the work presented in this thesis is aimed to increase the knowledge of the molecular biology of these organisms.
The main, and often only, infectious stage of these organisms is the asexually formed zoospore, and by studying gene expression in the zoospore we hoped to identify transcripts of potential importance for reproduction and pathogenicity. We used differential display RT-PCR to identify transcripts specific for the zoospore and mycelium stages of S. parasitica. One of the identified transcripts, puf1, encodes a putative mRNA binding protein, and is specifically expressed in encysted zoospores.
Another approach for studying pathogenesis is to study genes encoding enzymes of potential importance for the infection process. Using PCR with primers based on conserved chitinase domains a putative chitinase was cloned from the crayfish plague fungus, A. astaci, and the expression of this transcript was analysed under different growth condition and during different stages of the life cycle.
This work also includes a genetical and physiological study of five strains of
Aphanomyces, which had been isolated from freshwater crayfish in Spain and
Background
Parasitic fungi
The question what makes a fungus or another microorganism, a parasite or a pathogen is not simple, and there are multiple definitions of virulence and pathogenicity (Casadevall & Pirofski 1999). What adaptations will result in a successful parasitic life, why will a fungus be a pest on one host while it is harmless on another, often closely related host and how can a strain considered to be an opportunist, endophyte or saprophyte turn out to be a pathogen when brought into a new environment? Several attempts have been done to address these questions, in different host pathogen systems. In some cases mechanisms have been found that are unique to a particular host-parasite system, and maybe even attributed to a single gene. It may be the presence or absence of a gene product that is recognised by receptors in the host (Dangl & Holub 1997), the production of a potent toxin (Vilcinskas & Götz 1999a), or a gene product that inactivates the defence of the host (Vilcinskas & Götz 1999a). Adaptations to parasitism could also be more general treats that cannot easily be attributed to a single gene, such as adaptations in the regulation of life cycle or development or the formation of lytic enzymes. The interaction may often be a balance of offensive and defensive actions, so that a fungus producing the optimal mix of degrading enzymes, toxins etc may be virulent against one host because it is so optimised that it can successfully establish an infection, while it fails to infect a related species or cultivar. In such a case it will be almost impossible to correlate virulence with a particular gene (Schulz et al. 1999).
Oomycetes
Not all classes of fungi-like pathogens belong to the kingdom fungi. One of these groups is the oomycetes. This group comprises severe pests, like
Phytophthora infestans (Judelson 1997, Tyler 2001) causing the potatoe late
blight, A. astaci (Cerenius et al. 1988), causing crayfish plague and several fish pathogens from genera Aphanomyces, Achlya, and Saprolegnia (Cerenius & Söderhäll 1996). There is also at least one species with the potential of infecting humans, namely Pythium insidiosum (de Cock et al.1987; Mendoza
et al. 1993). Oomycetes are placed within the kingdom Chromista together
with for example brown algae. The Oomycetes are subdivided in orders (Fuller 1987). One being the Peronosporales comprising for example
Phytophthora spp which are parasites on among others potato and tomato,
and Peronospora. The Peronosporales often infect land living plants. Another order is the Saprolegniales, a group of water molds that are common in aquatic environments both as saprophytes and parasites on fish and crustaceans (Cerenius & Söderhäll 1996). The Saprolegniales also comprise specialised plant pathogens, such as Aphanomyces euteiches and Aphanomyces
cochlioides (Deacon 1996). In contrast to true fungi Oomycetes contain little
chitin in their cell walls, which instead is mainly composed of β-1-3-glucans (Söderhäll & Unestam 1979). The most well studied Oomycetes at the molecular level are plant parasites. One example is the study of avirulence genes from Phytophthora and Pythium that triggers gene-for-gene resistance in plants (Ching et al.1998; Lauge & de Wit 1998). These works have revealed the involvement of a particular class of proteins, elicitins (Yu 1995) that may act both as a virulence- and as avirulence factors. They are called elicitins due to their capacity for inducing defence responses in resistant plants. Among the animal pathogens the previously mentioned A. astaci has been studied for a long time. This parasite usually form a nonlethal chronical infection in several North American crayfish species, like the signal crayfish
Pacifastacus leniusculus while it causes very high moralities in European
species, including the noble crayfish, Astacus astacus (Unestam & Weiss 1970; Unestam 1972). There are also examples of A. astaci outbreaks in signal crayfish, when they are subjected to stress or in combination with a second infection (Persson & Söderhäll 1983; Persson et al. 1987).
Zoospores
Most Oomycetes form sexual spores, which may also serve as a resting stage. In many animal-pathogenic Saprolegniales though, no sexual stages have been observed (Beakes et al., 1994; Noga, 1993). Instead the main and often only infectious units are the asexually produced zoospores, (Cerenius & Söderhäll 1984a;1996) and it has been shown that suppression of the zoospore stage can prevent spreading of disease (Rantamäki et al. 1992). Because of the key role of zoospores in disease initiation much of the research on Oomycetes is focused on different aspects of the asexual life cycle. The details of the life cycle vary between the different genera (fig. 1). In
Saprolegnia spp. primary zoospores are formed in sporangia at the end of
hyphal tips (Heath 1987). The primary zoospore generally encysts after a short time and a secondary zoospore is released from primary cyst. As the secondary zoospore is the most long-lived stage and the most important for infection and dispersal it is also by far the most well studied. In the related
Aphanomyces spp the sporangium is a converted hyphal segment, almost
indistinguishable from normal hyphae, and the primary spores immediately encyst at the hyphal tip. Thus the secondary zoospore is the only motile stage in these species (Cerenius et al.1987).Yet another variation is seen in the
Peronosporales where undifferentiated sporangia may either germinate
directly or differentiate towards zoospore formation (Fuller 1987). When zoospores encyst they shed their flagella and within minutes they form a cell wall of glycoproteins. The spores attach to solid substrate by an adhesive material secreted from vesicles at the ventral part (Deacon & Donaldson 1993). Encystment is typically followed by germination, that is outgrowth of a hypha, but several species of Oomycetes including A. astaci (Cerenius & Söderhäll 1984b; Cerenius & Söderhäll 1985), and some species of
Saprolegnia (Diéguez-Uribeondo et al. 1994a) may under some conditions
release a new zoospore generation, leaving behind its old cyst coat, in a process named repeated zoospore emergence (Cerenius & Söderhäll 1985). The ultrastructure of zoospores have been described in detail (Cerenius et al. 1984; Olson et al. 1984; Deacon & Donaldson 1993). In summary zoospores are mononucleate, have two flagella, lack cell wall and communicate with the surrounding water through a water expulsion vacuole (Mitchell & Hardham 1999, Cerenius, Rennie & Fowke 1988) The properties of Oomycetes give rise to special problems in molecular studies, reviewed by Judelson (1997). Sexual stages are not always known and when they do the spores often arise from self- fertilisation.
Fig. 1: Asexual life cycle of Saprolegnia spp., Aphanomyces spp. and Phytophthora spp. I n Saprolegnia two different motile zoospore stages, with different physiology and
morphology, are present while in Aphanomyces and Phytophthora only zoospores
resembling secondary zoospores are produced. In Phytophthora undifferentiated sporangia
may be released from the mycelium, and may subsequently germinate or differentiate into a mature sporangium. Repeated zoospore emergence, that is the production of several generations of secondary zoospores, occurs in several parasitic and opportunistic strains
of oomycetes. In S. parasitica a recently formed secondary cyst is undetermined and may
either be triggered to germinate by the addition of nutrients during the first hour after encystment, or otherwise release a new zoospore. In contrast, the cysts of many species of
Aphanomyces, including A. astaci, becomes determined for zoospore release of germination
at the time of encystment, depending on the mode of encystment.
Consequently genetic approaches are often difficult or impossible. To further complicate the situation Oomycetes, in contrast to many true fungi, are diploid throughout the life cycle, making it hard to study mutants. Yet another problem is that successful transformation requires oomycete-derived promotors, of which only a few have been characterised so far (Judelson 1997). Even when successful transformations have been made the traditional ways of generating knockouts have faced problems because homologous recombination rarely takes place between the introduced DNA and the genome (Judelson 1997). The consequence of this is that the study of
Oomycetes progressed slowly compared with some model Ascomycetes, where genetic approaches were possible. Molecular knowledge on zoospores is thus largely restricted to biochemical characterisation, using monoclonal antibodies and lectins, summarised by Burr & Beaks (1994). With these approaches it has been possible to identify and characterise the vesicles responsible for deposition of the cyst coat and release of adhesive material (Deacon & Donaldson 1993; Burr & Beaks 1994) and to isolate and clone proteins from zoospore vesicles (Marshall et al. 2001). The experiments have also revealed biochemical differences between zoospores from different classes of Oomycetes and given some suggestions about the localisation of receptors involved in chemotaxis and encystment (Deacon & Donaldson 1993, Burr & Beaks 1994). The pathways regulating encystment and germination are poorly understood. However there are multiple lines of evidence that fluxes of Ca2+ play an important role in the early signalling (Donaldson & Deacon 1992), and Ca2+ can induce germination in vitro of zoospores from many oomycetes. It is also suggested that posphatidic acid (PA) acts as a second messenger (Zhang et al.1992; Deacon & Donaldson 1993). Only few attempts have been made to study gene expression in zoospores. Using indirect approaches such as in vitro translation and DNA hybridisation Gwynne & Brandhorst (1982) could show that substantial changes in mRNA expression take place during spore differentiation in
Achlya ambisexualis. After the zoospores have been released from sporangia
RNA and protein synthesis appear to be low (Söderhäll & Cerenius 1983; Penington et al 1989), and it is has consequently been assumed that the zoospores from that timepoint largely rely on stored macromolecules until the onset of germination.
Mechanisms of infection
For an infection to be established a number of steps have to be completed successfully and adaptations to parasitism, general or host specific, can involve one or several of these stages. Although fungi and Oomycetes are phylogenetically distant, many features of life cycle and pathogenic strategies, for example zoospore and appressorium formation, are present in both groups, probably as a result of convergent evolution (Tyler 2001).
First the pathogen must find and attach to its host. This could be a passive process as with the conidia of many fungi. In Oomycetes zoospores are attracted chemotactically to potential sources of nutrients in a manner similar to bacterial chemotaxis (Deacon & Donaldson 1993, Cerenius & Söderhäll, 1984a). While dispersal is largely a passive process also in zoosporic fungi the significance of swimming lies in selecting a site of encystment (Deacon
1996). Some attractants have been characterised chemically and may for example be some amino acids (Rand & Munden 1993), flavones (Tahara et
al. 1999), aldehydes and fatty acids (Cameron & Carlile1978). Chemotaxis is
likely to facilitate the finding of a host but may not contribute significantly to host specificity, since spores are also attracted to non-host substances. However it may help finding a wound or scratch (Diéguez-Uribeondo et al. 1994b). Since penetration often is most successful at joints, wounds and body openings (Unestam & Weiss 1970; Nylén & Unestam 1980) chemotaxis may help to find right spot of infection and could thus contribute to a successful infection.
The second step is adhesion. This often requires little more than a hydrophobic surface (Sing & Bartnicki-Garcia 1972, Svensson 1978), but requirements may be more specific, for example Lagenidium giganteum encysts on films of chitosan and chitin (Petersen et al 1997; Kerwin et al. 1991). In Oomycetes adhesion is associated with encystment and formation of a cellwall (Deacon & Donaldson 1993). The spore or conidia often attach to the substrate by secreting an adhesive material. In true fungi hydrophobins may play a role in adhesion (Ebbole 1997; Bidochka et al. 2001) while Oomycete zoospores secrete glycoproteins that may serve as adhesive material (Sing & Bartnicki-Garcia 1975, Deacon & Donaldson 1993).
The third step is germination, meaning the outgrowth of hyphae or penetration structures from the spore, or conidia. In many Oomycetes the trigger to encyst appears to be separate from that to germinate. The latter is often a nutrient, (Petersen et al 1997; Diéguez-Uribeondo et al. 1994b). For the highly specialised parasite A. astaci exudate or substances from the host are highly stimulatory for triggering germination, while other nutrient triggers are inefficient. (Svensson 1978). Several pathogenic species of Oomycetes including A. astaci (Cerenius & Söderhäll 1985), but also some opportunistic Saprolegnia spp. (Diéguez-Uribeondo et al. 1994a) will undergo repeated zoospore emergence if the place of encystment does not promote germination. The association of this ability with parasitic species have led some authors to suggest that it is an adaptation for parasitism (Cerenius & Söderhäll 1985). As this mechanism, in contrast to chemotaxis and encystment appears to be specifically regulated it may make an important contribution to host recognition (Cerenius & Söderhäll 1986).
To infect the germinating spore must then penetrate the host barriers. This can be achieved by mechanical force (Nylén & Unestam 1975), the action of lytic enzymes (Clarkson & Charnley 1996; Persson et al. 1984; Unestam 1966; Vilcinscas & Götz, 1999b) or by a combination of these factors. Many
true fungi (Clarkson & Charnley, 1996) and also Oomycetes of the genera
Phytophthora and Lagenidium produce appressoria on the surface on the host
from which a narrow hyphae with high turgor penetrates the host. Appressorium formation may be triggered by nutritional and topographic signals, signalling the presence of stomata (Bircher & Hohl 1997; St Leger et
al.1989; St Leger et al.1991a, St Leger et al.1992). The water mould A. astaci which do not grow on the outside of its host before penetration, instead
produces a narrow penetration-peg directly from the cyst into the crayfish cuticle (Nylén & Unestam 1975).
Lytic enzymes are often considered an important virulence factor in both animal and plant pathogenic fungi, essential both for penetrating the host and for releasing nutrients (Unestam 1966, Bidochka & Khachatounas 1990; Gupta et al. 1994; Hube et al.1994). Besides having the genes for the enzymes it is important that the genes are activated at the right stage of infection. In saprophytes and opportunists enzyme production is often inducible and catabolite regulated, while developmental regulation has been seen in some obligate parasites (Heiler et al. 1993; Rauscher et al.1995). The enzymatic activity is often adapted to the host niche (Bidochka et al.1999) and insect and arthropod pathogenic fungi produce for example proteases, chitinases and lipases (Samuels et al.1989) while plant pathogenic fungi produce other enzymes such as cutinases, pectinases, cellulases and xylanases (St Leger et al. 1997). Proteases are often considered to be most important for pathogenesis in insects and crustaceans. Not only is protein the main component of the cuticle, but proteases may also serve to detoxify antifungal compounds, impairing the host immune reactions and exhibit toxic effects (Vilcinscas & Götz 1999a) The proteases secreted by different fungal parasites may have different specificity's reflecting the host niche (St Leger et
al.1997). Different specificity of proteases has also been shown between
species of Aphanomyces as indicated by different sensitivity to inhibitors from crayfish blood (Diéguez-Uribeondo & Cerenius 1998) and one putative proteinase, AaSP2, is induced by crayfish serum suggesting a role for pathogenesis (Bangyeekhun et al. 2001). It is difficult to get evidence for the involvement of proteases in pathogenesis, since activities are often redundant. For example, inhibiting or knocking out trypsins may result in upregulation of metalloproteinases. Despite this there are cases where knockout strategies have been successful. For example knockout of genes for secreted aspartyl proteases, SAPS, have been shown to result in attenuated virulence in
Together with protein, chitin is a major component of insect and crustacean cuticle. Not surprisingly chitinases have also been associated with virulence in some cases, e.g. Metarhizium anisophliae (St Leger, et al. 1991b; Beauveria
bassiana (Havukkala et al. 1993) and Nomuraea rileyi (El-Sayed et al 1989).
Chitinases belong to different classes. The most important functional distinction is between endo- and exo-chitinases. However many chitinases exhibit both endo- and exo-chitinase activity. The insect pathogenic fungi that have been studied express their major chitinases in response to catabolites such as NAG released by the action of a constitutive chitinase expressed at a low level (Escott et. al. 1998; Schickler et al 1998; St Leger et al. 1996a; St Leger et al 1996b). In contrast to this the crayfish pathogen A. astaci constitutively expresses high levels of chitinase (Unestam 1966, Paper II). As with proteases evidence for their role in parasitism is hard to get due to a multiplicity of isoenzymes (St Leger et al.1993; Shickler et al 1998). It has been reported that overexpression of chitinase result in increased antifungal activity of the mycoparasite Trichoderma harzianum (Limon et al. 1999) but there is currently no reports of increased virulence of animal pathogens resulting from overexpression of chitinase.
Host response to fungi
For a successful infection to be established the parasites must also have the ability to cope with the hosts defence response. Invertebrates, including insects and crustaceans, depend solely on innate immunity (Ezekowitz & Hoffmann 1998; Söderhäll & Cerenius (1998), that is immunity based on germline encoded receptors recognising conserved molecules of microbial pathogens, such as β-1-3 glucans, which are important part of fungal and Oomycete cell walls, (Söderhäll & Unestam 1979), and bacterial peptidoglucans and LPS (Lee et al. 2000). Recognition activates defence mechanisms such as melanisation, phagocytosis, encapsulation, clotting and production of antimicrobial peptides (Roitt et al. 1998). There are also indications that fungi may be recognised by their enzymatic activities (Vilcinskas & Götz 1999c). Proteases released by A. astaci have the potential to activate the crayfish ProPO (Söderhäll 1978) but the biological significance is not clear. In Galleria mellonella the metalloproteinase thermolysin may induce a humoral response (Griesch, & Vilcinskas 1998; St Leger et al. 1991a) and overproduction of proteases result in heavy melanisation preventing further dispersal of the fungus (St Leger et al 1996c; Wedde et al. 1998).
There is also a passive defence based on fungistatic compounds and inhibitors in the blood and cuticle of the host. Preformed protease inhibitors may also act to inhibit penetration of the cuticle in both arthropods (Häll & Söderhäll 1983; Diéguez-Uribeondo & Cerenius 1998) and insects (Vilcinscas & Götz, 1999c). For example an inhibitor active against A. astaci proteases has been isolated from the cuticle of crayfish (Häll & Söderhäll 1983). Inhibitory effect of cuticular compounds has also been shown in some insects (Smith & Grula 1982).
Fungi infecting plants will encounter analogous defence systems including both a systemic acquired response, local hypersensitivity responses (Dangl & Holub 1997) and preformed fungistatic compounds. There are several examples of the gene for gene resistance where parasite avirulence genes are recognised by plant resistance genes. Pathogens have evolved several mechanisms to avoid the host immune system. True fungi often produce secondary metabolites, for example destruxin from M. anisophliae (Vilcinscas & Götz, 1999a; Clarkson & Charnley 1996) that have toxic effects and that may act to suppress the immune system, but very little is known about secondary metabolites from Oomycetes. There is also evidence from insect pathogenic fungi that proteases may serve as toxins that impair phagocytotic activity and attachment of haemocytes (Vilcinscas & Götz, 1999a; Griesch & Vilcinscas 1998). One way of not being recognised is to avoid expressing glucan-residues on the cell-surface. Protoplast growth inside insect is common among insect pathogenic fungi (Vilcinscas & Götz, 1999d) and in some cases the fungus even uses a coat that mimics the host (Vilcinscas & Götz, 1999d). It is currently not known whether these kinds of mechanisms take place also in Oomycetes. The activation of the host immune system by extracellular enzymes may be limited by restricting the release of these enzymes (Vilcinscas & Götz, 1999c) for example by keeping them associated with the cell wall.
Molecular mechanisms of parasitic adaptations.
In the last few years, molecular tools have been developed that can be applied also to Oomycetes. Using RAPD PCR and other techniques it is possible to study the genetic diversity of Oomycete pathogens and thus reveal population structures (Huang et al. 1994; Goodwin 1997), and recombination pathways (Judelson and Yang 1998). Techniques such as differential screening or differential display RT-PCR (Liang & Pardee 1992) can be used to find genes expressed at certain time points also in organisms where genetic approaches or screening for mutants are not possible. Using differential screening Pieterse et al. (1993), cloned several infection-associated genes from P.
infestans, including ipiO, which besides being induced during infectious
growth is expressed in the zoospores (van West et al 1998). Until recently few genes had been cloned from Oomycetes and only during the last few years have substantial genome- and EST-sequencing projects been launched (Tyler 2001). In the last decade transformation protocols have been optimised for Oomycetes that utilise Oomycete promotors (Judelson et al. 1992) and successful silencing of gene expression in Phytophthora have been achieved using antisense RNA (Judelson et al. 1993; van West et al.1999). With the enormous amount of sequence data accumulating genes of potential interest in parasitism can also be cloned by PCR, using information from other organisms. In my work I have tried to apply some of the recent advances in molecular biology for study the biology of the pathogenic water molds, S.
parasitica and A. astaci. The focus has been on studying the regulation of
genes potentially important for the infection process and evaluation of S.
Results and discussion
Identification of stage specific transcripts (I).
Given the importance of zoospores for the propagation of Saprolegniales it is quite possible that many genes, important for infection, are expressed in the zoospores. However, very little is known on the gene expression in Oomycete zoospores and it has not even been clear whether the transcripts found in the zoospores encode proteins that were synthesised when the spores were formed in the sporangium, proteins that are synthesised in the spore or proteins that will be made at the onset of germination. In a pilot study we have compared gene expression during the asexual life cycle of the water mould Saprolegnia
parasitica, strain Spt using differential display reverse-transcription PCR
(ddRT-PCR). The goals of this project were to obtain markers for the zoospore and mycelium life stage and to evaluate the use of ddRT-PCR for studying the biology of pathogenic water moulds. The organism in this study is an opportunistic parasite with some capacity for killing freshwater crayfish (Diéguez-Uribeondo et al. 1994b). It exhibits repeated zoospore emergence and could be a good model for studying this phenomenon.
Previous studies using in vitro translation and protein extraction (Gwynne & Brandhorst 1982; Kramer et al. 1997) has indicated that large changes in gene expression take place at the transition between the zoospore and mycelium life stages. Sufficient quantities of RNA can be obtained from zoospores and cysts of S. parasitica with relative ease and thus the species was a suitable model system for a pilot study with differential display.
Transcripts specific for zoospore and mycelium life-stages were isolated by ddRT-PCR. With five primer combinations, a total of 220 individual bands could be seen on the differential display gels, some examples of which are shown in fig. 2. Approximately 45 of these bands were specific for the spore stage and 40 were specific for the mycelium stage. Of the five subclones tested on northern blots, four were confirmed as stage specific, and a fifth was enriched in zoospores. This indicates that a large fraction of the bands represents differentially expressed transcripts and if this holds true, as much as 10 to 20% of all transcripts could be differentially expressed between
Fig. 2. Details of ddRT-PCR bands showing differentially expressed bands in spores, cysts induced to germinate (15 min after induction), germlings (90 min after induction) and
3-day-old mycelium. Total RNA was isolated from the different stages and 35S labelled R T
-PCR was performed with different arbitrary decanucleotide primers and an anchored poly-A primer. The PCR products were separated on a polyacrylamide gel and after exposure of film the differentially expressed bands were cut out from the gel, reamplified with the original primers, subcloned and sequenced. Numbers 1 and 2 show bands
corresponding to the transcripts puf1 and sst2 respectively.
spores and mycelium. A figure of this magnitude is reasonable, considering that at least 1000 of the 6200 protein-encoding genes in the budding yeast genome is reported to be either up or down regulated during sporulation (Chu
et al. 1998). Expression of two spore specific transcripts, puf1 and sst2 was
also analysed in sporulating mycelium. Both transcripts were detected when sporangia were abundant, but before any primary spores were released. This could indicate that they are expressed already in the non-flagellated primordial spores inside the sporangia. Only a limited number of bands were re-amplified and cloned and most of them could not be identified from the short nucleotide sequences. This drawback of the technique made it unsuitable for analysis of what functional classes of genes are expressed in different life stages, or for identifying possible virulence-associated genes, due to the time consuming cloning steps required to obtain longer sequences. The chances of identifying homologues of differentially expressed transcripts will however increase in the future with the expected increase in sequence information for genomic DNA and expressed sequence tags becoming available for the Oomycetes Phytophthora sojae and P. infestans (Tyler 2001).
Nevertheless the technique proved useful for obtaining stage specific markers which were subsequently used for studying the different phases of the asexual spore cycle. As a part of the project to identify transcriptional markers, we also studied the expression of two housekeeping genes, actin and elongation factor 1, during the asexual life cycle. Both transcripts were found by northern blot to be down regulated by at least one order of magnitude, but could still be detected, in encysted zoospores.
Fig. 3: Roles of Puf proteins in other organisms. Puf proteins in many organisms regulate development by repressing translation of fate determinants. (a) The Puf Protein binds to a specific sequence in the 3´UTR of the transcript, the sequence of which varies between
different organisms and examplified by D. melanogaster Nanos responsive element (NRE)
present in hunchback mRNA. (b) Pumilio and FBF requires the recruitment of other
proteins for translational repression. For example, in the D. melanogaster embryo,
Pumilio binds to the 3' UTR of the mRNA for the transcription factor hunchback, required for the formation of anterior structures. Repression of hunchback translation requires the presence of another protein, Nanos, expressed in the posterior region of the embryo. (c)The result is translational repression of hunchback in the posterior region (Wharton
et al. 1998), and thus suppressed formation of anterior structures. (d) In the germline of C. elegans hermaphrodites, FBF binds to a similar sequence element, named PME ( p o i n t
mutation element), in the 3' UTR of f e m - 3 and by repressing translation of this
transcript, FBF promotes the switch from sperm to oocyte production in the ageing
individuals (Zhang et al. 1997).(e) In the slime mold Dictyostelium discoideum the
protein PufA is expressed in the amoeboids where it represses translation of protein kinase C (PkaC). Upon starvation PufA is degraded allowing translation of PkaC, ultimately
A putative RNA binding protein (I).
One of the spore specific transcripts, puf1 encodes a putative protein belonging to the Puf-family. The members of this family are characterised by a large domain with sequence specific RNA binding activity (Zamore et al. 1999). Puf proteins recognise and bind to a motif, named NRE or PME, in the untranslated region of the target mRNA (fig. 3). After an mRNA has been targeted by a Puf protein, it may be subjected to post transcriptional regulation mechanisms, including inhibited translation, altered poly-A tail length and increased mRNA turnover (Zhang et al. 1997; Zamore et al. 1997; Olivas & Parker 2000; Tadauchi et al. 2001).
Several Puf-proteins from various organisms play roles in differentiation (fig. 3). For example in the Drosophila. melanogaster embryo, Pumilio is involved in translational repression of a transcription factor that promotes development of anterior structures (Hülskamp et. al. 1990), and PufA from
Dictyostelium discoideum is an effector of the YakA kinase pathway which
regulates multicellular development in response to starvation (Souza et al. 1999).
Alignments with Puf-proteins from other organisms showed that the predicted Puf1 protein contains a conserved putative RNA-binding domain (fig. 4) consisting of 8 repeats flanked by two short sequence motifs (Zhang et al., 1997; Zamore et al., 1997), named Csp1 and Csp2, by Zhang et al (1997). The three-dimensional structure of Pumilio has recently been resolved (Edwards et al. 2001). The molecule forms an extended rainbow shaped structure consisting of tandem repeats, each of which consists of three α-helices. The third helix of each repeat, roughly corresponding to positions 1 to 10 in fig. 4 was suggested to interact with RNA (Edwards et al. 2001). This highly conserved region contains the highest concentration of positively charged amino acid residues and several point mutations in this region have been shown to abolish RNA binding activity. The alignment in fig. 4 shows that sequence conservation between S. parasitica Puf1 and other Puf-proteins is very high within the region of each repeat predicted to be involved in RNA-binding, and all the positions where point mutations were shown to abolish RNA binding in Pumilio (Edwards et al. 2001), are conserved between this protein and Puf1 (fig. 4). Repressing translation of the targeted mRNAs may require interaction with other protein factors (Forbes & Lehmann. 1998; Kraemer et al. 1999). On such factor is Nanos, which is required for the activity of Pumilio in the D. melanogaster embryo (Forbes & Lehmann. 1998). Several homologues of Nanos are also involved in the action of Puf proteins in Caenorrhabditis elegans.
KH I I S R VQC V T DQNGNHV I QKC I E K IP T H L AI E S IE L D K Q IL L I G KL S G H I L S L T L Q MY G CRV I Q K EV DI K G HI A E F S K D QVG S R I I QQK I E N A KM K LELS SQ LLE D F R SV LL ND FR N ---T NKHKK W E L L SR LL ED FR ---N Q P YP NL QL R SR LL ED FR -NNR Y P NL QL R SS ML EE FK ---S N K T R G F E L A SR VL YL FH -ANKQR H F E L S RR GP ED -- ---P NGQ T P K T L Q SP LL EQ LR N SSS ----DKN S N SN M S L K SF GL SS ST ---SSS M V E I S A L P AE K Q R K I E E S SR ---F ADAV L DQY I G NNV L P T W S LD S N G ---EM R S RL S L SE V s. l e . . . l DAR GH M V E F AKDQHG S R F I QQK L E T AKAD -I KD I V F A DL A N H I VE FS QDQHG S R F I QQK L E R AT AA -E KQ M V F S EI A G H I M E FS QDQHG S R F I Q L K L E R AT P A -E R QL VF N EI A G H V V E FSS D Q YG S R F I QQK L E T AT S D -E K N M V Y E DI L G N V V L FST D Q HG S R F I QQK L A T A TEE -E R EA V F Q DI KNNV I E F AKDQHG S R F I QQK L E R AS L R --D K A A I F T DI F G H S L E FC KDQHG S R F I QR E L A T SP A S -E K EV I F N LR D L D Y I K LA TD Q F G C R F LQ K K L E T PSE S N M V R D L M Y E SI H SL C KDQHGC R F L Q KQ L D I LG S K --A A D R I F E LD SG DL MK F AVDK T G C Q F L E KAVKG S LTS -Y Q K F Q LF E Dq HG s r F i q EI --YP VA L T L M T DV F GNYV I Q K F F -D F GT S H HL S L L L R EI --LA AAY S L M T D V F GNYV I Q K F F -E F GT P E QKN T L G M EI -L QAAYQ L M V DV F GNYV I Q K F F -E F GS L E Q K LA LA E EI --MP HA L A L M T D V F GNYV I Q K F F -E H GL P P Q R R E L A D EI -A ST SC LQ LM MD IF GNYVVQKY F -E F GN E K QKQ I L L S -- ----PV --LE NA E E L M T DV F GNYV I Q K F F -E F GNN E Q R N Q L VG EI --RD DA I E L S NDV F GNYV I Q K F F -E F GS K I QKN T L VD QI -K P F F L D L I LD P F GNY L VQK L C -D Y LT A E Q K TLL I Q --KDY T V E L M T D S F GNY L I QK L L -E E VT T E Q R I V L T K VI G R KDD F L K L S T N I F GNY L V Q S V I G I S L A T N D D GY T K R Q EK LK N dv F GNYv i Q k TI S G R V L E L A L Q MG C R V I QKA LEL KN M P E K LH L I S QVKGHV L Q LA LQ M Y G C RV I Q KA LES I SPE Q Q Q E I V H RI R G H V L S L A L Q MY G C R V IQ K A LEF IP S DQQN E M VR KL F D N V L P L S L Q M Y GC R V I Q KA I E V VD L DQK I K M V K QI K G H V F S L S L Q M Y GC R VVQKA I E Y I SPE HQVQ L I Q TI R G N V M K L A LQ M Y G C R V IQ K A L E Y VE E KYQH E I L G QF K G N M K Q L S L Q M Y AC R V I Q KA L E Y ID S NQR I E L V L TI Y P N V F Q I S I N Q Y GT R S L Q K I I D T VDN E V Q I D L I I IS S P H F V E IS L N P H GT R A L Q KL I E C IK T D E E A Q IV V FI S S Q M T D M C LD K F A C R V I Q S S LQ N MD L S L A C K L V Q qm y g c R v i Q k -ELT GHV -L K C V KDQNGNHVVQKC I E I LP W K T P V A M D V G G F I L S -E L DGHV -L K C V KDQNGNHVVQKC I E C VD P VA ---L Q F I I N -E L DGHV -L K C V KDQNGNHVVQKC I E C VQ PQ S ---L Q F I I D -E L DGHV -M R C V R D QNGNHVVQKC I E C VP EE N ---I E F I I S -E L DGHV -L D C V C D QNGNHV I QKA I E C ID T GH ---L Q F I L R - --- ---E M E GQV -L K C V KDQNGNHV I QKV I E R VE PE R ---L Q F I I D -ELS D S V - ---L Q M I K D Q N G NHV I QKA I E T IP IE K ---L P F I L S KG F S Q E F T S I E Q VV T L I N D L NGNHV I QKC I F K FSPS K ---F G F I I D DS L R P Y T - ---V Q L S K D L NGNHV I QKC L Q R LK PE N ---F Q F I F D -A LP RD AR -L I A I C V DQNANHV I QKVVAV IP L K N W E F I V D Dq Ng NHV i Q K AF LG ----N V Y S L A T H P Y G C R V I Q R V L E H CT E AQ ---M A P I L K ---- ---AF KG ----Q V Y S L S T H P Y G C R V I Q R I L E H CT AE Q ---T T P I L D ---- ---AF KG ----Q V F A L S T H P Y G C R V I Q R I L E H CL P DQ ---T L P I L E ---- ---TF FG ----N V V T L S T H P Y G C R V I Q R V L E H CH DP D ----T Q S K V M D - - ---AL RP ----Q I H V L S A H P Y G C R V I Q R A I E H CH SE R -K L I I E ---- - --- ---- ---AF T K N N S D N V YT L S V H P YGC R V I Q R V L E Y CN EE Q ---K Q P V L D - - ---SL TG ----H I Y H L S T H S Y G C R V I Q R L L E F GS S E D ---Q E S I L N - - ---AI V E Q N --N I I T I S T H K H G C C V L Q K L L S V CT L QQ ---I F K I S V ---- ---AI SD ----S C I D I A T H R H G C C V L Q R C L D H GT TE Q ---C D N L C D - - ---F V A T P -E H L R Q I C S D KYGC R VVQ T I I E K LT AD S M N V D L T S AAQN L R E R A L Q R L M T hp y G C r V i Q r EI HDC C C L L V E DQYGNYV I QHV L E H GQ PSE RS ---Q V I N -EL HE HT EQ LI QDQYGNYV I QHV L E H GKQ ED KS ---I L I N --EL HQH T E Q L VQDQYGNYV I QHV L E H GR PE DK S ---K I V A -EI MS TI SM L AQDQYGNYV I QHV L E H GK PD ER T ---V I I K -EL LP HI LK LT QDQYGNYVVQH I L R T GS ES DKK ---Y I F D - - --AL QI HL KQ L V L DQYGNYV I QHV I E H G SPS DK E ---Q I V QD-EL KD F I P Y L I QDQYGNYV I QYV L Q Q DQ F T NK E M V D I K Q E I I E -KI V Q F L P G LI NDQ F G NY I I Q F L L D I KE LD FY LL A ----E L F N -KL LA L VDK L T L D P F G NYVVQY I I T K EA E KNKY D Y T -H K I VH L L SV T N R C Q E LA TN EY A N Y I I Q H I V S N DD L AVYR ---E C I I E K -dqygN Y v i Q h KVY P D I V R F S Y HK F A SN V I E K C L M Y A SV HQ LH VI V AHV ME A --- ---SV R G K V L V L S Q H K F AS NVV E K C V TH A TR GE RT GL ID EV CT F --- ---EI R G N V L V L S Q H K F AS NVV E K C V TH A SR TE RA VL ID EV CT M --- ---EL A G K I V Q M S QQK F A S NVV E K C L T F G GP EE RE FL VN EM LG TT D- ---LM I D H L L F L S CH K F A S NVV E R C I S Y I SD VD RR RI LN KI IS EK A- -- ---VI S D D L L K F AQHK F A S NV I E KC L T F G GHA E R N L I I DKVC GD - ---TV A N N V V E YS K H KF AS NVV E K S I LY G SK N Q K D L I I S K I LP R -D K NHA L N L RL S N E L C Q L S CL K F SS NVV E K F I KK L FR II TG FI VNNNGGA -S Q R T A V A S KP R -A I E L S I H KF GS N V I E KI L K TA IV SE PM IL EI LN NG G --- ---CL M R N L L S L S QE KF AS H V V E K AF L H A PL EL LA EM MD EI FD GYI P ---hK F a Sn V v E k c -- ----N ER GE CP LQ VM MK DQYANYVVQK L I DV ADA E E -R E R M VV -- -NDN -A L HV M M KDQYANYVVQK M I DV SE PT Q -L K K L -M T -- -NDG P H S A L Y T M M K DQYANYVVQK M I DV AE P GQ -R K I V M H -- ---EN E ----P L Q A M M K D Q F A N YVVQKV L E T C DDQQ -R E L I L G ---E N C S -I L M L MM K D KYANYV I Q K L L D A SP EEE -R D LL I S -- ---PN DP SP -P LL QM MK DP FA NYVVQK M L DV AD P QH -R KK I T L ---E D D S -P M I L MI K D Q F ANYV I Q K L VNV SE G E G -K K L I V I IN A SM N I L LTT I D I F T V N L N V L I RD N F G N YA L Q T L L DV KNY S P L L A Y NKN S N - -- ---ET G I -Q S L L N D S Y G N Y V L Q T A L D I S KQN D Y L Y KR L S E I -- ---HP DT G KDA LD IM MF HQ F G NYVVQC M L T I CC DAV S G -R R Q T K E dqya N Y vvQ k - ----I I K T Q A S H L K R - ---F N F G -K H I L N R ---L E K L T - ----K I R P H M A A L R K ----Y T Y G -K H I N A K ---L E K Y Y - ----K I R P H I A T L R K ----Y T Y G -K H I L A K ---L E K Y Y - ----R I K V H L N A L K K ----Y T Y G -K H I V A R ---V E K L V - ----Y I Y P H I S V L K K - ---F T Y G -K H L I M S ---V E R F R - - - ---L E Q P S - ----T I K P H I A T L R K ----Y N F G -K H I L R K Y I L K --L E K Y F - ----A I R A Y L D K L N K S ---N S L G N R H L A S ---V E K L A AI GQN S S S T L N Y G N F C N DF S L K I G N L I V L T -KE -L L P S I - -V A P L L V G P I R N ---T P H G -K R I I G - -- ---M L H L D -GGYD H A I S F Q D W L K K L H S R V T K E R H R L S R F S S GKK M I E T L A i. h . . l . k . t . g k h i. . . L E K Y 3 5 4 2 10 9 8 1 7 6 3 5 4 2 10 9 8 1 7 6 3 5 4 2 10 9 8 1 7 6 3 5 4 2 10 9 8 1 7 6 3 5 4 2 10 9 8 1 7 6 3 5 4 2 10 9 8 1 7 6 3 5 4 2 10 9 8 1 7 6 3 5 4 2 10 9 8 1 7 6 CSP1 Repeat 1 Repeat 2 Repeat 3 Repeat 4 Repeat 5 Repeat 6 Repeat 7 Repeat 8 Csp2 - - - - - - - -DD V - -Sp Pum Dm Pum Hs Pum At F16 S zp Pum Ce Pum S
cYLL ScUTH ScYGL CeFBF
EL N G H I -IQ F I ID SF H G H I Y Q LA TH P Y G CRV I Q R I L E H CA E K Q V A P I L D EL M RCA VS L VQDQYGNYV I QHV L E N GT P R DK S AI V C KL QGQ I Y N L S Q HK F A S N V I E K C V QHG CT A E R IL I I N E IL G D AN SPN SS N V L L K I L K D P YANYV IQ K I L D I VE P A Q -RD M I I N RI Q P F V P TL K K VT P G D d PufA NL SL E EK Q L V F D VVAVH SL MT DV F GNYV L Q K F F EH G T TE Q KR I L AD - -Sp Pum Dm Pum Hs Pum At F16 S zp Pum Ce Pum S
cYLL ScUTH ScYGL CeFBF
D d PufA Sp Pum Dm Pum Hs Pum At F16 S zp Pum Ce Pum S
cYLL ScUTH ScYGL CeFBF
D d PufA Sp Pum Dm Pum Hs Pum At F16 S zp Pum Ce Pum S
cYLL ScUTH ScYGL CeFBF
D d PufA E T - -Sp Pum Dm Pum Hs Pum At F16 S zp Pum Ce Pum S
cYLL ScUTH ScYGL CeFBF
D d PufA B BN BN N R R R R R
Point mutation analysis suggested that Pumilio interacts with Nanos by residues in the eighth repeat (Edwards et al. 2001) but in contrast to the RNA-binding motifs, the residues predicted to be required for recruitment of Nanos and another protein, Brat, are not conserved between different Puf-proteins. This region is also poorly conserved between Puf1 homologes from the different Aphanomyces spp. (paper II) and it is thus questionable whether the region is involved in protein-interactions in these proteins.
Many organisms harbour several Puf-homologues, which are more or less related and forming several subclasses (Zamore et al. 1997; Zhang et al. 1997). Puf1 shows homology to one subclass, which is present, with one or more representatives, in organisms from all eukaryotic kingdoms. Other members of this subclass, which have been characterised are Pumilio from D.
melanogaster and PufA from D. discoideum. The members of this subclass
are characterised by having similar Csp's and by a very high degree of conservation within the RNA binding repeats (Fig. 4). They also show higher similarity other members of the same subclass in other kingdoms than to other Puf-homologues in the same species, suggesting that it has evolved early in the evolution of eucaryotes. In contrast FBF, from C. elegans, and UTH4 from S. cerevisiae are more distant. Nevertheless they both exhibit
Fig. 4: Conservation of RNA binding between Puf-proteins from different kingdoms. Continous alignment of the RNA binding domain, arranged in blocks corresponding to the
repeats according to the 3-dimensional structure model (Edwards et al. 2001). "R" shows
positions where point mutation analysis destroys RNA binding activity in Pumilio. "N" and "B" shows positions were point mutations or insertions prevents interactions with the c o -factors Nanos and Brat respectively. The first six sequences may represent a subclass of the Puf-family that is conserved between the eukaryotic kingdoms.
These proteins are highly similar and almost identical within the predicted RNA binding region, indicated by pos 1 to 10 in the alignment, and most of them contain similar CSPs. Most organisms also harbor more distantly related Puf-homologues, represented in the alignment by YGL023, UTH4 and FBF, lacking the conserved Csp's and with significant differences in the RNA binding region.
The sequences are: SpPum = Puf1, Saprolegnia parasitica, accession number A J 2 4 5 4 4 1 ,
DmPum = Pumilio, Drosophila melanogaster, accession number A46221, HsPum =
KIA0099, Homo sapiens, accession number D43951, AtPum = F16P2.43, Arabidopsis
thaliana, accession number ACC95216, SzpPum = SPAC1687.22, Schizosaccharomyces pombe, accession number CAA22616, CePum = W06B11.2 Caenorrhabditis elegans,
accession number U39854, DdPufA = PufA, Dictyostelium discoideum , accession number
AF128626, ScYLL = YLL013c, Saccharomyces cerevisiae, accession number Z 7 3 1 1 8 ,
ScUTH = UTH4= MPT5, S. cerevisiae , accession number D26184, ScYGL = YGL023, S. cerevisiae, accession number S57889, CeFBF = FBF2, C. elegans, accession number
U23176. Boxes indicate amino acids conserved in more than 50% of the sequences. Dark boxes indicate identities, light boxes indicate similarities.
RNA binding activity, and act in post-transcriptional regulation (Zhang et al. 1997; Tadauchi et al. 2001). Taken together, the results of the sequence analysis support the assumption that S. parasitica Puf1 binds mRNA but it does not give any information about possible interaction sites with other proteins or the identity of targeted mRNAs.
puf1 is specific for undifferentiated cysts (I).
While it has previously been shown that germination of Oomycete zoospores is dependent on de novo synthesis of mRNA (MacLeod & Horgen 1979; Penington et al.1989; Söderhäll & Cerenius. 1983), the question of whether transcription occurs also in non-germinating spores has never been thoroughly assessed. To test whether changes in gene expression occurs also in non-germinating cysts we used RT-PCR to study in detail the expression of three spore-specific transcripts. Expression was analysed in zoospores and in cysts collected at different timepoints from encystment to zoospore emergence. One of the transcripts, puf1, was detected only in the cysts, while two other transcripts, sst2 and sst15 were present throughout the spore-stage.
Cysts of this Saprolegnia strain may be triggered for germination by the addition of a nutrient trigger during the first hour after encystment after which they become determined for releasing new zoospores. We found that the transcript puf1, which was absent in zoospores, was produced by de novo synthesis almost instantly following the trigger to encyst, and maintained until the cyst was either triggered to germinate or became determined for releasing a new zoospore. The results show that the cyst stage has two phases, of about equal duration, which are distinct with respect to physiology and
puf1 expression. Another conclusion is that the transcriptional machinery is
active in cysts also prior to the onset of germination.
puf1 in species with other regulation of germination (II)
While cysts of S. parasitica, can be triggered to germinate by addition of nutrients over an extended period of time, those of many Aphanomyces spp. including A. astaci only germinate in response to a trigger presented before or at the time of encystment. Spores of many saprophytic species, such as
Aphanomyces laevis, instead germinate following encystment, without the
need for additional triggers. We wanted to know whether puf1 expression is a general feature for cysts of different genera and whether the differences in regulation of spore development between species are associated with differences in the puf1 expression-pattern. By using PCR-primers directed against conserved regions of the protein family, transcripts with high
similarity to puf1 were amplified from cysts of A. laevis, A. astaci and
Aphanomyces stellatus. In A. astaci, where homogenous populations of
primary and secondary cysts can readily be produced we showed by RT-PCR that puf1 is expressed in both cyst-stages. We also found that in both A. astaci and A. laevis, homologues of puf1 were induced upon zoospore encystment, and subsequently degraded upon germination and that A. astaci cysts undergoing repeated zoospore emergence maintained the transcript for over two hours despite their inability to respond to germination triggers. The expression pattern of puf1 is thus a conserved feature of spore development in
Saprolegnia and Aphanomyces, and between primary and secondary cysts but
it was only in S. parasitica, however that a puf1-expressing secondary cyst remained undetermined.
Expression of puf1 was also detected in sexually reproducing A. stellatus. The sequence of this PCR product was identical to that from asexual cysts, indicating that the same gene is active during both sexual and asexual spore-formation. However, since a weak band was detected also in vegetative mycelium it cannot be excluded that the transcript detected actually arise from asexual spore-formation, although no signs of that could be seen in either vegetative or sexually sporulating mycelium.
Chitinase in the crayfish plague fungus, A. astaci (III,IV)
Using degenerate primers designed against conserved domains of chitinases we cloned a putative chitinase from A. astaci, which was designated AaChi1. The gene is related to the Bacillus class of endochitinases, which is found in bacteria, plants and fungi and the sequence of the putative active-site agrees well with the consensus sequence from this class of chitinases.
Southern blots indicate the presence of several chitinase genes in the genome of A. astaci. The presence of many different chitinases is a common feature of fungal parasites on invertebrates (Hearn et al. 1998; St Leger et al. 1993) and may reduce the sensitivity to host chitinase inhibitors, and it is also possible that they have different specific activity against different modified forms of chitin (Bishop et al. 1999). A. astaci has a gene for chitin synthase (unpublished results) and thus AaChi1 could in theory be involved in degradation of endogenous chitin during development, but several indices support the idea that the putative chitinase AaChi1 is involved in the extracellular chitinolytic activity. The mRNA expression profile of AaChi1 correlates perfectly with the production of chitinase in the culture medium and bands of the same size and intensity are detected on northern blots with probes containing the active site and the less conserved N-terminal region and
Strain Species Country Origin Year Chitinase -/+ chitin L 1 A.astaci type 1 Sweden A. astacus1 1 9 6 2 + / + G1 A. astaci type 1 Sweden A. astacus1 1 9 6 2 + / + Hö A.astaci type 2 Sweden A. astacus2 1 9 7 4 + / +
K v - 1 A.astaci type 3 Sweden P. leniusculus2 1 9 7 8 + / +
Pc A.astaci type 4 Spain P. clarkii3 1 9 8 8 + / +
An sp. Spain P. leniusculus2 1 9 9 8 - / + A r sp. Spain P. clarkii3 2 0 0 0 - / n . a . Fa sp. I t a l y P. clarkii3 1 9 9 9 - / n . a . Fb sp. I t a l y P. clarkii3 1 9 9 9 - / n . a . Se sp. Spain P. leniusculus2 2 0 0 0 - / + 1 0 7 - 5 2 A. laevis4 - - - - / +
4 6 5 - 6 4 A. laevis4 Germany soil - - / +
5 6 8 - 6 7 A. stellatus4 - - - - / +
P 1 9 7 A. invadans7 Thailand C. striata5,7 1 9 9 5 - / n . a .8
T99G2 A. invadans7 Thailand O. goramy6,7 1 9 9 9 - / n . a .8
1Noble crayfish, Astacus astacus
2 Signal crayfish, Pacifastacus leniusculus
3Red swamp crayfish, Procambarus clarkii
4Centralbureau voor Schimmelcultures, Bern, Nl
5Striped snakehead, Channa striata
6Giant goramy, Osphronemuc goramy
7Kindly provided by Jim Lilley, University of Stirling, Scotland, UK
8Failed to grow in this medium.
Table 1: Aphanomyces strains used in this study.
thus it seems less feasible that the signal detected is due to cross hybridisation with another transcript. In addition the probe detects a transcript in the
Aphanomyces sp. isolate Se (paper IV), only when grown in a
chitin-containing medium correlating with the production of chitinase. Constitutive chitinase expression, as determined by enzymatic activity against the fluorogenic substrate 4methylumbelliferyl-β-D-N,N',N''-triacetylchitotrioside and transcription of AaChi1, was observed in strains representing all four currently known genotypes of A. astaci (Table 1), while it was neither seen in the saprophytes A. laevis and A. stellatus nor in the fish pathogen
Aphanomyces invadans. As would be expected from the results of previous
studies on chitinolytic activity (Unestam 1966; Söderhäll et al. 1978), the transcript is constitutively expressed in growing mycelium but not in zoospores. Neither the enzyme production nor the transcription of AaChi1 was affected by the presence of glucose or chitin in the growth medium. As with several fungal chitinases a down regulation of the transcript and enzyme production is seen in response to high (1%) concentrations of NAG. This pattern of chitinase expression differs from that seen in the insect pathogenic
fungi Trichoderma harzianum (Schickler et al.1998) and M. anisophliae (St Leger et al. 1996a; St Leger et al. 1996b) and in Aspergillus fumigatus (Escott et al. 1998). These fungi instead regulate the expression of chitinases by a positive feedback mechanism. NAG released by a constitutive chitinase, expressed at a low level, stimulates the synthesis of their major chitinases. Induction of chitinases by the substrate, chitin, appears to be the case also in
Aphanomyces where it has been observed in the saprophytes A. laevis
(Unestam, 1966, paper III), and A. stellatus, (paper III) and also in two non-virulent Aphanomyces isolates, An and Se (paper IV), isolated from freshwater crayfish (table. 1).
High levels of chitinase was also produced when sporulation was induced by washing the mycelium with lake-water, despite the absence of substrate. This was seen also in the mutant strain G1, which does not show any signs of sporulation, indicating that it is the starving mycelium, rather than the sporangia that produce chitinase. These results also supports the conclusion that chitinase expression is constitutive, and not induced by substances in the peptone medium.
In summary, the pattern of chitinase expression in A. astaci reflects the parasitic life cycle of this organism. Chitin is first encountered upon infection when the penetration peg has reached the endocuticle and then both chitinase activity (Söderhäll et al. 1978) and AaChi1 mRNA (paper III) are detected. The fungus then mainly persists in the chitinous endocuticle until the death of the crayfish and in the dying crayfish hyphae penetrate the cuticle from the inside and form zoospores in contact with the surrounding water. The observed expression pattern ensures that chitinase is produced at all these stages of infection. In theory constitutive enzyme expression would be a waste of resources that could make A. astaci less competetive on other substrates. However, it is hardly ever found outside its host (Unestam 1965; Scott 1961) and its specific requirements for germination could lower the risk of germinating outside the host.
Since chitinase production is common to all described genotypes of A. astaci it may also be used as a criterion for identification of this pathogen and it gave an early indication that the new Spanish and Italian isolates were different from A. astaci. Since the chitinase assay only requires the collection of a few microliters of culture medium it may be applied to first mycelium obtained after isolation from the crayfish without sacrifying the culture, and may also be used to screen single zoospore isolates for A. astaci.
New Aphanomyces strains from freshwater crayfish (Paper IV).
Five strains of Aphanomyces were isolated from the freshwater crayfish
Procambarus clarkii and P. leniusculus during suspected outbreaks of
crayfish plague in Spain and Italy (table 1). The strains were not capable of killing crayfish of the crayfish plague susceptible species A. astacus or
Cherax destructor and ITS sequence analysis showed that the strains are not
related to A. astaci and instead show highest similarity to A. laevis 107-52 and 465-64, and to an unclassified isolate from fish. Comparisons of ITS-sequences showed a high degree of similarity between the new isolates although one isolate, Se is different from the other four isolates and may even represent a different species.
We also found that A. astaci belonging to each of the four currently known genotypes were identical in the ITS1 and ITS2 regions. RAPD PCR showed that the two Italian isolates were identical and may represent the same clone, while the other three isolates could be distinguished from each other. The RAPD profiles were also clearly different from those of A. astaci reference strains, from each of the four known A. astaci genotypes.
The new strains share some physiological properties, which could not be found in the A. astaci strains. They grow much faster than A. astaci, lack constitutive chitinase and their zoospores, in contrast to those of A. astaci, germinate in response to nutrients. The five strains are capable of repeated zoospore emergence and could produce 3 generations of zoospores (paper IV), similar with previous observations of A. astaci (Cerenius & Söderhäll, 1984). It has been suggested that repeated zoospore emergence is an adaptation to parasitism (Cerenius & Söderhäll 1985) but while the hypothesis appeared to hold true for Aphanomyces it was not supported in
Saprolegnia (Diéguez-Uribeondo et al.1994a). The finding of crayfish
isolates exhibiting repeated zoospores emergence, without being able to kill crayfish indicates that the association of repeated zoospore emergence with parasitism is not general within Aphanomyces. However, the proposed advantage of repeated zoospore emergence over automatic germination, would be that it offers a mechanism of substrate selection (Cerenius & Söderhäll 1985; Deacon & Donaldson 1993; Deacon 1996) and may thus rather reflect how narrow the niche is, where the fungus is competitive, than the ability to kill the host. While repeated zoospore emergence thus may be advantageous for parasites, which often have narrow host niches, it cannot be used to classify a strain as being a parasite or a saprophyte.
Conclusions
We have used S. parasitica strain Spt as a model for studying gene expression during Oomycete zoospore differentiation. The high synchrony of differentiation enabled us to study changes between different phases of cyst differentiation and germination. The relative ease with which RNA is isolated from synchronised spores in different phases of differentiation, makes this model suitable for identifying other phase specific transcripts.
We found a transcript, puf1 that was transiently expressed upon encystment showing that the transcriptional machinery is present and active in encysted zoospores, also prior to the onset of germination and emphasises that the zoospores of Saprolegniales are not dormant in contrast to many spores from true fungi. In S. parasitica puf1is a marker of undetermined cysts that can be triggered to germinate and it disappears when the cysts are either triggered to germinate or begin producing new zoospores. Also in A. laevis where zoospores automatically germinate upon encystment and in A. astaci where zoospores can only be triggered to germinate at the time of encystment, puf1 homologues are expressed with similar patterns of expression as in S.
parasitica. Thus the pattern of puf1 expression is a conserved feature of
zoospore development taking place regardless of the germination triggers of the species.
The undetermined cysts have the capacity for both germination and repeated zoospore emergence, and are also the structures from which penetration pegs emerge. They may thus harbour several transcripts that play important roles during the early events of establishing an infection and thus it may be worthwhile to look in the future for other transcripts that are differentially expressed between the different phases of zoospore differentiation.
The putative Puf1 protein is member of a novel family of sequence specific mRNA-binding proteins, named the Puf-family. The high degree of
conservation indicates that the putative protein also binds mRNA but its biological role cannot be predicted since the different Puf-proteins that has been studied participate in completely different signal pathways. However the expression pattern of this transcript, together with the functions described for other Puf-proteins, suggests a role for Puf1 protein in blocking translation of selected transcripts in the cyst-stage, or in regulating mRNA turnover upon germination or zoospore release.
Constitutive chitinase expression is conserved between the different genotypes of A. astaci in contrast to saprophytic species and some newly isolated Spanish and Italian Aphanomyces isolates. The constitutive expression may be an adaptation to parasitism in A. astaci, which is a highly specialised parasite that does not live in nature without its host (Scott 1961; Unestam 1972). The pattern of chitinase production reflects the parasitic life cycle of this species, taking place during mycelial growth and in starving mycelium prior to sporulation, but not in zoospores, or early germlings, which are unlikely to encounter chitin.
Five isolates of Aphanomyces sp. that showed morphological similarities to A.
astaci, were isolated from freshwater crayfish in Spain and Italy. The
molecular data however showed that they were unrelated to this species, emphasising that either molecular data or reinfection-experiments are necessary for a reliable diagnosis. ITS sequencing proved to be a convenient method for identification of species within Aphanomyces, and offers a simple diagnosis for crayfish plague. However, the ITS-sequences of the four known genotypes of A. astaci are identical and it is thus necessary to perform RAPD-PCR analysis for tracing dispersal of individual clones. Since constitutive chitinase production appears to be conserved within A. astaci in contrast to other Aphanomyces spp. it may be used for getting a preliminary diagnose, and for selecting single zoospore isolates. The Spanish and Italian isolates of Aphanomyces isolates exhibits repeated zoospore emergence, apparently without being pathogens indicating that the ability to perform repeated zoospore emergence can not be used to classify a strain of
Acknowledgements.
I wish to thank
My supervisors Kenneth Söderhäll and Lage Cerenius
All former and present member of department of Comparative Physiology for helping me with with various things in the lab and for making this my PhD period a really good time.
My wife Mahsa for love and support and my doughter Sepi for reminding me that there are other things in life than work.
The work has been supported by a PhD fellowship from the Swedish Council for Forestry and Agricultural Research, by grants from Carl Tryggers
Stiftelse, from the Swedish Council for Forestry and Agricultural Research (SJFR) and from the Swedish Research Council forEnvironment, Agricultural Sciences and Spatial Planning (FORMAS).
