Control of morphogenesis in the human fungal pathogen Penicillium marneffei

Int. J. Med. Microbiol. 292, 331 ± 347 (2002) ¹ Urban & Fischer Verlag http: // www.urbanfischer.de/journals/ijmm

Control of morphogenesis in the human fungal pathogen Penicillium marneffei Alex Andrianopoulos Department of Genetics, University of Melbourne, Parkville, Victoria, 3010, Australia

Abstract Fungal pathogens are an increasing threat to human health due to the increasing population of immunocompromised individuals and the increased incidence of treatment-derived infections. Penicillium marneffei is an emerging fungal pathogen endemic to South-east Asia, where it is AIDS defining. Like many other fungal pathogens, P. marneffei is capable of alternating between a filamentous and a yeast growth form, known as dimorphic switching, in response to environmental stimuli. P. marneffei grows in the filamentous form at 25 8C and in the yeast form at 378C. During filamentous growth and in response to environmental cues, P. marneffei undergoes asexual development to form complex multicellular structures from which the infectious agents, the conidia, are produced. At 37 8C, P. marneffei undergoes the dimorphic switching program to produce the pathogenic yeast cells. These yeast cells are found intracellularly in the mononuclear phagocyte system of the host and divide by fission, in contrast to the budding mode of division exhibited by most other fungal pathogens. In addition, P. marneffei is evolutionarily distinct from most other dimorphic fungal pathogens and is the only known Penicillium species which exhibits dimorphic growth. The unique evolutionary history of P. marneffei and the rapidly increasing incidence of infection, coupled with the presence of both complex asexual development and dimorphic switching programs in one organism, makes this system a valuable one for the study of morphogenesis and pathogenicity. Recent development of molecular genetic techniques for P. marneffei, including DNA-mediated transformation, have greatly facilitated the study of these two important morphogenetic programs, asexual development and dimorphic switching, and we are beginning to uncover important determinants which control these events. Understand these programs is providing insights into the biology of P. marneffei and its pathogenic capacity. Key words: morphogenesis ± asexual development ± dimorphic switching ± Penicillium marneffei ± GTPase ± transcriptional regulation

Introduction Fungi are an extremely diverse group of eukaryotic organisms closely related to animals (Zeng et al., 2001). Despite the many beneficial applications of fungi in medicine, agriculture and industry, a number of species are important pathogens. Fungal

pathogens are a significant and increasing public health problem. A number of studies have shown that nosocomial fungal infections nearly doubled between 1980 ± 1990 (Beck-Sague and Jarvis, 1993; Fisherhoch and Hutwagner, 1995; Sternberg, 1994) and that about 40% of deaths from nosocomial

Corresponding author: Alex Andrianopoulos, Department of Genetics, University of Melbourne, Parkville, Victoria, 3010, Australia. Phone: ‡ 61 3 8344 5164, Fax: ‡ 61 3 8344 5139, E-mail: [email protected]

1438-4221/02/292/5±6-331 $ 15.00/0

332

A. Andrianopoulos

infections were due to fungi (Edmond et al., 1999; Sternberg, 1994). This trend shows no signs of abating (Henderson and Hirvela, 1996; Weinberger et al., 1997). Fungal infections are particularly devastating for immunocompromised individuals such as AIDS patients where treatment is rarely curative (Walsh and Groll, 1999). At similar risk are the increasing population of immunosuppressed individuals after cancer chemotherapy or as a consequence of therapy after transplantation surgery. Surprisingly, the most marked increases in infection rates have not been in transplant or oncology units, but in other surgical and medical areas demonstrating that fungal infections are no longer restricted to severely immunosuppressed individuals (Beck-Sague and Jarvis, 1993; Edmond et al., 1999). This problem is compounded by the lack of safe and effective antimycotic drugs, especially for systemic infections, the increase in pathogen resistance to antimycotics and the alarming emergence of new pathogens (Gleason et al., 1997; Sternberg, 1994; Walsh and Groll, 1999). Effective control of fungal disease is hindered by the lack of a basic understanding of pathogenesis ± a combination of the biology of the fungal pathogen and its capacity to cause disease. Fungal pathogens are a diverse group of organisms that can range from commensals such as Candida albicans to free living opportunists such as Penicillium marneffei and Cryptococcus neoformans and aggressive pathogens such as Histoplasma capsulatum and Coccidioides immitis (Alexopoulos et al., 1996; Kwon-Chung and Bennett, 1992). With the exception of commensals, these pathogenic fungi must have mechanisms to enter the host and then to survive in the host. Interestingly many pathogenic fungi are dimorphic, and recent studies have demonstrated that dimorphic switching is directly associated with pathogenicity (for example see (Braun and Johnson, 1997; Csank et al., 1998; Leberer et al., 1997; Lo et al., 1997; Stoldt et al., 1997; Yaar et al., 1997). P. marneffei is an important emerging fungal pathogen capable of dimorphic growth. This review describes recent studies into the molecular mechanisms which control development and morphogenesis in this opportunistic pathogen and how these relate to development and pathogenesis in other fungi. The pathology and epidemiology of P. marneffei infections is also addressed but the reader is referred to a number of excellent recent reviews which deal with these issues in greater detail (Cooper, 1998; Vossler, 2001). In addition, this review highlights recent developments in techniques for the molecular analysis of P. marneffei.

P. marneffei is a unique pathogen The Penicillium group of species, comprising the teleomorphic genera Eupenicillium and Talaromyces and the anamorphic Penicillium genus, consists of a number of important industrial and food important species (Pitt, 1979). Molecular and morphological phylogenetic analyses suggest that P. marneffei is most closely related to Talaromyces species with biverticilliate conidiophore morphology (LoBuglio et al., 1993; LoBuglio and Taylor, 1995). Although P. marneffei is one of the few medically important Penicillium species, there are a number of species in closely related genera which are also important pathogens, such as Aspergillus fumigatus. However, amongst the Penicillium group and the other closely allied genera of the Eurotiales, P. marneffei is the only know dimorphic species, exhibiting all of the filamentous characteristics of this group of organisms (filamentous growth and asexual development), in addition to being capable of switching to and growing as a yeast (Alexopoulos et al., 1996; Kwon-Chung and Bennett, 1992; Pitt, 1979). Other dimorphic pathogenic fungi such as H. capsulatum, Blastomyces dermatitidis, C. immitis are distantly related to P. marneffei but closely related to each other (Fig. 1). This makes P. marneffei an important species to study in terms of molecular medical mycology in order to address the problem of how organisms acquire pathogenicity traits. In conjunction with studies in species such as H. capsulatum and C. immitis, studies in P. marneffei will also be an important point of comparison to identify differences and similarities in the molecular strategies used by these species for their pathogenic life cycles. Lifecycle The P. marneffei life cycle can be divided into three stages: Filamentous vegetative growth at 25 8C, asexual reproduction (conidiation) at 25 8C and unicellular yeast-like vegetative growth at 37 8C (Fig. 2). At 25 8C and under suitable conditions, a uninucleate conidium (spore) germinates, growing isotropically within the first 6 hours and then polarising to produce a germ tube by 12 hours. The germ tube grows apically to produce a hypha, and cellular compartments are established behind the growing tip by septation. Unlike apical cells, sub-apical cells are capable of repolarising to produce branched cells with a new apical growth point. The compartments of actively growing cells are multinucleate indicating

P. marneffei morphogenesis and pathogenicity

333

Fig. 1. P. marneffei is a unique dimorphic fungal pathogen. The phylogenetic relationship of a number of fungal pathogens is shown with the classifications listed at the base of the branches. Pathogens which exhibit a dimorphic switching program are shown in bold. Other important monomorphic fungal pathogens of animals or plants are shown for reference.

an uncoupling of nuclear and cellular division, while most older cellular compartments are often uninucleate. The reproductive asexual development program results in the production of complex, multicellular structures (conidiophores) bearing uninucleate spores (conidia) (Fig. 3). The program proceeds at 25 8C after hyphal growth is fully established and is initiated in the presence of an air interface, such as growth on a solid medium in a Petri dish. From specialised vegetative hyphal cells known as foot cells, multinucleate aerial stalk cells are produced which grow away from the vegetative mycelium. The stalk cells are often septate and may produce secondary stalks called rama. The tips of these stalk cells differentiate to produce the sterigmata cell types (metulae and phialides) which are uninucleate, unlike actively growing hyphal cells, and produced by a budding division rather than by apical growth and septation. The first budded cell type is the metula, which subsequently buds one or more of the sporogenous cell type known as a phialide. Unlike the acropetal mode of division of all other cells produced at 25 8C, phialides produce conidia in a basipetal mode such that older spores are displaced by younger spores budded from the phialide. This produces a chain of spores which, when mature, have a pale green colouration. In addition, under conditions of carbon limitation, P. marneffei undergoes a rudimentary form of asexual development. Instead of the long conidiophore stalks and rama,

very short unbranched stalks are produced which bud sterigmata and conidia in the usual manner (Fig. 3). In some cases no stalks are produced and sterigmata appear to bud directly from foot cells. Carbon limitation also influences the temporal pattern of asexual development. Under carbonlimiting conditions such as 0.1% glucose, conidiophores appear within 2 days whereas under carbonsufficient condition (1% glucose) conidiophores are evident only after 5 days. Yeast morphogenesis in P. marneffei occurs at 37 8C producing uninucleate elongate yeast cells which divide by fission, similar to that for Schizosaccharomyces pombe (Chan and Chow, 1990; Garrison and Boyd, 1973; Pitt, 1979; Segretain, 1959) (Fig. 4). Germinated conidia produce highly branched hyphal cells at 37 8C, and morphogenesis begins 48 h after germination with the coupling of the nuclear and cell division cycles to produce hyphal compartments which contain a single nucleus. These hyphal cells are shorter (20 mm) along their long axis than vegetative hyphal cells (40 mm) grown at 25 8C and appear to be a transitory state. These prearthroconidial cells are separated by easily discernible double septa and evident by 72 h. Pre-arthroconidial cells separate as the cell wall material between the double septa is degraded to liberate uninucleate, single cells. This process is termed arthroconidiation and the resultant single cells are called arthroconidia. Arthroconidia initiate polarised growth, maintaining the elongated cell shape,

334

A. Andrianopoulos

Fig. 2. P. marneffei life cycle. Diagrammatic representation of the P. marneffei life cycle showing the various growth and developmental stages. The solid circles represent nuclei in cells. The filamentous phase exhibited at 25 8C is shown above the dashed line and characterised by hyphal growth and the asexual development program which produces conidia. The yeast phase exhibited at 37 8C is shown below the dashed line and is characterised by the arthroconidiation program, yeast growth and division by fission. The dimorphic switch which links the yeast and hyphal growth phases in response to temperature is shown.

and divide by fission after nuclear division to produce the true yeast cells. Hyphae produced at 25 8C and shifted to 37 8C undergo a similar morphogenic process beginning with the coupling of nuclear and cell division in the previously multinucleate apical cells. Arthroconidiation has

been described in a number of fungal species, including the pathogen C. immitis (Alexopoulos et al., 1996; Kwon-Chung and Bennett, 1992; Pitt, 1979), however concomitant maintenance of a yeast growth state following arthroconidiation is uncommon amongst dimorphic fungi.

P. marneffei morphogenesis and pathogenicity

Fig. 3. Asexual development in P. marneffei at 25 8C. A. A P. marneffei conidiophore showing the basal hypha (H), rama (R), metula (M), phialide (P) and conidium (C) cell types. Cultures were grown on medium containing 1% glucose as a carbon source. B. A rudimentary P. marneffei conidiophore showing the phialide (P) and conidium (C) cell types. Cultures were grown on medium containing 0.1% glucose as a carbon source. C. P. marneffei conidia. Cell walls were stained with calcofluor and photographed using fluorescence optics. Scale bars represent 20 mm.

When conidia are germinated at 37 8C, they do not undergo direct morphogenesis to produce yeast cells. Initially, germination of conidia at 25 8C and 37 8C is indistinguishable with conidia swelling isotropically and then polarising to produce a germ tube and eventually a hyphal network. However, unlike hyphae produced at 25 8C, the resultant hyphae at 37 8C are highly branched. By 48 h post germination, these branched hyphae have begun arthroconidiation and yeast cells are clearly evident by 72 h. Therefore, conidia must undergo a certain period of growth before gaining the ability to arthroconidiate. Interestingly, P. marneffei-infected macrophages only contain yeast cells, and the establishment of a hyphal network by P. marneffei conidia after they have been phagocytosed by macrophages (see Pathogenesis below) has not been described (Chan and Chow, 1990; Garrison and Boyd, 1973). Whether this reflects the lethal effect of extensive hyphal growth on macrophage integrity or suggests that under these conditions a direct conidium-toyeast cell morphogenesis program is activated remains to be determined. Although P. marneffei yeast cells grow apically like hyphae, unlike hyphae they do not branch, they exhibit coupled nuclear and cell division, and divide by fission where cytokinesis is followed by cell separation. When yeast cells are shifted from 37 8C

335

Fig. 4. Yeast morphogenesis of P. marneffei at 37 8C. A. Arthroconidiation of hyphae grown at 25 8C and transferred to 37 8C for 3 days. Septa are visible in the hyphae (black arrowheads) using differential interference contrast optics (top) Staining with DAPI shows a single nucleus (white arrowhead) in each cellular compartment (bottom). B. Yeast cells after 4 days at 37 8C shown under differential interference contrast optics (left) and after staining with DAPI (right) using fluorescence optics. Scale bars represent 20 mm.

to 25 8C apical growth continues but nuclear division becomes uncoupled. Elongated, multinucleate cells divide by septation with no concomitant cell separation and these cells can branch to produce a hyphal network (Fig. 5). Therefore, dimorphic switching in P. marneffei occurs in both the hyphal-yeast and the yeast-hyphal directions and is triggered by a simple temperature cue. While other factors such as nutritional conditions can influence the growth state of P. marneffei, these factors cannot effect the dimorphic switch in the absence of the temperature signal.

Epidemiology and pathology P. marneffei is endemic to the South-East Asian region, including Burma, Cambodia, southern China, Indonesia, Laos, Malaysia, Thailand, and Vietnam. It is the third most frequent opportunistic infection in AIDS patients and as a result of its

336

A. Andrianopoulos

Fig. 5. The yeast-to-hyphal dimorphic switch. P. marneffei yeast colonies, after growth for 7 days at 37 8C, were shifted to 25 8C and photographed after 0, 1, 2, 3, and 7 days. After 1 day the colonies become filamentous at the periphery and begin to produce a red pigment. The colonies continue filamentous growth and pigment production and eventually undergo asexual development by 7 days to produce pale green conidia.

rapidly escalating incidence now represents an `AIDS defining pathogen' in the region (see (Cooper, 1998; Vossler, 2001) and references therein). There is clear evidence of P. marneffei strain dissemination and establishment. A recent study in Taiwan demonstrated the presence of disseminated isolates from patients who had travelled to Thailand as well as distinct `endemic` isolates suggesting that P. marneffei is rapidly colonising new regions (Hseuh et al., 2000). Confirmed cases of P. marneffei infection have been found in Europe, North America, Africa, and Australia although most of these incidents have been associated with travel through South-East Asia (see (Cooper, 1998; Wong et al., 1999) and references therein). Furthermore, the prevalence is likely to have been underestimated until recently as a result of misdiagnosis due to the similarity of P. marneffei infections with those of H. capsulatum and the cross-reactivity of some of the earlier serological tests with other fungal pathogens (Cooper, 1998; Kaufman et al., 1995; Ungpakorn, 2000). Despite the clinical data, the ecology of P. marneffei remains obscure. Studies to identify P. marneffei in the environment have proved disappointing. It has been suggested that bamboo rats (Rhizomys pruinosus, R. sinensis, R. sumatrensis, and Cannomys badius) prevalent in endemic regions, may represent a reservoir for P. marneffei in the environment as the organism has been found in the organs and faeces of these rats and these rats are found in the region where P. marneffei is endemic (Chariyalertsak et al., 1996a, b). However, there is little evidence to support this hypothesis. In one case

study which examined penicilliosis in Northern Thailand, no link between the disease and exposure to bamboo rats could be established (Chariyalertsak et al., 1996a). Although P. marneffei is closely related to many saprophytic fungi which can readily be isolated from the environment, is capable of growing on defined minimal medium and a wide range of carbon and nitrogen sources, it has rarely been isolated. A survey of soil samples, including those from around the burrows of bamboo rats, only identified one P. marneffei isolate in 95 samples (Chariyalertsak et al., 1996b). Despite this, infection by P. marneffei is common in the region. The environmental niche of this organism and its population biology must be determined if we are to understand how infections are acquired. P. marneffei has been primarily associated with HIV-positive individuals. However, a number of cases of acquired infection in apparently immunocompetent individuals have been reported. In some of these cases undiagnosed, underlying immunodeficiencies were later identified (see (Cooper, 1998; Vossler, 2001) and references therein). In both of these groups, untreated P. marneffei infections are fatal. Although infections respond to treatment with antimycotics such as amphotericin B and a number of the azole drugs, they are rarely curative and continued treatment is required to avoid relapse, especially in immunocompromised individuals (Sirisanthana and Supparatpinyo, 1998; Sirisanthana et al., 1998; Supparatpinyo et al., 1992; Supparatpinyo et al., 1993, 1994). As has been proposed for many other fungal pathogens, infection is likely to occur through the

P. marneffei morphogenesis and pathogenicity

inhalation of the asexual spores produced by the filamentous form. In infected individuals, pulmonary alveolar macrophages and peripheral blood mononuclear cells contain multiple P. marneffei cells in the yeast form (Deng and Conner, 1985). Cellmediated immune responses via pulmonary alveolar macrophages appear to be the primary defence mechanism against P. marneffei. However, P. marneffei yeast cells are capable of dividing and killing macrophages in immunocompromised hosts and this strategy may provide a mode of dissemination throughout the host. Severe P. marneffei infections can affect bone marrow, liver, spleen, kidneys, lungs, lymph nodes, skin, and soft tissues; and mycosis is associated with a number of clinical manifestations (Hilmarsdottir et al., 1993; Supparatpinyo et al., 1994). In AIDS patients infected with P. marneffei, fever, weight loss, anaemia, skin lesions, hepatomegaly, adenopathy, and pulmonary infiltrates were noted (Hilmarsdottir et al., 1993; Supparatpinyo et al., 1994; Ungpakorn, 2000).

Cellular and developmental programs Fungi exhibit a diverse array of cellular and developmental programs in their life cycles. For dimorphic fungi, understanding the molecular mechanisms of these programs is central to understanding the capacity of the organism to infect and cause disease. In P. marneffei, the infectious agent is likely to be the asexual spore while the pathogenic agent is the yeast form which develops upon infection. Therefore, it is important to understand the molecular mechanisms which govern these two programs. Unlike other dimorphic pathogens, P. marneffei has a morphologically elaborate and experimentally amenable asexual development program as well as a defined dimorphic switching program. The presence of both developmental programs in P. marneffei makes it a valuable system for studying the relationship between these processes that have hyphal-yeast transitions in common. Control of asexual development Regulatory genes The P. marneffei asexual development program is morphologically similar to that of many other related species, including Aspergillus nidulans in which this program is best characterised (Adams et al., 1998). Asexual development in A. nidulans is controlled by two parallel regulatory pathways. The brlA and abaA transcriptional regulatory genes,

337

encoding a C2H2 zinc finger protein and an ATTS/ TEA protein respectively, make up the core pathway (Adams et al., 1988; Andrianopoulos and Timberlake, 1994; Clutterbuck and Timberlake, 1992; Mirabito et al., 1989). The brlA gene is activated by conidiation induction signals such as an air interface and light, and this in turn activates expression of the abaA gene (Adams et al., 1988; Clutterbuck and Timberlake, 1992; Mirabito et al., 1989). The abaA gene feedback regulates both brlA and abaA expression resulting in a feedback loop which eliminates the need for continued induction of this core pathway (Andrianopoulos and Timberlake, 1994; Mirabito et al., 1989). Loss-of-function mutation in these genes results in a complete failure to produce spores and loss of many of the cell types associated with spore production (Boylan et al., 1987; Clutterbuck and Timberlake, 1992). The stuA and medA genes represent a pathway which modifies the core developmental program. Loss-of-function mutations in these genes reduce but do not block spore production and result in aberrant conidiophore morphology (Clutterbuck and Timberlake, 1992; Miller et al., 1992). In P. marneffei, the abaA gene plays a similar role in asexual development (Borneman et al., 2000). The gene is highly homologous to the abaA gene from A. nidulans, and homologues exist in both pathogenic and nonpathogenic fungi including Saccharomyces cerevisiae, C. albicans and A. fumigatus. Expression of abaA is tightly regulated such that the transcript is not detected during vegetative growth but induced 30-fold during asexual development. Deletion of abaA has no detectable phenotypic effect during vegetative growth. During asexual development, the abaA deletion strain produces conidiophore stalks, rama and metulae. However, in place of the sporogenic phialides, aberrant cells are produced which are phialide-like in shape but metula-like in their acropetal division pattern and their capacity to branch. These aberrant cells are incapable of producing spores (Borneman et al., 2000). A. nidulans abaA mutants have identical phenotypes (Sewall et al., 1990). Although a few of the downstream targets of abaA in A. nidulans have been identified the exact cellular role of AbaA remains obscure (Mirabito et al., 1989). One clue has come from ectopic overexpression studies of abaA in P. marneffei vegetative hyphae using an inducible expression system. Ectopic expression results in aberrant apical hyphal cells which are swollen at their distal tips and contain excessive nuclear material, either due to uncontrolled nuclear division or endoreplication (Borneman et al., 2000). This data suggests that AbaA functions to control

338

A. Andrianopoulos

cell cycle events during conidiation. There is also evidence to suggest that AbaA controls the cell cycle during P. marneffei yeast growth (see below). In A. nidulans, cell cycle control during conidiation has been shown to be dependent on the brlA gene, which in turn activates abaA (Ye et al., 1999). The stuA gene is a member of the APSES group of transcription factors which possess a basic helixloop-helix (bHLH) DNA binding motif. These factors have been identified in a number of fungi including S. cerevisiae, C. albicans and Neurospora crassa (Aramayo et al., 1996; Gimeno and Fink, 1994; Miller et al., 1992; Stoldt et al., 1997; Ward et al., 1995). Like abaA, expression of the stuA gene in P. marneffei is tightly regulated (Borneman et al., 2002). The transcript is not detectable during vegetative hyphal growth but strongly induced during conidiation. Deletion of stuA results in the loss of the sterigmata cells, metulae and phialides. Despite this, spores are still produced directly from the tips of conidiophore stalks. Therefore stuA is required for the specification of these two cell types but the core conidiation pathway is still active and can convert the tip of the conidiophore into a sporogenous cell capable of budding spores basipetally (Borneman et al., 2002). As for abaA, the P. marneffei stuA mutant phenotype is similar to the corresponding A. nidulans stuA mutant. In A. nidulans, both the brlA and abaA genes are targets for StuA (Miller et al., 1992). Signalling genes A number of genes that activate the brlA/abaA pathway have been identified and characterised in A. nidulans and these include genes encoding a Ga subunit of a heterotrimeric G protein (fadA) and a regulator of G protein signalling (flbA) (Wieser et al., 1997; Yu et al., 1996). Ga subunits have been characterised in many fungi and they play key roles in the control of many morphogenic programs, such as dimorphic switching and conidiation, in addition to affecting pathogenicity in a number of species (for review see (Lengeler et al., 2000)). In P. marneffei, deletion of the Ga subunit gene gasA results in delayed conidiation under inducing conditions (Zuber et al., 2002). A similar result is observed when a dominant negative allele of gasA is introduced into P. marneffei. Under conditions which suppress conidiation in the wild type, such as growth in liquid, the strain carrying a dominant negative allele conidiates and the gasAD strain produces thickened hyphae similar to those of conidiophores. These results are consistent with Northern experiments which show brlA expression in the dominant negative and deletion strains in liquid culture.

Conversely, strains carrying a dominant active allele of gasA fail to produce conidiophores and show no brlA expression. These data suggest that gasA inhibits conidiation and is therefore involved in the decision between growth and development (Zuber et al., 2002). In A. nidulans, the fadA gene is also involved in regulating secondary metabolism (Wieser et al., 1997). P. marneffei produces a red pigment which is characteristic of the species and this pigment is likely to be a secondary metabolite. The pigment is only produced during filamentous growth at 25 8C and the amount of pigment varies depending on the growth medium. For example, pigment production is low when grown on medium containing ammonium as a sole nitrogen source but high when GABA (g-amino butyric acid) is the nitrogen source. Strains carrying the gasA dominant negative and deletion alleles produced less pigment than the wild type, which in turn produces less than strains carrying the dominant activating allele (Zuber et al., 2002). Hence, like fadA in A. nidulans, the gasA-encoded Ga subunit controls both asexual development and at least one secondary metabolic pathway. Although many Ga genes have been isolated from fungi and have been implicated in regulating developmental programs, few of these are from animal pathogens. In C. neoformans the GPA1 gene encodes a Ga subunit which is required for mating and regulates virulence (Alspaugh et al., 1997). Gpa1 like many other characterised Ga subunits has a direct effect on the levels of cAMP, and a number of downstream targets have been identified. Despite the number of Ga subunits that have been isolated from various fungi, very few of the Gb and Gg components have been identified and even fewer of the transmembrane receptors which link to these G proteins (Lengeler et al., 2000). Control of dimorphic switching Dimorphic switching is a common program in an evolutionarily diverse set of fungi and is prominent amongst the pathogenic fungi (Alexopoulos et al., 1996; Kwon-Chung and Bennett, 1992). Different signals can trigger this switch including temperature, oxygen limitation, nutrient limitation, mating, pH, serum, or combinations of these (for example see (Gow, 1995; Kulkarni and Nickerson, 1981; Lengeler et al., 2000; Orlowski, 1991; Szaniszlo et al., 1983) and references therein). In P. marneffei the dimorphic switch is triggered by temperature and this leads to cellular changes such as the coupling and uncoupling of the nuclear and cell division cycles, septation with and without cell separation,

P. marneffei morphogenesis and pathogenicity

and directional and branching growth polarity. The close association between pathogenicity and dimorphic switching has led to the current interest of many research groups, using a number of dimorphic fungi, in the mechanisms which control dimorphic switching. Regulatory genes In P. marneffei loss of the transcriptional activator abaA leads to cellular defects during yeast cell morphogenesis at 37 8C (Borneman et al., 2000). The pre-arthroconidial cells, the transition point between multinucleate hyphal cells and uninucleate yeast cells, fail to correctly couple nuclear and cell division, resulting in multinucleate cells. Furthermore, many of the nuclei in these cells are clustered and incorrectly positioned relative to the long axis of the cell which is the normal division axis. This supports the hypothesis that abaA controls coupling events during the nuclear and cellular division cycles and that some of these defects extend to nuclear positioning and movement. In vegetative hyphal cells of the wild type and abaA mutant grown at 25 8C, where abaA expression is not detectable, the nuclei are evenly distributed along the long axis of the cell. Despite these defects, cell separation proceeds and the individual arthroconidial cells are liberated. These cells grow and give rise to yeast cells by fission which are also multinucleate. With extended incubation times, the number of yeast cells with multiple nuclei decreases (Borneman et al., 2000). Therefore, although loss of abaA affects coupling of the nuclear and cell division cycles in yeast cells, another mechanism must be activated which replaces it, suggesting that the primary role of AbaA during yeast cell morphogenesis is in the pre-arthroconidial and arthroconidial cell types. This is in addition to its role in asexual development where it is required for differentiation of the yeast-like phialide cell, the switch from acropetal to basipetal division and conidium production (see above). The S. cerevisiae and C. albicans ATTS/TEA genes, both known as TEC1, are required for pseudohyphal development and filamentation, respectively (Gavrias et al., 1996; Madhani and Fink, 1997; Schweizer et al., 2000). Although these transcription factors share little homology outside of the DNA-binding region when compared to highly divergent species, they are functionally homologous. Forced expression of the A. nidulans abaA gene in S. cerevisiae drives the pseudohyphal growth program (Gavrias et al., 1996), and there are a number of putative AbaA/ TEC1-binding sites in the promoters of important pseudohyphal genes such as PHD1, FLO11 and

339

TEC1 itself. Therefore, it is clear that ATTS proteins play important roles in fungal morphogenesis and in the transitions between hyphal and yeast growth forms. The studies in P. marneffei indicate that this may involve aspects of cell cycle regulation. In S. cerevisiae and C. albicans, two other genes play central roles in the yeast-hyphal switch. These are the APSES proteins-coding genes and the STE12like genes (Gavrias et al., 1996; Gimeno and Fink, 1994; Liu et al., 1994; Lo et al., 1997; Madhani and Fink, 1997; Stoldt et al., 1997; Ward et al., 1995). APSES genes have been identified in a number of fungi, both dimorphic and monomorphic, and encode transcription factors with a basic helixloop-helix (bHLH) DNA-binding domain (Aramayo et al., 1996; Gimeno and Fink, 1994; Miller et al., 1992; Stoldt et al., 1997; Ward et al., 1995). Deletion of the APSES-encoding stuA gene from P. marneffei has no effect on the dimorphic switch (Borneman et al., 2002). Morphologically normal yeast cells are produced at 37 8C and these yeast cells revert to the hyphal form when the temperature is lowered to 25 8C. In contrast, P. marneffei stuA mutants show defects in asexual development at 25 8C, failing to produce the yeast-like sterigmata cells (metulae and phialides) and this asexual development phenotype is similar to that in A. nidulans stuA mutants (see above). In S. cerevisiae, there are two APSES-related genes PHD1 and SOK2 which encode a potent activator and a repressor of pseudohyphal development, respectively (Gimeno and Fink, 1994; Ward et al., 1995). In C. albicans the APSES homologue EFG1 plays a more central role in regulating dimorphic growth such that efg1 homozygous mutants fail to produce true hyphal cells under most inducing conditions and are attenuated for virulence in mouse models (Braun and Johnson, 2000; Lo et al., 1997; Stoldt et al., 1997). In addition, efg1 mutants are unable to form chlamydospores, which are formed under certain starvation and embedded growth conditions by budding from hyphal filaments (Sonneborn et al., 1999). The cellular roles of APSES proteins in A. nidulans, S. cerevisiae and C. albicans suggested that these proteins are major regulators of hyphal-yeast transitions (for example see (Sonneborn et al., 1999)). However, P. marneffei stuA appears to play no role in the hyphal-yeast transition. The difference between the P. marneffei dimorphic switch and that in S. cerevisiae and C. albicans is that the hyphal-yeast transition in the latter involves a budding mode of division whereas the P. marneffei hyphal-yeast transition is fission based. In contrast, the cellular transitions during conidiation in A. nidulans and P. marneffei, and chlamydospore

340

A. Andrianopoulos

formation in C. albicans, are budding hyphal-yeast transitions. Therefore, APSES proteins may not be general regulators of all hyphal-yeast transitions, but may instead be limited to controlling those pathways in which a budding mode of growth is required, and a different mechanism(s) must be present in P. marneffei and other dimorphic fungi such as C. immitis, which produce yeast cells by arthroconidiation. The S. cerevisiae STE12 gene encodes a homeodomain protein required for pseudohyphal development through its interaction with TEC1 (Gavrias et al., 1996; Madhani and Fink, 1997). In addition, STE12p regulates mating and is the target of the mating signalling MAPK cascade in response to pheromone stimulation (Herskowitz, 1995). The C. albicans STE12 homologue, CPH1, is also required for filamentation (Liu et al., 1994; Lo et al., 1997; Singh et al., 1994). Loss of cph1 leads to defects in filamentation and attenuated virulence (Liu et al., 1994; Lo et al., 1997). In contrast the P. marneffei STE12 homologue, stlA, is not required for dimorphic switching, and deletion mutants exhibit wild-type germination, vegetative growth and asexual development (Borneman et al., 2001). STE12 homologues have been studied in a number of other fungi where they appear to be only involved in mating (A. nidulans, Candida lusitaniae) or virulence (C. neoformans) (Chang et al., 2000; Vallim et al., 2000; Wickes et al., 1997; Young et al., 2000; Yue et al., 1999). Although the virulence of P. marneffei stlA mutant strains remains to be examined, it is clear that P. marneffei does not require stlA for hyphal-yeast switching. Signalling and cell polarity genes The P. marneffei dimorphic switch, like that of a number of dimorphic fungal pathogens, is triggered by temperature. Although the molecular details of this signal remain to be determined, the signalling pathways for dimorphic switching in other fungi are better understood (for review see (Lengeler et al., 2000)). In S. cerevisiae pseudohyphal development is transmitted through both mitogen-activated protein kinase (MAPK) and cAMP-dependent signalling pathways. The trigger for the MAPK pathway is unknown and the pathway shares a number of components with the pheromone signalling pathway including the downstream target for the MAP kinase, Ste12p (Cook et al., 1997; Liu et al., 1993; Madhani and Fink, 1997). The cAMP-dependent signal is triggered by the action of the Gpr1p transmembrane receptor and the heterotrimeric Ga subunit Gpa2p (Kubler et al., 1997; Lorenz and Heitman, 1997; Lorenz et al., 2000). Analogous

studies extended into C. albicans have shown that both MAPK and cAMP signalling pathways control dimorphic switching (Csank et al., 1998; Guhad et al., 1998; Kohler and Fink, 1996; Leberer et al., 1996; Sabie and Gadd, 1992; Sonneborn et al., 2000). The Ga subunit-encoding gene gasA from P. marneffei is involved in asexual development but does not appear to affect dimorphic switching (Zuber et al., 2002). Neither deletion of gasA nor expression of dominant active (gasAG42R) and negative (gasAG203R) forms affect the morphogenesis of pre-arthroconidial hyphae or subsequent yeast cell formation. Sequence comparison with other fungal Ga proteins has shown that GasA is more closely related to Gpa1p from S. cerevisiae than to Gpa2p (Zuber et al., 2002). Therefore, if the temperature sensor functions through a Ga subunitpotentiated signalling cascade, then GasA is either not involved or redundant. Whether a P. marneffei homologue of Gpa2 is involved in P. marneffei dimorphic switching remains to be determined, as do the downstream targets of such a signalling pathway. CDC42, a member of the Rho family of small GTPase-encoding genes, is a downstream target of such a signalling pathway in S. cerevisiae and plays a key role in both signalling and polarity determination (for review see (Johnson, 1999; Madden and Snyder, 1998)). Cdc42p interacts with Ste20p, a p21-activated kinase (PAK), which signals to the MAPK cassette required for pheromone signalling and stimulating pseudohyphal growth. In addition, Cdc42p is necessary for cell polarisation and bud emergence (for review see (Johnson, 1999; Madden and Snyder, 1998)). The P. marneffei CDC42 homologue, cflA, is required for correct morphogenesis of yeast cells (Boyce et al., 2000). A dominant negative mutant allele of cflA (cflAG14V) results in a failure to produce true yeast cells. Instead aberrant cells with irregular shapes, branching morphology and poor septation and cell separation are produced. In contrast, a dominant activated allele of cflA (cflAD120A) does not block the production of yeast cells but the resultant cells are swollen and show multiple and random regions of depolarised growth. Despite these defects these yeast and yeast-like cells are still capable of switching to filamentous growth upon incubation at 25 8C (Boyce et al., 2000). This suggests that CflA may not be required for the signalling events which control dimorphic switching but is crucial for cell polarisation events. The cflA mutants also show a number of polarisation defects in hyphal growth at 25 8C, with severe

P. marneffei morphogenesis and pathogenicity

effects at the growing tips of hyphae, and in germination. Intriguingly, no defects in asexual development are evident. The conidiophores are morphologically normal and conidiogenesis proceeds as for the wild type (Boyce et al., 2000). This clearly points to the action of other GTPases in regulating the polarity events during conidiation. In other dimorphic fungi, such as Exophiala dermatitidis, the equivalent CDC42G14V mutation results in a similar loss of polarised growth yielding enlarged and multinucleate cells that are repressed for hyphal growth. However, in contrast to P. marneffei, the dominant-negative mutation (CDC42D120A) has no effect on cell polarisation (Ye and Szaniszlo, 2000). Mutations in the C. albicans CDC42 homologue have been shown to affect cell polarisation of the yeast cells, as for S. cerevisiae (Leberer et al., 1997; Mirbod et al., 1997). In addition, expression of dominant mutant alleles of the C. albicans CDC42 gene have recently been shown to result in the formation of aberrant germ tubes upon induction but did not block polarisation completely (Ushinsky et al., 2002).

A sexual cycle? Many fungi are presumed to be asexual because mating has never been observed. In some cases, more convincing evidence exists from population geneticbased studies. P. marneffei is considered an imperfect ascomycete as no sexual cycle has been discovered (for a review see (Alexopoulos et al., 1996; Kwon-Chung and Bennett, 1992)). Phylogenetic studies have shown that P. marneffei has a number of close sexual relatives (LoBuglio et al., 1993; LoBuglio and Taylor, 1995). This close phylogenetic association between a particular sexual species and an asexual species has been clearly demonstrated for many species in both the Penicillium and Aspergillus group of organisms and is suggestive of the recent loss of a sexual mode of reproduction (Geiser et al., 1996; LoBuglio et al., 1993). Despite the lack of a known sexual cycle, P. marneffei has a highly conserved STE12 homologous gene, stlA. Deletion of stlA appears to have no phenotypic effect in P. marneffei, yet the gene appears fully functional as it is capable of complementing the sexual cycle defects of a steA mutant in A. nidulans (Borneman et al., 2001). The level of sequence conservation between stlA and steA, the A. nidulans homologue, is equivalent to that of other known functional regulatory genes in P. marneffei.

341

These data may indicate that P. marneffei has a cryptic sexual cycle and stlA is a key component of this cycle, therefore selection is still operating on stlA. If P. marneffei has a sexual cycle, then it may not have been detected for a number of possible physiological reasons. For example, P. marneffei may be heterothallic, requiring the presence of two different mating types for fruitful mating. If one of these mating types is rare in the environment or avirulent, and therefore rarely sampled, then the correct pairing may have never been tested. A similar situation exists for C. neoformans serotype D strains where the MATa strains are more prevalent in the environment and in infections than the MATa strains and the latter are less virulent (Kwon-Chung and Bennett, 1978; Kwon-Chung et al., 1992). Another possibility is that appropriate growth conditions have not been used when testing for the sexual cycle. Alternatively, loss of the sexual cycle in P. marneffei may be a recent evolutionary event and there has been insufficient time for measurable divergence of stlA. It has been recently shown that the asexual yeast C. albicans has many conserved genes involved in mating, including the mating type loci homologous to those in S. cerevisiae (Hull and Johnson, 1999; Raymond et al., 1998). Mutation of one of each of these mating loci from reciprocal strains or generation of strains homozygous for each mating type allows C. albicans to be induced to mate and to form tetraploid cells (Hull et al., 2000; Magee and Magee 2000). If P. marneffei is a recently derived asexual species, then this raises an important question regarding the origin of dimorphic switching in P. marneffei. Dimorphic switching has not been demonstrated in any other Penicillium species including the closest relatives of P. marneffei, the teleomorphic species Talaromyces stipitatus, T. intermedius and T. flavus and the anamorphic species P. duclauxii (LoBuglio and Taylor, 1995). This suggests that dimorphic growth and pathogenicity have arisen at the same time as the sexual cycle has been lost and this has occurred after P. marneffei diverged from its closest relatives. The alternate hypothesis that dimorphic switching is evolutionarily ancient and has been repeatedly lost by all other Penicillium species is less likely as it requires the loss of dimorphism many times during the evolution of these organisms. It is tempting to speculate that the two events, gain of dimorphic switching and loss of mating, are directly related in P. marneffei. This is clearly not a universal phenomenon as there are many examples of dimorphic fungi with a sexual cycle. The stuA gene of A. nidulans is required for both asexual and sexual development (Borneman et al.,

342

A. Andrianopoulos

2002). The P. marneffei stuA gene is highly conserved with A. nidulans stuA and can complement the sexual development defects of an A. nidulans stuA deletion strain if expressed from the A. nidulans promoter. However, P. marneffei stuA expressed from its own promoter is not capable of complementing the sexual cycle defects in A. nidulans (Borneman et al., 2002). Therefore, the sexual function of P. marneffei stuA gene product has been conserved but the promoter, and subsequently the expression pattern, has diverged. In contrast, P. marneffei StuA fully complements the asexual development defects of an A. nidulans stuA deletion strain whether it is expressed from the P. marneffei or the A. nidulans promoter sequences (Borneman et al., 2002).

Tools for the molecular genetic study of P. marneffei A large repertoire of molecular genetic techniques have been applied to P. marneffei making it a highly amenable genetic system. Based on the P. marneffei type strain (FRR 2161, ATCC 18224), nitrate reductase (niaD) and orotidine 5'-monophosphate decarboxylase (pyrG) mutants have been selected (Borneman et al., 2001). Using heterologous niaD and pyrG genes, a high-efficiency DNA-mediated transformation system (> 1000 transformants/mg DNA/107 protoplasts) has been developed based on the PEG-mediated protoplast fusion method. A dominant selection transformation scheme using the bacterial bleomycin resistance-encoding gene has also been established (Austin et al., 1990; Borneman et al., 2001). Stable transformation in P. marneffei is based on integration of the transforming DNA. Although the nature of most integration events has not been characterised, directed or homologous integration occurs at a sufficiently high frequency (1 in 10 ± 50) to make allelic replacement or targeted mutations such as gene `knock out' manageable (Borneman et al., 2000 2001, 2002; Zuber et al., 2002). Based on molecular and genetic studies it appears that P. marneffei is essentially haploid. This allows powerful mutagenesis screens for the isolation of desired mutants, such as those defective in dimorphic switching or pathogenicity, to be used. A highefficiency transformation protocol then facilitates the efficient cloning of these genes by complementation of mutant strains. Furthermore, high-efficiency transformation makes sequence-tagged mutagenesis methods such as REMI (restriction enzyme-medi-

ated integration) feasible (Brown and Holden, 1998; Riggle and Kumamoto, 1998). Standard molecular techniques for the isolation of nucleic acids and proteins are available and many cell biology techniques have been established (Borneman et al., 2000, 2001, 2002; Boyce et al., 2000; Zuber et al., 2002). Reporter genes are an essential tool for studying gene expression and gene product localisation. A number of reporter gene systems have been developed for P. marneffei including the Escherichia coli lacZ gene (b-galactosidase), the E. coli uidA gene (b-glucuronidase) and the Aequorea victoria gfp gene (green fluorescent protein) (Andrianopoulos, unpublished) (Borneman et al., 2001). Regulated promoters for the controlled expression of genes based on the ethanol-inducible A. nidulans alcA and the xylose-inducible P. chrysogenum xylP promoters are functional in P. marneffei (Borneman et al., 2000; Boyce et al., 2000; Graessle et al., 1997; Waring et al., 1989). The molecular and genetic tools available for studying P. marneffei allow the cloning and analysis of important developmental and morphogenetic genes. To understand the role these genes play in pathogenicity and virulence, both mammalian cell culture and animal models have been developed for P. marneffei (Cogliati et al., 1997; Kudeken et al., 1996, 1997). The cell culture pathogenicity assay, which relies on the fact that P. marneffei is an intracellular pathogen of macrophages, is an ideal assay by which a large number of mutants can be screened for effects on pathogenicity and the results subsequently tested in an animal model.

Conclusions P. marneffei is an important emerging fungal pathogen of humans which has both a morphologically elaborate asexual development program and a dimorphic switching program. The combination of both of these developmental programs in one organism has made it possible to examine the relationship between these two programs and to assess the level of functional conservation of the regulatory, cell polarity and signalling factors involved in these programs across other dimorphic species (Fig. 6). From these studies it is clear that some genes are functionally conserved between programs and across species, others are conserved at the functional level but used in different ways, and yet others are conserved at the sequence level but functionally diverged. These comparisons have significantly strengthened our understanding of the

P. marneffei morphogenesis and pathogenicity

343

Acknowledgements. Michael Hynes, Kylie Boyce, Richard Todd, and Sophie Zuber are acknowledged for critical comments on the manuscript, and the other members of the laboratory for their contributions. This work was supported by grants from the Australian Research Council and Novozymes A/S.

References

Fig. 6. Control of dimorphic switching and asexual development in P. marneffei. Schematic representation of yeast and hyphal growth and the asexual development structure (conidiophore). The action of the AbaA and StuA transcription factors, the CflA small GTPase and the GasA Ga heterotrimeric subunit in the dimorphic switching and asexual development programs is shown.

biology of P. marneffei and provided insights into mechanisms which control dimorphic switching in other pathogens. With the molecular techniques that are currently available for analysis of P. marneffei, the ease of manipulation of the morphogenic and pathogenic programs, the low risk in the laboratory and its unique biology, P. marneffei makes an ideal model for studying morphogenesis and pathogenicity. It complements other excellent systems in predominantly yeast-like organisms such as C. albicans and C. neoformans. Understanding the biology of pathogenic fungi and how it relates to their pathogenic capacity affords the option of a rational target choice to which drugs for systemic infections can be developed.

Adams, T. H., Boylan, M. T., Timberlake, W. E.: brlA is necessary and sufficient to direct conidiophore development in Aspergillus nidulans. Cell 54, 353 ± 362 (1988). Adams, T. H., Wieser, J. K., Yu, J. H.: Asexual sporulation in Aspergillus nidulans. Microbiol. Mol. Biol. Rev. 62, 35 ± 54 (1998). Alexopoulos, C. J., Mims, C. W., Blackwell, M.: Introductory Mycology. Wiley, New York 1996. Alspaugh, J. A., Perfect, J. R., Heitman, J.: C. neoformans mating and virulence are regulated by the Gprotein a subunit GPA1 and cAMP. Genes Dev. 11, 3206 ± 3217 (1997). Andrianopoulos, A., Timberlake, W. E.: The Aspergillus nidulans abaA gene encodes a transcriptional activator that acts as genetic switch to control development. Mol. Cell. Biol. 14, 2503 ± 2515 (1994). Aramayo, R., Peleg, Y., Addison, R., Metzenberg, R.: Asm1‡, a Neurospora crassa gene related to transcriptional regulators of fungal development. Genetics 144, 991 ± 1003 (1996). Austin, B., Hall, R. M., Tyler, B. M.: Optimized vectors and selection for transformation of Neurospora crassa and Aspergillus nidulans to bleomycin and phleomycin resistance. Gene 93, 157 ± 162 (1990). Beck-Sague, C., Jarvis, W. R.: Secular trends in the epidemiology of nosocomial fungal infections in the United States, 1980 ± 1990. J. Infect. Dis. 167, 1247 ± 1251 (1993). Borneman, A., Hynes, M. J., Andrianopoulos, A.: The abaA homologue of Penicillium marneffei participates in two developmental programs: Conidiation and dimorphic growth. Mol. Microbiol. 38, 1038 ± 1047 (2000). Borneman, A., Hynes, M. J., Andrianopoulos, A.: A basic helix-loop-helix protein with similarity to the fungal morphological regulators, Phd1p, Efg1p and StuA, controls conidiation but not dimorphic growth in Penicillium marneffei. Mol. Microbiol. 44, 621 ± 631 (2002). Borneman, A. R., Hynes, M. J., Andrianopoulos, A.: An STE12 homolog from the asexual, dimorphic fungus Penicillium marneffei complements the defect in sexual development of an Aspergillus nidulans steA mutant. Genetics 157, 1003 ± 1014 (2001). Boyce, K. J., Hynes, M. J., Andrianopoulos, A.: The CDC42 homolog of the dimorphic fungus Penicillium marneffei is required for correct cell polarization

344

A. Andrianopoulos

during growth but not development. J. Bacteriol. 183, 3447 ± 3457 (2000). Boylan, M. T., Mirabito, P. M., Willett, C. E., Zimmerman, C. R., Timberlake, W. E.: Isolation and physical characterization of three essential conidiation genes from Aspergillus nidulans. Mol. Cell. Biol. 7, 3113 ± 3118 (1987). Braun, B. R., Johnson, A. D.: Control of filament formation in Candida albicans by the transcriptional repressor TUP1. Science 277, 105 ± 109 (1997). Braun, B. R., Johnson, A. D.: TUP1, CPH1 and EFG1 make independent contributions to filamentation in Candida albicans. Genetics 155, 57 ± 67 (2000). Brown, J. S., Holden, D. W.: Insertional mutagenesis of pathogenic fungi. Curr. Opin. Microbiol. 1, 390 ± 394 (1998). Chan, Y., Chow, T. C.: Ultrastructural observations on Penicillium marneffei in natural human infection. Ultrastruct. Pathol. 14, 439 ± 452 (1990). Chang, Y. C., Wickes, B. L., Miller, G. F., Penoyer, L. A., Kwon-Chung, K. J.: Cryptococcus neoformans STE12 regulates virulence but is not essential for mating. J. Exp. Med. 191, 871 ± 882 (2000). Chariyalertsak, S., Sirisanthana, S., Supparatpinyo, K., Nelson, K. E.: Seasonal variation of disseminated Penicillium marneffei infections in Northern Thailand: A clue to the reservoir? J. Infect. Dis. 173, 1490 ± 1493 (1996a). Chariyalertsak, S., Vanittanakom, P., Nelson, K. E., Sirisanthana, S., Vanittanakom, N.: Rhizomys sumatrensis and Cannomys badius, new natural hosts of Penicillium marneffei. J. Med. Vet. Mycol. 34, 105 ± 110 (1996b). Clutterbuck, A. J., Timberlake, W. E.: Genetic regulation of sporulation in the fungus Aspergillus nidulans. In: Development: The molecular genetic approach. (D. J. Cove, S. Brody, S. Ottolenghi, V. E. A. Russo, eds.), pp. 103 ± 120. Springer-Verlag, New York 1992. Cogliati, M., Roverselli, A., Boelaert, J. R., Taramelli, D., Lombardi, L., Viviani, M. A.: Development of an in vitro macrophage system to assess Penicillium marneffei growth and susceptibility to nitric oxide. Infect. Immun. 65, 279 ± 284 (1997). Cook, J. G., Bardwell, L., Thorner, J.: Inhibitory and activation functions for MAPK Kss1 in S. cerevisiae filamentous-growth signalling pathway. Nature 390, 85 ± 88 (1997). Cooper, C. R.: From bamboo rats to humans: The odessey of Penicillium marneffei. ASM News 64, 390 ± 397 (1998). Csank, C., Schroppel, K., Leberer, E., Harcus, D., Mohamed, O., Meloche, S., Thomas, D. Y., Whiteway, M.: Roles of the Candida albicans mitogenactivated protein kinase homolog, Cek1p, in hyphal development and systemic candidiasis. Infect. Immun. 66, 2713 ± 2721 (1998). Deng, Z. L., Conner, D. H.: Progressive disseminated penicilliosis caused by Penicllium marneffei: Report of eight cases and differentiation of the causative

organism from Histoplasma capsulatum. Am. J. Clin. Pathol. 84, 323 ± 327 (1985). Edmond, M. B., Wallace, S. E., McClish, D. K.: Nosocomial bloodstream infections in United States hospitals: A three year analysis. Clin. Infect. Dis. 29, 239 ± 244 (1999). Fisherhoch, S. P., Hutwagner, L.: Opportunistic candidiasis ± an epidemic of the 1980s. Clin. Infect. Dis. 21, 897 ± 904 (1995). Garrison, R. G., Boyd, K. S.: Dimorphism in Penicillium marneffei as observed by electron microscopy. Can. J. Microbiol. 19, 1305 ± 1309 (1973). Gavrias, V., Andrianopoulos, A., Gimeno, C. J., Timberlake, W. E.: Saccharomyces cerevisiae TEC1 is required for pseudohyphal growth. Mol. Microbiol. 19, 1255 ± 1263 (1996). Geiser, D. M., Timberlake, W. E., Arnold, M. L.: Loss of meiosis in Aspergillus. Mol. Biol. Evol. 13, 809 ± 817 (1996). Gimeno, C. J., Fink, G. R.: Induction of pseudohyphal growth by overexpression of PHD1, a Saccharomyces cerevisiae gene related to transcriptional regulators of fungal development. Mol. Cell. Biol. 14, 2100 ± 2112 (1994). Gleason, T. G., May, A. K., Caparelli, D., Farr, B. M., Sawyer, R. G.: Emerging evidence of selection of fluconazole-tolerant fungi in surgical intensive care units. Arch. Surgery 132, 1197 ± 1201 (1997). Gow, N. R.: Yeast-hyphal dimorphism. In: The Growing Fungus (N. Gow, G. Gadd, eds.), pp. 403 ± 422. Chapman and Hall, London 1995. Graessle, S., Haas, H., Friedlin, E., Kurnsteiner, H., Stoffler, G., Redl, B.: Regulated system for heterologous gene expression in Penicillium chrysogenum. Appl. Environ. Microbiol. 63, 753 ± 756 (1997). Guhad, F. A., Jensen, H. E., Aalback, B., Csank, C., Mohamed, O., Harcus, D., Thomas, D. Y., Whiteway, M., Hau, J.: Mitogen-activated protein kinase-defective Candida albicans is avirulent in a novel model of localised murine candidiasis. FEMS Microbiol. Lett. 166, 135 ± 139 (1998). Henderson, V. J., Hirvela, E. R.: Emerging and reemerging microbial threats ± nosocomial fungal infections. Arch. Surgery 131, 330 ± 337 (1996). Herskowitz, I.: MAP kinase pathways in yeast: For mating and more. Cell 80, 187 ± 197 (1995). Hilmarsdottir, I., Meynard, J. L., Rogeaux, O., Guermonprez, G., Datry, A., Katlama, C., Brucker, G., Coutellier, A., Danis, M., Gentilini, M.: Disseminated Penicillium marneffei infection associated with human immunodeficiency virus: A report of two cases and a review of 35 published cases. J. Acquir. Immune Defic. Syndr. 6, 466 ± 471 (1993). Hseuh, P.-R., Teng, L.-J., Hung, C.-C., Hsu, J.-H., Yang, P.-C., Ho, S.-W., Luh, K.-T.: Molecular evidence for strain dissemination of Penicillium marneffei: An emerging pathogen in Taiwan. J. Infect. Dis. 181, 1706 ± 1712 (2000). Hull, C. M., Johnson, A. D.: Identification of a mating type-like locus in the asexual pathogenic

P. marneffei morphogenesis and pathogenicity yeast Candida albicans. Science 285, 1271 ± 1275 (1999). Hull, C. M., Raisner, R. M., Johnson, A. D.: Evidence for mating of the ™asexual™ yeast Candida albicans in a mammalian host. Science 289, 307 ± 310 (2000). Johnson, D. I.: Cdc42: An essential Rho-type GTPase controlling eukaryotic cell polarity. Microbiol. Mol. Biol. Rev. 63, 54 ± 105 (1999). Kaufman, L., Standard, P. G., Anderson, S. A., Jalbert, M., Swisher, B. L.: Development of specific fluorescent-antibody test for tissue form of Penicillium marneffei. J. Clin. Microbiol. 33, 2136 ± 2138 (1995). Kohler, J. R., Fink, G. R.: Candida albicans strains heterozygous and homozygous for mutations in mitogen-activated protein kinase signalling components have defects in hyphal development. Proc. Natl. Acad. Sci. USA 93, 13223 ± 13228 (1996). Kubler, E., Mosch, H. U., Rupp, S., Lisanti, M. P.: Gpa2p, a G-protein a-subunit, regulates growth and pseudohyphal development in Saccharomyces cerevisiae via a cAMP-dependent mechanism. J. Biol. Chem. 272, 20321 ± 20323 (1997). Kudeken, N., Kawakami, K., Kusano, N., Saito, A.: Cellmediated immunity in host resistance against infection caused by Penicillium marneffei. J. Med. Vet. Mycol. 34, 371 ± 378 (1996). Kudeken, N., Kawakami, K., Saito, A.: CD4‡ T cellmediated fatal hyperinflammatory reactions in mice infected with Penicillium marneffei. Clin. Exp. Immunol. 107, 468 ± 473 (1997). Kulkarni, R. K., Nickerson, K. W.: Nutritional control of dimorphism in Ceratocystis ulmi. Exp. Mycol. 5, 148 ± 154 (1981). Kwon-Chung, K. J., Bennett, J. E.: Distribution of a and a mating types of Cryptococcus neoformans among natural and clinical isolates. Am. J. Epidemiol. 108, 337 ± 340 (1978). Kwon-Chung, K. J., Bennett, J. E.: Medical Mycology. Lea and Febiger, Philadelphia 1992. Kwon-Chung, K. J., Edman, J. C., Wickes, B. L.: Genetic association of mating types and virulence in Cryptococcus neoformans. Infect. Immun. 60, 602 ± 605 (1992). Leberer, E., Harcus, D., Broadbent, I. D., Clark, K. L., Dignard, D., Ziegelbauer, K., Schmidt, A., Gow, N., Brown, A., Thomas, D. Y.: Signal transduction through homologs of the Ste20p and Ste7p protein kinases can trigger hyphal formation in the pathogenic fungus Candida albicans. Proc. Natl. Acad. Sci. USA 93, 13217 ± 13222 (1996). Leberer, E., Ziegelbauer, K., Schmidt, A., Harcus, D., Dignard, D., Ash, J., Johnson, L., Thomas, D. Y.: Virulence and hyphal formation of Candida albicans require the STE20p ± like protein kinase CaCLA4p. Curr. Biol. 7, 539 ± 546 (1997). Lengeler, K. B., Davidson, R. C., D`Souza, C., Harashima, T., Shen, W.-C., Wang, P., Pan, X., Waugh, M., Heitman, J.: Signal transduction cascades regulating fungal development and virulence. Microbiol. Mol. Biol. Rev. 64, 746 ± 785 (2000).

345

Liu, H. P., Styles, C. A., Fink, G. R.: Elements of the yeast pheromone response pathway required for filamentous growth of diploids. Science 262, 1741 ± 1744 (1993). Liu, H. P., Kohler, J., Fink, G. R.: Suppression of hyphal formation in Candida albicans by mutation of a STE12 homolog. Science 266, 1723 ± 1726 (1994). Lo, H.-S., Kohler, J. R., DiDomenico, B., Loebenberg, D., Cacciapuoti, A., Fink, G. R.: Nonfilamentous C. albicans mutants are avirulent. Cell 90, 939 ± 949 (1997). LoBuglio, K. F., Pitt, J. I., Taylor, J. W.: Phylogenetic analysis of two ribosomal DNA regions indicates multiple independent losses of a sexual Talaromyces state among asexual Penicillium species in subgenus Biverticillium. Mycologia 84, 592 ± 604 (1993). LoBuglio, K. F., Taylor, J. W.: Phylogeny and PCR identification of the human pathogenic fungus Penicillium marneffei. J. Clin. Microbiol. 33, 85 ± 89 (1995). Lorenz, M. C., Heitman, J.: Yeast pseudohyphal growth is regulated by GPA2, a G protein alpha homolog. EMBO J. 16, 7008 ± 7018 (1997). Lorenz, M. C., Pan, X., Harashima, T., Cardenas, M. E., Xue, Y., Hirsch, J. P., Heitman, J.: The G proteincoupled receptor Gpr1 is a nutrient sensor that regulates pseudohyphal differentiation in Saccharomyces cerevisiae. Genetics 154, 609 ± 622 (2000). Madden, K., Snyder, M.: Cell polarity and morphogenesis in budding yeast. Annu. Rev. Microbiol. 52, 687 ± 744 (1998). Madhani, H. D., Fink, G. R.: Combinatorial control required for the specificity of yeast MAPK signaling. Science 275, 1314 ± 1317 (1997). Magee, B. B., Magee, P. T.: Induction of mating in Candida albicans by construction of MTLa and MTLalpha strains. Science 289, 310 ± 313 (2000). Miller, K. Y., Wu, J., Miller, B. L.: StuA is required for cell pattern formation in Aspergillus. Genes Dev. 6, 1770 ± 1782 (1992). Mirabito, P. M., Adams, T. H., Timberlake, W. E.: Interactions of three sequentially expressed genes control temporal and spatial specificity in Aspergillus development. Cell 57, 859 ± 868 (1989). Mirbod, F., Nakashima, S., Kitajima, Y., Cannon, R. D., Nozawa, Y.: Molecular cloning of a RHO family Cacdc42 gene from Candida albicans and its mRNA expression changes during morphogenesis. J. Med. Vet. Mycol. 35, 173 ± 179 (1997). Orlowski, M.: Mucor dimorphism. Microbiol. Rev. 55, 234 ± 258 (1991). Pitt, J. I.: The genus Penicillium and its teleomorphic states Eupenicillium and Talaromyces. Academic Press, Inc., London 1979. Raymond, M., Dignard, D., Alarco, A. M., Mainville, N., Magee, B. B., Thomas, D. Y.: A Ste6p/P-glycoprotein homologue from the asexual yeast Candida albicans transports the a-factor mating pheromone in Saccharomyces cerevisiae. Mol. Microbiol. 27, 587 ± 598 (1998).

346

A. Andrianopoulos

Riggle, P. J., Kumamoto, C. A.: Genetic analysis in fungi using restriction-enzyme-mediated integration. Curr. Opin. Microbiol. 1, 395 ± 399 (1998). Sabie, F. T., Gadd, G. M.: Effect of nucleosides and nucleotides and the relationship between cellular adenosine 3'-5'-cyclic monophosphate (cyclic AMP) and germ tube formation in Candida albicans. Mycopathologia 119, 147 ± 156 (1992). Schweizer, A., Rupp, S., Taylor, B. N., Rollinghoff, M., Schroppel, K.: The TEA/ATTS transcription factor CaTec1p regulates hyphal development and virulence in Candida albicans. Mol. Microbiol. 38, 435 ± 445 (2000). Segretain, G.: Penicillium marneffei n. sp., agent d'une mycose du systeme reticulo-endothelial. Mycopathologia 11, 327 ± 353 (1959). Sewall, T. C., Mims, C. W., Timberlake, W. E.: abaA controls phialide differentiation in Aspergillus nidulans. Plant Cell 2, 731 ± 739 (1990). Singh, P., Ganesan, K., Malathi, K., Ghosh, D., Datta, A.: Acpr, a STE12 homologue from Candida albicans, is a strong inducer of pseudohyphae in Saccharomyces cerevisiae haploids and diploids. Biochem. Biophys. Res. Commun. 205, 1079 ± 1085 (1994). Sirisanthana, T., Supparatpinyo, K.: Epidemiology and management of penicilliosis in human immunodeficiency virus-infected patients. Int. J. Infect. Dis. 3, 48 ± 53 (1998). Sirisanthana, T., Supparatpinyo, K., Perriens, J., Nelson, K. E.: Amphotericin B and itraconazole for treatment of disseminated Penicillium marneffei infection in human immunodeficiency virus-infected patients. Clin. Infect. Dis. 26, 1107 ± 1110 (1998). Sonneborn, A., Bockmuhl, D. P., Ernst, J. F.: Chlamydospore formation in Candida albicans requires the Efg1p morphogenetic regulator. Infect. Immun. 67, 5514 ± 5517 (1999). Sonneborn, A., Bockmuhl, D. P., Gerads, M., Kurpanek, K., Sanglard, D., Ernst, J. F.: Protein kinase A encoded by TPK2 regulates dimorphism of Candida albicans. Mol. Microbiol. 35, 386 ± 396 (2000). Sternberg, S.: The emerging fungal threat. Science 266, 1632 ± 1634 (1994). Stoldt, V. R., Sonneborn, A., Leuker, C. E., Ernst, J. F.: Efg1p, an essential regulator of morphogenesis of the human pathogen Candida albicans, is a member of a conserved class of bHLH proteins regulating morphogenetic processes in fungi. EMBO J. 16, 1982 ± 1991 (1997). Supparatpinyo, K., Chiewchanvit, S., Hirunsri, P., Uthammachai, C., Nelson, K. E., Sirisanthana, T.: Penicillium marneffei infection in patients infected with human immunodeficiency virus. Clin. Infect. Dis. 14, 871 ± 874 (1992). Supparatpinyo, K., Nelson, K. E., Merz, W. G., Breslin, B. J., Cooper, C. R., Kamwan, C., Sirisanthana, T.: Response to antifungal therapy by human immunodeficiency virus-infected patients with disseminated Penicillium marneffei infections and in vitro susceptibilities of isolates from clinical specimens.

Antimicrob. Agents Chemother. 37, 2407 ± 2411 (1993). Supparatpinyo, K., Khamwan, C., Baosoung, V., Nelson, K. E., Sirisanthana, T.: Disseminated Penicillium marneffei infection in Southeast Asia. Lancet 344, 110 ± 113 (1994). Szaniszlo, P. J., Jacobs, C. W., Geis, P. A.: Dimorphism: Morphological and biochemical aspects. In: Fungi Pathogenic for Humans and Animals. (D. H. Howard, ed.) pp. 323 ± 436. Marcel Dekker, Inc., New York 1983. Ungpakorn, R.: Cutaneous manifestations of Penicillium marneffei infections. Curr. Opin. Infect. Dis. 13, 129 ± 134 (2000). Ushinsky, S. C., Harcus, D., Ash, J., Dignard, D., Marcil, A., Morchhauser, J., Thomas, D. Y., Whiteway, M., Leberer, E.: CDC42 Is required for polarized growth in human pathogen Candida albicans. Eukaryotic Cell 1, 95 ± 104 (2002). Vallim, M. A., Miller, K. Y., Miller, B. L.: Aspergillus SteA (Sterile12-like) is a homeodomain-C2/H2-Zn2‡ finger transcription factor required for sexual reproduction. Mol. Microbiol. 36, 290 ± 301 (2000). Vossler, J. L.: Penicillium marneffei: An emerging fungal pathogen. Clin. Microbiol. Newslett. 23, 25 ± 29 (2001). Walsh, T. J., Groll, A. H.: Emerging fungal pathogens: Evolving challenges to immunocompromised patients in the twenty-first century. Transpl. Infect. Dis. 1, 247 ± 261 (1999). Ward, M. P., Gimeno, C. J., Fink, G. R., Garrett, S.: SOK2 may regulate cyclic AMP-dependent protein kinase-stimulated growth and pseudohyphal development by repressing transcription. Mol. Cell. Biol. 15, 6854 ± 6863 (1995). Waring, R. B., May, G. S., Morris, N. R.: Characterization of an inducible expression system in Aspergillus nidulans using alcA and tubulin-coding genes. Gene 79, 119 ± 130 (1989). Weinberger, M., Sacks, T., Sulkes, J., Shapiro, M., Polacheck, I.: Increasing fungal isolation from clinical specimens ± experience in a university hospital over a decade. J. Hosp. Infect. 35, 185 ± 195 (1997). Wickes, B. L., Edman, U., Edman, J. C.: The Cryptococcus neoformans STE12a gene: A putative Saccharomyces cerevisiae STE12 homologue that is mating type specific. Mol. Microbiol. 26, 951 ± 960 (1997). Wieser, J., Yu, J. H., Adams, T. H.: Dominant mutations affecting both sporulation and sterigmatocystin biosynthesis in Aspergillus nidulans. Curr. Genet. 32, 218 ± 224 (1997). Wong, S. S. Y., Siau, H., Yuen, K. Y.: Penicilliosis marneffei ± West meets East. J. Med. Microbiol. 48, 973 ± 975 (1999). Yaar, L., Mavarech, M., Koltin, Y.: A Candida albicans RAS-related gene (CaRSR1) is involved in budding, cell morphogenesis and hypha development. Microbiology 143, 3033 ± 3044 (1997).

P. marneffei morphogenesis and pathogenicity Ye, X., Szaniszlo, P. J.: Expression of a constitutively active Cdc42 homologue promotes development of sclerotic bodies but represses hyphal growth in the zoopathogenic fungus Wangiella (Exophiala) dermatitidis. J. Bacteriol. 182, 4941 ± 4950 (2000). Ye, X. S., Lee, S. L., Wolkow, T. D., McGuire, S. L., Hamer, J. E., Wood, G. C., Osmani, S. A.: Interaction between developmental and cell cycle regulators is required for morphogenesis in Aspergillus nidulans. EMBO J. 18, 6994 ± 7001 (1999). Young, L. Y., Lorenz, M. C., Heitman, J.: A STE12 homolog is required for mating but dispensable for filamentation in Candida lusitaniae. Genetics 155, 17 ± 29 (2000). Yu, J. H., Wieser, J., Adams, T. H.: The Aspergillus FlbA RGS domain protein antagonizes G-protein signalling

347

to block proliferation and allow development. EMBO J. 15, 5184 ± 5190 (1996). Yue, C., Cavallo, L. M., Alspaugh, J. A., Wang, P., Cox, G. M., Perfect, J. R., Heitman, J.: The STE12 homolog is required for haploid filamentation but largely dispensable for mating and virulence in Cryptococcus neoformans. Genetics 153, 1601 ± 1615 (1999). Zeng, Q., Morales, A. J., Cottarel, G.: Fungi and humans: Closer than you think. Trends Genet. 17, 682 ± 694 (2001). Zuber, S., Hynes, M. J., Andrianopoulos, A.: G-protein signaling mediates asexual development at 25 8C but has no effect on yeast-like growth at 37 8C in the dimorphic fungus Penicillium marneffei. Eukaryotic Cell 1, 440 ± 447 (2002).