APP1 transcription is regulated by inositol-phosphorylceramide synthase 1-diacylglycerol pathway and is controlled by ATF2 transcription factor in Cryptococcus neoformans.

Inositol-phosphorylceramide synthase 1 (Ipc1) is a fungal-specific enzyme that regulates the level of two bioactive molecules, phytoceramide and diacylglycerol (DAG). In previous studies, we demonstrated that Ipc1 regulates the expression of the antiphagocytic protein 1 (App1), a novel fungal factor involved in pathogenicity of Cryptococcus neoformans. Here, we investigated the molecular mechanism by which Ipc1 regulates App1. To this end, the APP1 promoter was fused to the firefly luciferase gene in the C. neofor-mans GAL7:IPC1 strain, in which the Ipc1 expression can be modulated, and found that the luciferase activity was indeed regulated when Ipc1 was modulated. Next, using the luciferase reporter assay in both C. neoformans wild-type and GAL7:IPC1 strains, we investigated the role of DAG and sphingolipids in the activation of the APP1 promoter and found that treatment with 1,2-dioctanoylglycerol does increase APP1 transcription, whereas treatment with phytosphingosine or ceramides does not. Two putative consensus sequences were found in the APP1 promoter for ATF and AP-2 transcription factors. Mutagenesis analysis of these sequences revealed that they play a key role in the regulation of APP1 transcription: ATF is an activator, whereas AP-2 in a negative regulator. Finally, we identified a putative Atf2 transcription factor, which is required for APP1 transcription and under the control of Ipc1-DAG pathway. These studies provide novel regulatory mechanisms of the sphingolipid pathway involved in the regulation of gene transcription of C. neoformans.

Inositol-phosphorylceramide synthase 1 (Ipc1) 3 is a fungal-specific enzyme of the sphingolipid pathway ( Fig. 1) that regulates the level of phytoceramide and diacylglycerol (DAG), two well established bioactive molecules in mammalian cells, which regulate key cellular functions such as cell growth and viability (1)(2)(3)(4)(5)(6). On the other hand, the role of these lipids in signaling in yeast cells is poorly understood.
Although the presence of sphingolipid enzymes has been demonstrated in Saccharomyces cerevisiae (7), and in pathogenic fungi, such as Aspergillus fumigatus (8), Candida albicans, and Cryptococcus neoformans (9), studies of sphingolipid-mediated signaling transduction in pathogenic fungi are in their infancy. Studies in S. cerevisiae showed that Ipc1 modulates the level of phytoceramide and DAG (10) but whether these lipids regulate signaling in this microorganism has yet to be elucidated. Whereas in mammalian cells DAG is a well known activator of protein kinase C (PKC), DAG does not activate the fungal homolog Pkc1 in S. cerevisiae (11,12) or in C. albicans (13). Thus, if DAG regulates signaling in S. cerevisiae or C. albicans this regulation would be exerted through proteins other than Pkc1.
On the other hand, C. neoformans Pkc1 contains a putative DAGbinding domain, or C1 domain, which is highly homologous to the C1 domain of DAG-dependent mammalian PKCs (14). In recent studies, we showed that Ipc1 activates Pkc1 in C. neoformans through a DAGdependent mechanism (14). This activation is mediated by the C1 domain of Pkc1 and regulates the localization and function of laccase (15), an enzyme that catalyzes melanin, which is required for the pathogenicity of C. neoformans.
Additional studies revealed that Ipc1 also plays a role in the regulation of phagocytosis of C. neoformans, through the modulation of a novel fungal factor called antiphagocytic protein 1 (App1), which inhibits the attachment and ingestion of fungal cells by macrophages (16). Ipc1 controls App1 at the mRNA level, suggesting a transcriptional or posttranscriptional activation. Therefore, we sought to examine the possibility that Ipc1 and the lipids that it modulates, such as phytoceramide or/and DAG, may regulate the transcriptional activation of fungal factor(s), such as App1.
To study the mechanism of the transcriptional regulation of the APP1 gene by Ipc1, the APP1 promoter was fused to the firefly luciferase gene in both C. neoformans wild-type and the GAL7:IPC1 strain, in which Ipc1 can be regulated by growing yeast cells in glucose or galactose. We find that the luciferase activity is modulated by the expression of Ipc1. Treatment with DAG activates luciferase in a dose-and time-dependent manner only when the luciferase gene is under the control of the APP1 promoter, whereas treatment with ceramide/phytoceramide or sphingosine/phytosphingosine does not affect APP1 transcription. We identified two consensus sequences in the APP1 promoter for AP-2 and ATF cis-acting elements. Deletion of the ATF consensus sequence in the APP1 promoter abolishes luciferase activity, whereas mutation of ATF abrogates the DAG-dependent activation. Deletion or mutation of AP-2 causes a significant increase of APP1 transcription, suggesting that this consensus sequence acts as a negative regulator. Finally, we identified and deleted the putative C. neoformans ATF2 gene by homologous recombination and found that loss of Atf2 abrogates luciferase activation driven by the APP1 promoter and regulated by Ipc1 or DAG. Thus, these studies suggest that APP1 transcription is under the control of the Ipc1-DAG pathway through the Atf2 transcription factor and two consensus sequences (AP-2 and ATF) present in the APP1 promoter.

MATERIALS AND METHODS
Strains, Growth Media, and Reagents-C. neoformans var. grubii serotype A strain H99 and derivative mutants used in this study are illustrated in TABLE ONE. The strains were routinely grown in yeast extract/peptone/dextrose (YPD) medium. Yeast extract peptone (YP) supplemented with 20 g/liter glucose or 20 g/liter galactose was used to down-or up-regulate the expression of IPC1 gene, respectively. Nourseothricin (Werner BioAgents, Germany) at a concentration of 100 g/ml was added to YPD plates for selection of the IPC1/APP1: Nuclear Run-on Assay-The nuclear run-on assay was performed according to Hirayoshi and Lis (17). Briefly, C. neoformans IPC1 (WT) and GAL7IIPC1 strains were grown on YP-glucose medium in a shaker incubator for 24 h at 30°C. Cells were washed 3 times in sterile-distilled water (SDW) and then incubated in YNB broth containing 2% glucose or 2% galactose for 24 h at 30°C. Cells were collected at 2800 ϫ g, washed 3 times in 0.5 M NaCl, 50 mM EDTA, suspended in 9.5 ml of SDW containing 0.5 ml of ␤-mercaptoethanol, and incubated at 37°C for 1 h. The cell pellet was then collected at 2800 ϫ g, suspended in 4 ml of spheroplastic solution (1 M sorbitol, 0.1 M Na citrate, pH 5.8, 0.01 M EDTA), and placed on ice for 10 min. Then, 1 ml of spheroplastic solution containing 10 mg of lysing enzyme (Sigma number L-1412) was added, and cells were incubated at 37°C for 1 h. Next, 2.5 ϫ 10 7 cells were harvested at 3000 ϫ g for 6 min at room temperature and washed with TMN buffer (10 mM Tris-HCl, pH 7.4, 100 nM NaCl, 5 mM MgCl 2 ). Cells were suspended and kept on ice for 15 min in Sarkosyl solution (0.95 ml of SDW, 0.05 ml of 10% (w/w) N-lauroylsarcosine (Sigma L-9150)). After centrifugation at 2500 ϫ g for 1 min at 4°C, the cells were suspended in 100 l of reaction buffer (50 mM Tris-HCl, pH 7.9, 100 mM KCl, 5 mM MgCl 2 , 1 mM MnCl 2 , 2 mM dithiothreitol, 0.5 mM ATP, GTP, CTP, 100 Ci of [␣-32 P]UTP, and 1 units of RNasin). The CDase, ceramidase; DH-Cer, dihydroceramide; DH-CDase, dihydroceramidase; Gcs1, glucosylceramide synthase 1; IPC, inositol phosphorylceramide; Ipt1, inositol phosphotransferase 1; Isc1, inositol phosphosphingolipids-phospholipase C1; MIPC, mannose-inositol-P-ceramide; M(IP) 2 C, mannose-(inositol-P) 2 -ceramide; M2IPC, dimannose-inositol-P-ceramide; GalM2IPC, galactose-dimannose-inositol-P-ceramide; IP, inositol phosphate; CP, choline phosphate; PI, phosphatidylinositol; PC, phosphatidylcholine; SMase, sphingomyelinase; SMS, sphingomyelin synthase; SPH, sphingosine. Adapted from Ref. 6. reaction was incubated at 25°C for 8 min and terminated by the addition of 1 l of 1 mg/ml ␣-amanitin and 20 l of 50 mg/ml DNase I. The mixture was incubated for 10 min at 30°C and then an equal volume of stop buffer (ϳ120 l) containing 20 mM Tris-HCl, pH 7.4, 10 mM EDTA, 2% SDS, and 200 g/l proteinase K was added and the mixture incubated at 42°C for 30 min. The labeled mRNA was extracted using phenol/chloroform/isoamyl alcohol (25:24:1) (Sigma number P-3803) and precipitated with ethanol. The labeled mRNA was suspended in 100 l of diethyl pyrocarbonate water and added to the hybridization chamber containing nytran membranes in which actin, IPC1, and APP1 cDNA were previously immobilized using a slot blot apparatus. Before adding the labeled mRNA, the nytran membranes were pre-hybridized for 3 h at 58°C in hybridization solution (10% dextran sulfate, 1% SDS, 50% formamide, 6ϫ standard saline citrate (SSC)). Once the labeled mRNA was added, the nytran membranes were hybridized at 58°C for 16 h. After hybridization, the membranes were washed 3 times at 58°C for 10 min each in 2ϫ SSC and 1% SDS, 3 times at 68°C for 10 min each in 2ϫ SSC and 1% SDS, and 3 times at 68°C for 10 min each in 0.1ϫ SSC and 1% SDS. Membranes were then air-dried and exposed to phosphoscreen at room temperature for 6 days. Each band was quantified by a phosphoimager STORM 840.
Generation of C. neoformans Strains Carrying the Luciferase Gene under the Control of Wild-type or Mutated Forms of the APP1 Promoter-Plasmid pSK/APP1/LUC/NAT1/3ЈUTR was generated as follows: the NAT1 gene under the control of the C. neoformans actin (ACT) promoter was amplified from the pNAT1 vector (kindly provided by Dr. John Perfect, Duke University Medical Center, Durham, NC) using primers XB-NAT-F (5Ј-CTAATCTAGAGCGAGGATGTGAGCTGGAGAGCGG-3Ј) and XB-NAT-R (5Ј-CGCGTCTAGAGAAGAGATGTAGAAACTAGCTT-CC-3Ј), which contain XbaI sites (bold and underlined). The resulting fragment was cloned into pCR2.1-TOPO vector (Invitrogen), generating pCR-NAT1. The LUC gene was amplified using primers Luc3Ј, 5Ј-GATC-TTTCCGCCCTTCTT-3Ј, and Luc5Ј, 5Ј-GCATGCCAGAGATCCTAT-3Ј, and plasmid pGL3 basic vector (Promega) as a template. The resulting fragment was digested with HindIII and BamHI and cloned into the HindIII-and BamHI-digested pCR-NAT1 vector. The resulting plasmid (pCR-NAT1/LUC) was digested with HindIII and EcoRV, yielding a fragment containing LUC:NAT1 that was subcloned into HindIII-and Eco-RV-digested pSK vector (Invitrogen). The resulting plasmid, pSK/LUC/ NAT1, was digested with EcoRV and SacI to insert the EcoRI-blunted and SacI-restricted 3Ј-UTR region of the APP1 gene from the p⌬app1 plasmid (3), generating pSK/LUC/NAT1/3ЈUTR. Next, a 800-bp fragment corresponding to the APP1 promoter was obtained by digesting that p⌬app1 plasmid with XhoI and EcoRI and subcloned into XhoI and SalI-blunted restricted pSKLUC/NAT1/3ЈUTR plasmid creating plasmid pSK/APP1/LUC/NAT1/3Ј/ UTR, which was biolistically transformed into C. neoformans WT and GAL7:IPC1 to generate IPC1/APP1:LUC and GAL7:IPC1/APP1:LUC strains. This 800-bp APP1 sequence was chosen because it was highly predicted to be the promoter of APP1: 1) it represents the 5Ј-UTR sequence immediately upstream of the APP1 mRNA transcribed region (GenBank TM accession number AY965856); 2) it contains the TATA box (Ϫ72 bp from ATG) and two putative consensus sequences for transcription factors AP-2 (Ϫ236 bp from ATG) and ATF (Ϫ139 bp from ATG); and 3) it was highly predicted to be a promoter region when blasted into the BioInformatics & Molecular Analysis Section (BIMAS) at the National Institutes of Health (bimas.dcrt.nih.gov/molbio/proscan/ index.html).
The above plasmids were sequenced before the biolistic transformation to confirm appropriate insertion of the corresponding fragments. Stable resistant transformants were selected for further analysis after five passages onto non-selectable YPD agar medium, and genomic DNA was extracted and subjected to Southern analysis with appropriate probes to identify a double crossover event at the APP1 locus without ectopic integrations (data not shown).
Ipc1 and Luciferase Enzymatic Activities-Ipc1 activity was performed as described previously (20). Luciferase activity was performed according to the Promega protocol described in the Luciferase Reporter Gene Assay. Proteins were extracted according to Luberto et al. (20). Then, 20 l of cell lysate was added to 100 l of luciferase assay reagent (Promega) and the production of luciferase was immediately measured by a Reporter Microplates Luminometer (Turner Designs). Results were normalized per 1 g of proteins.
Lipid Treatments-The C. neoformans strains were grown on the appropriate medium in a shaking incubator for 24 h at 30°C. Cell pellets were washed 2 times with SDW, resuspended in fresh medium, and incubated in the appropriate medium in a shaking incubator for 16 h at 30°C. Next, cell pellets were washed twice with SDW, resuspended in YP medium, and counted. Next, 5 ϫ 10 6 cells/ml were inoculated in 40 ml of appropriate medium containing 0, 5, 10, and 20 M 1,2-dioctanoylglycerol (DiC8) for 2 h and 30 min at 30°C in a shaking incubator. Proteins were then extracted, quantified, and luciferase activity was measured.
Statistical Analysis-Statistical analysis was performed using Student's t test.

RESULTS AND DISCUSSION
Ipc1 Regulates Transcription of APP1-In previous studies, we found that APP1 mRNA levels are regulated by Ipc1 modulation (16). Thus, we investigated whether Ipc1 regulates APP1 gene expression at the transcriptional level. A nuclear run-on assay was used to measure the level of APP1 mRNA upon Ipc1 modulation. Fig. 2 shows that when Ipc1 is up-regulated (GAL7:IPC1 strain grown in galactose) the transcription of IPC1 and APP1 mRNA increases compared with the WT strain. These results suggest that up-regulation of Ipc1 increases APP1 transcription. The nuclear run-on assay was also performed in conditions in which Ipc1 was down-regulated (glucose). We found that, using this method, neither IPC1 nor APP1 mRNA transcripts decrease in the GAL7:IPC1 strain grown on glucose compared with the WT strain (data not shown), suggesting that the sensitivity of the method is not sufficient.
To investigate how the APP1 transcription is regulated by Ipc1, a ϳ800-bp fragment upstream of the ATG of the APP1 gene was fused to the luciferase reporter gene of the firefly Photinus pyralis in both WT and GAL7:IPC1 strains, producing IPC1/APP1:LUC and GAL7:IPC1/ APP1:LUC strains. The IPC1/APP1:LUC and GAL7:IPC1/APP1:LUC strains were grown in glucose and galactose, and luciferase activity was measured. Up-regulation of Ipc1 determined a significant increase of luciferase activity in the GAL7:IPC1/APP1:LUC strain, confirming previous results with the nuclear run-on assay, in which up-regulation of Ipc1 increases APP1 transcription. Importantly, there was no significant difference in the luciferase activity between glucose and galactose cultures of the IPC1/APP1:LUC strain, suggesting that the different carbon source does not affect APP1 transcription (Fig. 3A). Interestingly, under conditions in which Ipc1 is down-regulated (GAL7:IPC1/APP1:LUC grown on glucose), luciferase activity does not decrease below the wildtype level. These results suggest that when Ipc1 is down-regulated, potential compensatory mechanism(s) may exist leading to an increase of DAG level by pathways other than that regulated by Ipc1. This hypothesis is supported by evidence that a decrease of DAG under conditions of Ipc1 down-regulation is transitory and occurring in the very early log-phase of growth (14). It is also possible that because of the leakiness of the GAL7 promoter, the effect of GAL7 down-regulation cannot be measured using these assays at a given time point. On the other hand, in previous studies we showed that APP1 mRNA levels analyzed by reverse transcriptase-PCR are significantly decreased when Ipc1 is down-regulated (16). Thus, it is possible that under conditions in which Ipc1 is down-regulated, the APP1 mRNA level decreases as a result of an increased degradation of APP1 mRNA. Clearly, this potential regulatory mechanism awaits further characterization.
Thus, to better address whether APP1 transcription can be turned "on," "off," and on again by Ipc1, the GAL7:IPC1/APP1:LUC strain was first grown on galactose, then switched to glucose, and finally placed back into galactose medium. Under these conditions, both Ipc1 and  luciferase activities were measured. As shown in Fig. 3, B and C, downregulation of Ipc1 decreases luciferase activity, whereas switching of the cells to Ipc1-inducing conditions (galactose) induces an increase of luciferase activity. These results suggest that modulation of Ipc1 regulates APP1 transcription.
As a positive control, we fused the actin promoter to the luciferase gene into the C. neoformans H99 wild-type strain, creating a IPC1/ACT: LUC strain. As a negative control, we used a ϳ876-bp long nucleotide sequence (5Ј-UTR) upstream of the APP1 promoter and fused it to the luciferase gene in the GAL7:IPC1 strain, creating a GAL7:IPC1/5ЈUTR: LUC strain (see "Materials and Methods" for details). Strains were grown in glucose or galactose, and luciferase activity was measured. As expected, the GAL7:IPC1/5ЈUTR:LUC strain did not show any luciferase activity, under conditions in which Ipc1 was up-or down-regulated (Fig. 4). The IPC1/ACT:LUC strain showed a high level of luciferase activity that was not modulated by the different carbon sources (Fig. 4). Importantly, luciferase activity was detected at a high level, as expected, in IPC1/ACT:LUC. Taken together, these results suggest that the nucleotide sequence corresponding to the APP1 promoter activates luciferase activity and that APP1 transcription is under the control of Ipc1.
DAG Activates APP1 Transcription-Because in previous studies we showed that C. neoformans Ipc1 regulates the level of phytoceramide and DAG (14), we next wondered whether treatment with DAG or phytoceramide would affect luciferase activity. Thus, the IPC1/APP1: LUC, GAL7:IPC1/5ЈUTR:LUC, and IPC1/ACT:LUC strains were treated with different concentrations of DAG, and luciferase activity was measured. We found a dose-and time-dependent increase of luciferase activity in the IPC1/APP1:LUC strain treated with DAG (Fig. 5, A  and B). Importantly, treatment with DAG did not increase luciferase activity in the IPC1/ACT:LUC strain (Fig. 5C), and did not activate luciferase in the GAL7:IPC1/5ЈUTR:LUC strain (Fig. 5D). Because Ipc1 also regulates the level of phytoceramide (14), we wondered whether luciferase activity would be modulated by treatment with ceramides or phytosphingosine. We found no effect on luciferase activity when cells were exposed to different concentrations of C2-, C6-ceramide, C6-phytoceramide, or phytosphingosine (data not shown). These results suggest that DAG may be the effector of the Ipc1 regulation over the APP1 promoter.

APP1 Transcription Is Regulated by ATF and AP-2 cis-Acting
Elements-To identify potential sites of regulation of the APP1 promoter, a blast search for consensus sequences for transcription factors was performed (bimas.dcrt.nih.gov/molbio/proscan/index.html). The search revealed the presence of two putative consensus sequences normally recognized by transcription factors AP-2 and ATF, located at Ϫ236 bp (AP-2) and Ϫ139 bp (ATF) from ATG, respectively. In mammalian cells, it has been suggested that DAG may exert a regulatory effect on the transcription factor(s) ATF/cAMP-responsive element (21)(22)(23). Also, it has been proposed that AP-2 mediates transcriptional activation either in response to DAG and the DAG-PKC pathway or in response to the cAMP-PKA pathway (24). According to our previous findings in which Ipc1 activates C. neoformans Pkc1 through DAG in vitro (14) and in vivo (15), it was particularly intriguing to hypothesize that Ipc1-DAG would regulate APP1 transcription through ATF or/and AP-2.   To address this hypothesis, we focused on a stretch of 366 bp of the APP1 promoter containing the two cis-acting elements (AP-2 and ATF). The 366-bp fragment was fused to the LUC gene and it was either integrated into the APP1 locus in the WT strain yielding the IPC1/366: LUC strain (Fig. 6), or transformed episomally in the GAL7:IPC1/ 5ЈUTR:LUC strain using the pTel/366:LUC plasmid (Fig. 6) yielding the Tel/366:LUC strain. Fig. 7A shows that the 366-bp fragment is sufficient to activate luciferase activity in both strains in which the 366:LUC fragment is integrated chromosomally or episomally (Fig. 7A), and luciferase activity controlled by the 366-bp fragment is not significantly different from that one controlled by the ϳ800-bp APP1 promoter (Fig. 3A). Importantly, up-regulation of Ipc1 significantly increased luciferase activity when the LUC gene is driven episomally by the 366-bp fragment (Fig. 7A), similar to what was observed in the GAL7:IPC1/APP1:LUC strain (Fig. 3A). Next, the IPC1/366:LUC and Tel/366:LUC strains were treated with DAG, and luciferase activity was measured. As shown in Fig. 7, B and C, following DAG treatment, luciferase activity increased in a dose-dependent manner. These results clearly suggest that the 366-bp fragment is sufficient to drive luciferase activation by Ipc1 and DAG in strains in which the 366-bp fragment is either integrated or episomal.
Next, we determined the effect of the deletion or mutations of AP-2 and ATF consensus sequences on luciferase activity. Thus, plasmid constructs carrying AP-2 or ATF or AP-2/ATF deletions or mutations were transformed episomally in the GAL7:IPC1/5ЈUTR:LUC strain (Fig. 8), as indicated. Stable transformants were screened by Southern analysis of undigested genomic DNA and transformants carrying only the episomal plasmid(s) were selected (data not shown). In the deletion mutant strains, the AP-2 or ATF consensus sequence were removed from the APP1 promoter, whereas in the mutated strains two nucleotides in the ATF or AP-2 consensus sequence were mutated (see "Materials and Methods" for details). As shown in the figure, deletion of ATF abolished luciferase activity, whereas deletion of AP-2 dramatically increased luciferase activity by ϳ20-fold (Fig. 8A). When the ATF consensus sequence was mutated (from TGACGTCA into TGAAATCA), the  basal luciferase activity was not altered (Fig. 8A), although DAG activation was completely abrogated (Fig. 8B). Treatment with DAG did not increase luciferase activity in either the AP-2-deleted or mutated strains (Fig. 8B), suggesting that DAG might have a negative effect on AP-2 activity.
These results might suggest that the two CG nucleotides in the ATF consensus sequence are responsible for the DAG-mediated activation of APP1 transcription. This hypothesis is supported by studies in mammalian cells, in which mutation of these two nucleotides in the ATF cis-acting element abrogates the activation of the ecto-5Ј-nucleotidase promoter by phorbol 12-myristate 13-acetate (25). Thus, our findings support other reports that describe the ATF cis-acting site as a phorbol 12-myristate 13-acetate/DAG responsive element (22,23,25).
Interestingly, because activation by DAG was also abolished when the AP-2 consensus sequence was either deleted or mutated, it would suggest that DAG might also have an effect on AP-2. Thus, we hypothesize that ATF is involved in the constitutive expression of APP1 and that DAG positively controls APP1 expression through ATF and negatively through AP-2. It is possible, however, that because luciferase activity is greatly increased upon AP-2 deletion or mutation, the addition of DAG cannot further activate luciferase or that DAG inhibits AP-2 activity that is lost when AP-2 is deleted or mutated. Further investigations are required to determine whether DAG and Ipc1 modulation would affect the binding of transcription factor(s) to ATF or/and AP-2 cis-acting elements of the C. neoformans APP1 promoter.
Role of ATF2 Gene in APP1 Activation-Once the putative C. neoformans ATF2 gene ORF was identified (see "Materials and Methods" for details), the putative C. neoformans ATF2 gene was deleted in both IPC1/APP1:LUC and GAL7:IPC1/APP1:LUC strains, creating IPC1/ APP1:LUC/⌬atf2 and GAL7:IPC1/APP1:LUC/⌬atf2 (Fig. 9, A and B). Strains were grown in glucose or galactose or treated with DAG, and luciferase activity was measured. We found that both strains show a dramatic reduction of luciferase activity, which is no longer under the control of Ipc1 modulation (Fig. 9, C and D) or DAG treatment (Fig. 9, E and F). Taken together, these results suggest that the C. neoformans Atf2 transcription factor may be activated by the Ipc1-DAG pathway to increase the transcription of APP1 through ATF and/or AP-2 cis-acting elements of the APP1 promoter.
The molecular interaction between Atf2 and the APP1 promoter has yet to be elucidated. It is tempting to speculate that because an AP-2 homolog was not found in the C. neoformans genome data base the Atf2 transcription factor may bind to both ATF and AP-2 cisacting elements and that the presence of DAG (perhaps promoting Atf2 phosphorylation) may favor the release of the binding to AP-2 while maintaining the binding to ATF. As a result, APP1 transcription increases. On the other hand, it is possible that DAG negatively regulates a transcription factor yet to be identified that binds to the putative AP-2 consensus sequence. The model proposed in Fig. 10 takes into consideration this possibility. Clearly, this model awaits further biochemical characterization.
The question on the potential mechanism by which Ipc1-DAG activates the transcription factor(s) involved in the regulation of APP1 transcription is also intriguing. In mammalian cells, although Atf2 (also known as CRE-BP-1) can be phosphorylated by PKC (26), ATF tran-scription factors, are mainly activated by the cAMP-PKA pathway (reviewed in Ref. 27). On the other hand, in the fission yeast Schizosaccharomyces pombe, the Atf1 transcription factor is homologous to the mammalian Atf2 and is positively regulated by stress-activated mitogen-activated protein kinase pathway but not by the cAMP-PKA pathway (28 -30). When the S. pombe Atf1 protein (GenBank BAA09841) was blasted into the C. neoformans H99 data base, the same (and only) sequence identified by blasting the human Atf2 (Chr2-piece9) was found with an E value of 3e-22. This suggests that the putative C. neoformans ATF2 gene that we identified is closely related to the human Atf2 and S. pombe Atf1 transcription factors. Importantly, the role and mechanisms by which ATF transcription factor(s) is regulated in human pathogenic fungi, such as C. albicans, C. neoformans, and A. fumigatus, are basically unexplored.
With these studies, we propose that C. neoformans Atf2 is under the control of the Ipc1-DAG pathway. This control may be exerted by Pkc1 because C. neoformans Pkc1 is also regulated by the Ipc1-DAG pathway (14,15). Importantly, preliminary studies showed that treatment with calphostin C, which, in addition to the mammalian PKC, also inhibits C. neoformans Pkc1 (14), abolishes the activation of luciferase by DAG in the IPC1/APP1:LUC strain, 4 supporting the hypothesis that DAG may activate Atf2 through Pkc1. Considering that Ipc1 controls Pkc1 in C. neoformans (15) and that, in mammalian cells, Atf2 phosphorylation is required for its activation (31,32), the hypothesis that Ipc1-DAG-Pkc1 may activate APP1 transcription through the Atf2 transcription factor and ATF and/or AP-2 cis-acting elements is particularly attractive.
Conclusions-It is anticipated that the study of regulation of gene transcription(s) that favors C. neoformans to escape the host response and cause disease is a future area of investigation given the recent completion of the genome sequence of this organism (33) and the construction of C. neoformans microarrays (genome.wustl.edu/ projects/cneoformans/microarray/). With this study, we developed the use of luciferase as a reporter for gene expression as an important additional tool for studying gene transcription of C. neoformans in vitro. By using this approach, we dissected the Ipc1-App1 pathway and we propose that the biochemical mechanism by which Ipc1 regulates APP1 transcription is through the production of DAG and the potential activation of Atf2 (Fig. 10). The presence of stimulatory (ATF) and putative inhibitory (AP-2) factors in the regulation of APP1 transcription would suggest that the gene may be activated only under certain circumstances, perhaps during different stages of the cryptococcal infection. In previous studies, App1 protein was found in sera of AIDS patients affected by cryptococcosis and, given its characteristic nature of being anti-phagocytic, C. neoformans might regulate the production of App1 protein to escape internalization by phagocytic cells and down-regulate it once yeast cells would be internalized. This intriguing regulation would support the hypothesis that, to survive, pathogens would need to show adaptation and plasticity during their interaction with the host. In being both an extra-and intracellular microorganism, C. neoformans may have developed unique features, such as the production of App1 protein (or other antiphagocytic factors such as the polysaccharide capsule), which can be positively and negatively controlled and, thus, allow the fungus to counteract the immune response depending on the immune status of the host. How these microbial features are regulated and how they promote fungal survival within the host is a very attractive area of investigation.