Portal de Periódicos da CAPES

A cyclin-dependent kinase family member (PHOA) is required to link developmental fate to environmental conditions in Aspergillus nidulans

1998; Springer Nature; Volume: 17; Issue: 14 Linguagem: Inglês

10.1093/emboj/17.14.3990

1460-2075

Henk‐Jan Bussink, Stephen A. Osmani,

Protist diversity and phylogeny

Article15 July 1998free access A cyclin-dependent kinase family member (PHOA) is required to link developmental fate to environmental conditions in Aspergillus nidulans Henk-Jan Bussink Henk-Jan Bussink Carlsberg Laboratory, Department of Physiology, GI, Carlsbergvej 10, DK-2500 Copenhagen Valby, Denmark Search for more papers by this author Stephen A. Osmani Corresponding Author Stephen A. Osmani Henry Hood Research Program, Weis Center for Research, Pennsylvania State University College of Medicine, Danville, PA, 17822 USA Search for more papers by this author Henk-Jan Bussink Henk-Jan Bussink Carlsberg Laboratory, Department of Physiology, GI, Carlsbergvej 10, DK-2500 Copenhagen Valby, Denmark Search for more papers by this author Stephen A. Osmani Corresponding Author Stephen A. Osmani Henry Hood Research Program, Weis Center for Research, Pennsylvania State University College of Medicine, Danville, PA, 17822 USA Search for more papers by this author Author Information Henk-Jan Bussink2 and Stephen A. Osmani 1 1Henry Hood Research Program, Weis Center for Research, Pennsylvania State University College of Medicine, Danville, PA, 17822 USA 2Carlsberg Laboratory, Department of Physiology, GI, Carlsbergvej 10, DK-2500 Copenhagen Valby, Denmark *Corresponding author. E-mail: [email protected] The EMBO Journal (1998)17:3990-4003https://doi.org/10.1093/emboj/17.14.3990 PDFDownload PDF of article text and main figures. ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions ShareFacebookTwitterLinked InMendeleyWechatReddit Figures & Info We addressed the question of whether Aspergillus nidulans has more than one cyclin-dependent kinase gene and identified such a gene, phoA, encoding two PSTAIRE-containing kinases (PHOAM1 and PHOAM47) that probably result from alternative pre-mRNA splicing. PHOAM47 is 66% identical to Saccharomyces cerevisiae Pho85. The function of this gene was studied using phoA null mutants. It functions in a developmental response to phosphorus-limited growth but has no effect on the regulation of enzymes involved in phosphorus acquisition. Aspergillus nidulans shows both asexual and sexual reproduction involving temporal elaboration of different specific cell types. We demonstrate that developmental decisions in confluent cultures depend upon both the initial phosphorus concentration and the inoculation density and that these factors influence development through phoA. In the most impressive cases, absence of phoA resulted in a switch from asexual to sexual development (at pH 8), or the absence of development altogether (at pH 6). The phenotype of phoA deletion strains appears to be specific for phosphorus limitation. We propose that PHOA functions to help integrate environmental signals with developmental decisions to allow ordered differentiation of specific cell types in A.nidulans under varying growth conditions. The results implicate a putative cyclin-dependent kinase in the control of development. Introduction Cyclin-dependent kinases (CDKs) were first identified during genetic screens aimed at isolating cell cycle-specific genes from yeast (Nasmyth and Reed, 1980; Beach et al., 1982; Nurse, 1990; Nasmyth, 1993). Since their original isolation, numerous other protein kinases have been isolated (Uesono et al., 1987; Toh-e et al., 1988; Meyerson et al., 1992) which are structurally similar to the founding members Cdc28 and Cdc2. These kinases display the unifying feature of requiring a bound cyclin partner for function (Peeper et al., 1993; Kaffman et al., 1994; Jeffrey et al., 1995). In the yeast Saccharomyces cerevisiae, CDC28 encodes the major cell cycle-specific CDK that promotes cell cycle transitions when bound to different cyclin partners (Nasmyth, 1993). The nearest relative to Cdc28 in S.cerevisiae is the Pho85 (Toh-e et al., 1988) protein kinase (51% identity). PHO85 was first identified as a negative regulator of phosphate-repressible genes involved in the acquisition of phosphate (Hirst et al., 1994; Kaffman et al., 1994; Schneider et al., 1994; O'Neill et al., 1996; for review, see Lenburg and O'Shea, 1996). PHO85 subsequently has been implicated not only in the regulation of acid phosphatase expression but also in cell cycle regulation (Espinoza et al., 1994; Measday et al., 1994), regulation of carbon metabolism (Gilliquet and Berben, 1993; Huang et al., 1996; Timblin et al., 1996), repression of stress response genes (Timblin and Bergman, 1997) and vacuole inheritance (Nicolson et al., 1995). Consistent with these apparently disparate functions, Pho85 interacts with numerous different cyclin partners. To regulate acid phosphatase expression, Pho85 binds to the Pho80 cyclin-like molecule (Kaffman et al., 1994; Schneider et al., 1994; O'Neill et al., 1996). To contribute to the regulation of Start, Pho85 interacts with the G1-specific cyclins Pcl1 and Pcl2 (Espinoza et al., 1994; Measday et al., 1994). Recently, Pho85 has been shown to interact with a further seven cyclin-like proteins (Measday et al., 1997), bringing the total estimate to 10 potential cyclin partners for Pho85. Thus Pho85 interacts with multiple cyclins which presumably direct it to fulfill numerous non-essential functions in the life cycle of S.cerevisiae. One CDK gene (nimXcdc2) has been isolated from Aspergillus nidulans, which encodes NIMXcdc2, a protein kinase essential to promote transitions through the cell cycle (Osmani et al., 1994). To study the role of other CDKs in a genetically facile multicellular microorganism, we sought to isolate other potential CDK genes using a PCR approach in A.nidulans. This filamentous fungus produces 10 different cell types (Champe and Simon, 1992; Champe et al., 1994), some of which are genetically programmed for elaboration only during asexual or sexual spore formation (Timberlake, 1990). In addition, this fungus has a wide and diverse biosynthetic capacity and is able to grow from completely inorganic components when supplied with a suitable carbon source. Our studies identify a gene with high identity to the S.cerevisiae PHO85 gene which, unlike S.cerevisiae PHO85, is not required for regulation of scavenging phosphatase production. The A.nidulans gene, phoA, is instead essential at the interphase between environmental conditions and the developmental programs that allow switching from vegetative growth and commitment to either asexual or sexual spore formation. The data implicate a putative CDK in the modulation of the developmental program of a multicellular eukaryote. Results Molecular cloning of the phoA gene Preliminary attempts to identify cdc2-related genes by hybridization, using a nimXcdc2 cDNA as the probe and employing hybridization conditions of moderate stringency, did not yield indications of a gene highly related to nimXcdc2. Therefore, a PCR-based approach was adopted using degenerate primers that had been designed on the basis of sequences of kinase domains I (Hanks et al., 1988) [oligo ERK1(F)], VIII [ERK3(R)] and VII [ERK4(R)] of both functional cdc2 homologs and more distantly related members of the cdc2 protein kinase family (Simon et al., 1986; Toh-e et al., 1988; Boulton et al., 1991; Meyerson et al., 1992). PCR products derived from six putative A.nidulans protein kinase genes, including those of the previously cloned nimXcdc2 (Osmani et al., 1994) and crkA (M.Mischke and N.R.Morris, personal communication) genes, were identified as detailed in Materials and methods. One of the newly isolated sequences showed a single open reading frame that has the capacity to encode a protein sequence 60% identical to that of the C-terminal domain kinase (Cismowski et al., 1995) encoded by the KIN28 gene (Simon et al., 1986) in the region specified by, but not including, primers ERK1(F) and ERK4(R). In the other three cases, the presence of intron sequences in the PCR fragments recovered was assumed on the basis of alignments of the conceptual translation products and protein kinases of known structure, and also the introns tentatively identified showed sequences identical or similar to those in other filamentous fungal introns (Gurr et al., 1987). Two of the derived amino acid sequences showed highest similarity to MAP kinases and the other one to Pho85. The latter was the only one showing a sequence highly similar to the so-called PSTAIRE motif (Pines and Hunter, 1991), characteristic of functional cdc2 homologs, and the corresponding gene was therefore selected for further analysis. It was designated phoA, on the basis of the phenotype of A.nidulans strains lacking this gene. Structure of the phoA gene Most of the protein-coding region of the phoA gene (Figure 1) could be assigned on the basis of homology of the encoded protein, PHOA, with other protein kinases. The presence of the two downstream introns was confirmed by showing their absence in a PCR fragment amplified from an A.nidulans λgt10 cDNA library. The gene has two possible in-frame translation initiation codons, the more upstream one being located in a 66 bp sequence that shows all the sequence characteristics of an Aspergillus intron. Preliminary attempts to amplify fragments from high-quality cDNA libraries using forward primers located before this putative intron were unsuccessful. Therefore, a first-strand cDNA synthesis was carried out using a phoA-specific primer (RACE1), its products were used directly in PCR reactions employing a forward primer (IP1) that has its 3′-end three bases before the putative intron, and the two phoA-specific products produced were cloned and sequenced. The shorter product showed the absence of all the three introns as depicted in Figure 1, establishing that the upstream translation initiation codon is present in an intron. However, this intron had not been removed in the longer product, although the two downstream introns had been spliced out. The longer product was the slightly more abundant product, in duplicate PCR reactions, suggesting that inefficient splicing of the first intron might be of physiological relevance, resulting in two forms of PHOA that differ by the presence or absence of an N-terminal amino acid sequence of 46 amino acids. Figure 1.Structure of the phoA gene. Intron sequences are in lower case, the PHOA-encoding sequence is separated in codons, and alternative start methionines are indicated in the three-letter code. The putative translation termination codon (end) and the start of the poly(A)+ tail (A+) of transcripts derived from an immediately upstream gene are indicated below the sequence. The sequence between the underlined DraI and EcoRV sites has been replaced in the phoA1 allele. The closed triangles indicate the start of phoA sequence in expression plasmids pPAP(Met1) and pPAP(Met47). DDBJ/EMBL/GenBank accession No. u59215. Download figure Download PowerPoint To investigate the occurrence of these two forms of PHOA in vivo, and to study gene function (see below), strains were generated in which the major portion of the phoA gene (between the DraI and EcoRV sites indicated in Figure 1) had been replaced by the A.nidulans pyrG gene (Oakley et al., 1987). This deletion allele was designated phoA1. A phoA1 strain was transformed with plasmids that contained the A.nidulans pyroA gene as the selectable marker and one of two versions of the phoA structural gene under the control of the promoter of the alcohol dehydrogenase-encoding alcA gene (Waring et al., 1989). The phoA sequence present in plasmid pPAP(M1) has the capacity to encode the larger form of PHOA starting at amino acid position 1 (Figure 1), whereas plasmid pPAP(M47) can only encode the smaller form of PHOA starting at position 47. The wild-type A.nidulans strain R153, the phoA1 mutant and complemented strains obtained with plasmids pPAP(M1) and pPAP(M47) were grown for 18 h on minimal medium containing glucose as the carbon source and extracts were then prepared for Western blot analysis (Figure 2). The antibody raised against Escherichia coli-expressed PHOA is not entirely PHOA-specific as it recognizes some other proteins in the extract from the phoA1 mutant of a mobility that can be expected for a CDK. However, it is quite clear that deletion of the phoA gene results in the absence of two immunoreactive proteins of apparent mol. masses of 32 and ∼40 kDa found in the wild-type strain. The larger form of PHOA, PHOAM1, is expressed from plasmid pPAP(M1), while plasmid pPAP(M47) only directs expression of the smaller form, PHOAM47. These results show that both possible translation initiation codons in the phoA gene are indeed used. It can be seen that PHOAM47 is also expressed at a relatively low level from plasmid pPAP(M1), possibly resulting from read-through of the first translation codon or from initiation of transcription downstream of this codon, which could be due to the use of the alcA promoter. As expected, much higher levels of PHOA were observed when its synthesis in the transformants was induced with ethanol, but it was also much more degraded compared with glucose-grown cells (data not shown). Preliminary experiments gave no indications of regulation of the ratio of the two PHOA forms. Figure 2.The phoA gene encodes two polypeptides, PHOAM1 and PHOAM47, that are expressed from different translation initiation codons. The phoA deletion strain PHOΔ17 (lane 1), the wild-type strain R153 (lane 2) and strain PHOΔ17 complemented with plasmid pPAP(M1) (lane 3) or pPAP(M47) (lane 4) were grown for 18 h in minimal medium and mycelial extracts were analyzed by Western blotting using a polyclonal antibody raised against E.coli-expressed PHOA. Download figure Download PowerPoint The sequence of phoA is preceded by a long open reading frame, from position 2 to 604 and in the same direction of transcription as phoA, that is likely to extend much further upstream on the basis of sequence obtained from only one strand of DNA not included in Figure 1. Putative transcripts having a poly(A) tail following the bases at positions 689 and 697 were identified using a 3′-RACE procedure. The partial conceptual translation product shows low, but probably significant, similarity to Sir2-related proteins (Brachmann et al., 1995). Homology of PHOA to other protein kinases The sequence of PHOA shows all the invariant residues found in protein kinases (Hanks et al., 1988), except that an aspartic acid (D224) rather than a glutamic acid is present in subdomain VIII, as has been observed recently for several protein kinases (Figure 3). PHOA is most similar to Pho85 (Toh-e et al., 1988), showing 66.3% identity to PHOAM47 (Figure 3). It is also highly similar (64.7% identical) to the Dictyostelium discoidum protein tentatively designated Crp kinase (Michaelis and Weeks, 1993). While the function of this protein is unknown, the amount of crp mRNA is very low in vegetative cells and increases dramatically during development. The vertebrate protein kinases showing highest similarity to PHOA found in databases were human cdk5 (Meyerson et al., 1992) and Xenopus laevis cdk2 (Paris et al., 1991). Figure 3.Similarity between PHOA and other proteins. (A) Alignment of PHOAM1 and related protein kinases. Conserved amino acids are indicated by dashes, and where alignment has been improved by the introduction of gaps, this is marked by points. The perfect PSTAIRE motif in the cdk2Xl sequence is underlined twice and Ser(212) and Asp(224) in PHOA are indicated by closed and open diamonds, respectively. Similarities were calculated on the basis of PHOAM47 sequence. (B) Similarity between the N-terminal extension of PHOAM1, sequences in the N-terminal non-catalytic regions of yeast MAP kinase kinases and Neurospora crassa NUC-1. See text for details. Download figure Download PowerPoint PHOA contains a sequence highly similar to the 16 amino acid PSTAIRE motif present in functional cdc2 homologs (Pines and Hunter, 1991) that has been identified as a key element in the activation of cdk2 by cyclin A (Jeffrey et al., 1995). The two amino acid substitutions in PHOA with respect to this PSTAIRE motif are also found in Pho85. Cdc2 and cdk2 are activated further by phosphorylation of a specific threonine residue by CDK-activating kinase. A serine, rather than a threonine, is found at the corresponding positions in PHOA (S212) and some other cdc2-related protein kinases. While mutagenesis of this serine to alanine results in a non-functional Pho85 kinase (Santos et al., 1995), the activity of cdk5 does not depend on phosphorylation (Qu et al., 1995). The 46 amino acid N-terminal extension of PHOAM1 is rich in serine, proline and leucine, together comprising 50% of the residues. The similarity of the PHOAM1 extension to regions of other proteins (Figure 3B), essentially consisting of a serine/threonine-rich region of six amino acids and a proline-rich region separated by ∼20 amino acids, may suggest a regulatory function for the extension (see Discussion). Absence of PHOA does not lead to constitutive expression of phosphatases Because deletion of PHO85 leads to constitutive expression of the phosphatase genes under its control, the A.nidulans phoA1 strains were also tested for the activity of three phosphate-regulated extracellular phosphatases, alkaline phosphatase, acid phosphatase and phosphodiesterase, by activity staining of colonies grown in the presence or absence of inorganic phosphate (Pi). Agar contains sufficient phosphate to sustain growth. Deletion of the phoA gene did not result in constitutive synthesis of any of the three activities on Pi-rich medium (11 mM), while the induced activities on medium without Pi were comparable with those of the wild-type strain (Figure 4). Measurement of acid phosphatase activity produced in liquid culture using phosphate-depleted rich media that had been fully reconstituted with phosphate, or with only 0.1 mM phosphate (low-Pi), confirmed that the regulation of acid phosphatase gene expression in response to phosphorus availability was not affected by the phoA1 deletion. The phoA1 strain also grew at the same rate as the wild-type under different Pi concentration growth conditions as revealed by dry weight measurements (Table I). While yeast pho85 null mutants show reduced growth on acetate, ethanol and glycerol (Gilliquet and Berben, 1993), no obvious phenotype was observed when the phoA1 strain was grown on these carbon sources (data not shown). Figure 4.Normal regulation of excreted phosphatase activities in the absence of phoA. Colonies were stained for (A) alkaline phosphatase, (B) phosphodiesterase and (C) acid phosphatase. In each panel, three strains containing the phoA1 mutation (top) and non-deleted control strains (lower) were grown on agar plates containing 11 mM Pi or no added Pi as indicated for 22 h prior to activity staining. Download figure Download PowerPoint Table 1. PHOA is not required to repress the synthesis of extracellular acid phosphatase under phosphate-sufficient growth conditions Strain Acid phosphatasea Biomassb R153 (wild-type) 11 mM Pi 0.13 1.56 0.1 mM Pi 32.0 0.90 PHOΔ17 (phoA1) 11 mM Pi 0.10 1.69 0.1 mM Pi 21.5 0.84 a Acid phosphatase activity of culture filtrates expressed as nmol of p-nitrophenol liberated/mg of mycelial dry weight/min. b Grams of mycelial dry weight/l after 14 h. A temperature-sensitive phoA1–nimX3cdc2 double mutant was obtained to investigate if the absence of PHOA enhances the effect of impaired NIMXcdc2 function as may be expected if PHOA functioned redundantly with nimXcdc2. However, no significant effect of the presence of the phoA1 allele was observed when the nimX3cdc2 mutant and the double mutant were grown at 37°C, at which temperature nimX3cdc2 strains grow poorly compared with wild-type strains. phoA1 strains appeared to form conidia (asexual spores) normally and could be crossed to each other to yield viable recombined progeny. The phoA gene is therefore apparently not essential for any part of the A.nidulans life cycle under standard laboratory growth conditions. PHOA functions in the cellular response to phosphorus limitation When the phoA1 mutant was grown at pH 8 in the presence of normal Pi (11 mM) levels, it grew and differentiated like the wild-type strain, and both strains produced similar amounts of a brown pigment (Figure 5). However, on low-Pi medium, only very little pigment was formed in the wild-type strain, whereas the mutant overproduced the brown pigment. The contrast in pigment production between the phoA1 and wild-type strains was greatly diminished when tested at pH 6.5, because the wild-type strain produced more of the brown pigment at this pH, while the phoA1 strain produced less than at pH 8. Figure 5.PHOA is required to repress the synthesis of a pigment during phosphate-limited growth at pH 8. Upper panel: two colonies each of the wild-type strain R153 (on the left) and the phoA deletion strain PHOΔ17 (on the right) were grown for 5 days on regular pH 8 medium, or on media with only 0.1 mM P ior no added sulfate. The back of the plates is shown. Lower panel: complementation of strain PHOΔ17 with plasmids pPAPnc, pPAP(M1) and pPAP(M47). Colonies were grown for 5 days in the presence of 0.1 mM Pi. Strain PHOΔ17 was inoculated at the top of the plates and strain R153 in the lower left corner. The upper face of the plates is shown. Download figure Download PowerPoint The production of the pigment started around 2 days after inoculation, after a colony of considerable size had been formed, and it appeared to be secreted into the agar medium. All phoA deletion strains isolated showed this phenomenon, and in a cross of strains PHOΔ17 and HB23, this trait and uridine prototrophy, i.e. the phoA1 allele, co-segregated as a single Mendelian marker. While we have not analyzed this pigment, it may be a phenolic compound (Hermann et al., 1983; Champe and Simon, 1992; Champe et al., 1994). An in situ staining method for sexual development (Hermann et al., 1983) depends on the activity of laccase II, a developmentally regulated phenoloxidase, and coupling of the reaction product to DMA (3,5-dimethylaniline). phoA1 colonies (carrying the yA2 allele to inactivate laccase I) producing the pigment showed evenly distributed staining upon application of this method, even when the laccase substrate was omitted from the reaction mixture (data not shown) presumably due to reaction of the pigment and DMA. Wild-type strains produced only few cleistothecia (sexual fruiting bodies) on low-Pi medium, whereas the phoA1 strain showed abundant sexual development under these conditions. This can be seen in Figure 6 where the green color is caused by conidia and the yellow/tan structures on the surface of the colonies indicate sexual development. This yellow color stems from so-called hulle cells whose appearance is the first morphological manifestation of sexual development, and sexual spores are formed in developing cleistothecia that are surrounded by these hulle cells. On complex MAG medium, the phoA1 strain produced typical large clusters of cleistothecia that were, in most experiments, absent in the wild-type strain R153, although some of these clusters were also seen in R153 in an experiment where they were particularly abundant in the phoA1 strain. Figure 6.Morphology of wild-type and mutant strains grown on 0.1 mM Pi medium for 7 days. Strains HB38 (WT), HB9 (ΔphoA), HB37 (palcA1) and HB34 (ΔphoA, palcA1) are shown. Download figure Download PowerPoint The phoA1 mutation was complemented with the cloned gene, as described above using plasmids pPAP(M1) and pPAP(M47). These plasmids do not contain any of the gene upstream of phoA, eliminating the possibility that these plasmids could integrate in the upstream gene by homologous recombination and thus complement a conceivable phenotype of the phoA1 mutant that might be due to truncation of the 3′-non-coding region of the upstream gene, rather than the absence of PHOA. In addition, a frameshift mutation was introduced in the PHOA-encoding region of pPAP(Met47), and this plasmid, pPAPnc, was used as a negative control. The pyroA+ transformants obtained with these plasmids were first grown on low-Pi medium with glucose as the carbon source and scored for pigmentation (Figure 5). Plasmid pPAPnc was not able to complement the phoA1 mutation, and plasmids pPAP(M1) and pPAP(M47) gave similar complementation frequencies of 65 and 61%, respectively. Essentially identical results were obtained when the transformants were retested on low-Pi medium with the alcA-inducing carbon source ethanol. It thus follows that low-level expression of PHOA from the alcA promoter during growth on the 'repressing' carbon source, glucose, is sufficient to complement the phoA1 mutant. This low level of expression from the alcA promoter in the presence of glucose previously has been observed by others (Som and Kolaparthi, 1994). The pyroA+ transformants that were not complemented with respect to the phoA1 mutation are likely to result from plasmid-dependent repair of the pyroA4 mutation or from plasmid integrations in which the phoA gene has been inactivated. The transformants were tested further on MAG medium for the formation of the large clusters of cleistothecia. Plasmids pPAP(M1) and pPAP(M47) both also complemented this sexual differentiation characteristic of the phoA1 mutant at high frequency; ∼95% of the transformants obtained with these plasmids, that did not produce the brown pigment, showed normal sexual differentiation on MAG medium, whereas all strains tested that did produce the brown pigment, including the transformants constructed using plasmid pPAPnc, showed the hypersexual phenotype of the phoA1 mutant (data not shown). To test the nutritional specificity of the phenotype of the phoA1 strain, several other nutrients were made limiting for growth. The colonies of the phoA1 strain and the wild-type strain formed under conditions of carbon, nitrogen and sulfur limitation, respectively, were similar. The response to the individual limitations was, however, quite different. Upon sulfur limitation (Figure 5), for example, unlike the case of phosphorus limitation, strong pigmentation was seen in both the wild-type and phoA1 strains, and both strains also showed strong sexual development. These data demonstrate that there is a mechanism that represses sexual development and pigmentation specifically during phosphorus-limited growth which requires PHOA's function. Interaction between phoA and palcA The positive-acting regulatory palcA gene controls the syntheses of several Pi-repressible phosphatases and probably also a phosphate permease (Caddick et al., 1986). Loss-of-function palcA mutations (palcA1, palcA40) also lead to strong pigmentation (Figure 6), suggesting that the phoA and palcA gene products may function in a common pathway that represses pigmentation. While a strong correlation was observed between the synthesis of the pigment and sexual development, no cleistothecia were observed in the palcA1 strain on low-Pi medium. However, morphologically abnormal hyphae were observed in the later stages of colony development, suggesting that the absence of sexual development might be due to low intracellular phosphorus concentrations caused by the palcA1 mutation. This was tested by constructing a phoA1/palcA1 double mutant strain, in which process the phoA gene was also mapped onto the left arm of chromosome II, linked (8% recombination) to palcA (Figure 7). The double mutant did not show sexual development and, thus, sexual development under these conditions requires the presence of a functional palcA gene. This result also indicates that while pigmentation may be coordinated with sexual development, it is not caused by advanced stages of sexual development. This notion is supported by the observation that pigment production also occurred when palcA1 and phoA1 strains, but not wild-type strains, were grown under specific Pi-limiting conditions in shaken liquid cultures where differentiation was not observed (data not shown). Figure 7.Map position of phoA on the left arm of linkage group II, linked to palcA. The position of the centromere (●) and the distance between palcA and drkB are from Clutterbuck (1993), and other distances are from the cross shown. Distances are expressed as uncorrected recombination frequencies. There appears to be a slight selection against the phoA1 allele (43.9% recovery) which was also observed in a cross between strains PHOΔ17 and HB23 (44.3% recovery of phoA1; 23.6% recombination between wA and phoA; 512 progeny analyzed). Download figure Download PowerPoint The phoA deletion affects development in confluent plate cultures depending on inoculation density as well as phosphate concentration While the phenotype of the phoA deletion strain depends on the Pi concentration of the growth medium, the differences with respect to the yeast PHO85 system raise the question of whether PHOA is involved specifically in transmitting a phosphorus-related signal (see Discussion). The possibility that another factor might play an important role was suggested initially by the fact that sexual development in the phoA deletion strain is particularly strong at the border of a colony (Figure 6), either in the case of a single large colony or where colonies meet each other. In wild-type strains, sexual development occurs typically in the older part of a colony, near the center (Champe et al., 1987). In order