Aspergillus fumigatus strain akuB^KU80 and the isogenic pyrG auxotrophic strain akuB^KU80 pyrG^- were used in all experiments. A. fumigatus cultures were grown on glucose minimal medium (GMM) at 37^°C. Escherichia coli DH5α competent cells were used for subcloning. A. fumigatus was transformed as described earlier (Steinbach et al., 2006). For radial growth analyses, 10⁴ conidia of the akuB^KU80 and the other strains were cultured on GMM agar at 37^°C, with colony diameters determined every 24 h over a period of 5 days. All growth experiments were performed in triplicate.

While a ΔcbpA strain was previously reported by our group (Pinchai et al., 2009), we generated a second deletion strain in the akuB^KU80 background for this study in order to maintain the same genetic background for all mutated strains. Deletion of cbpA was performed using a previously designed construct in the plasmid pJW24 with the Aspergillus parasiticus pyrG gene as the auxotrophic marker (Steinbach et al., 2006). The cbpA deletion construct contained 1 kb promoter and 780 bp terminator flanking regions to facilitate homologous recombination. The ∼5 kb cbpA deletion construct was PCR-amplified using the primers cbpA-promo-KpnI-F and cbpA-term-HindIII-R (Supplementary Table S1) and transformed into the A. fumigatus akuB^KU80 pyrG^- strain. Transformants selected in the absence of uracil/uridine (Steinbach et al., 2006) were screened by PCR (Supplementary Table S1) and Southern analyses to confirm deletion of the cbpA gene (Supplementary Figure S1).

In order to purify the CbpA protein for phosphorylation analyses, the cbpA-tagged egfp expression construct at the cbpA native locus was generated by a fusion PCR strategy. First, the 864 bp cbpA genomic DNA without the stop-codon and the 722 bp egfp gene were PCR-amplified separately. Next, the two PCR fragments were combined and used as templates to amplify the final fusion PCR product containing the cbpA-egfp fusion product, which was then digested with KpnI and NotI to clone at the KpnI-NotI sites in pUCGH (gift from Axel Brakhage) with the hygromycin B resistant gene (hph) as the selection marker. The cloned cbpA genomic DNA was sequenced to ensure accuracy (Primers listed in Supplementary Table S1). To facilitate homologous recombination, the 780 bp cbpA terminator region was cloned into the SbfI-HindIII sites in pUCGH (Langfelder et al., 2001). The final plasmid obtained was designated as pUCGH-CbpA, and the entire construct was linearized by KpnI-HindIII digestion to obtain a fragment of ∼6.7 kb. The linearized construct was then transformed into the A. fumigatus akuB^KU80 strain as previously described and transformants selected by resistance to hygromycin B. The transformants obtained were verified for homologous integration by PCR and also by fluorescent microscopy to visualize GFP fluorescence.

Site-directed mutation of phosphorylated residues in CbpA was performed using primers listed in Supplementary Table S1 and pUCGH-cbpA as a template. Briefly, the first PCR used complementary primers overlapping the regions to be mutated and the respective primers at the opposite ends to amplify two PCR fragments. Next, a fusion PCR was performed by using an equiproportional mixture of the two PCR fragments as a template and amplifying the mutated PCR fragment using primers at the opposite ends (Supplementary Table S1). Each fragment cloned into pUCGH was sequenced to confirm the respective mutation and then linearized for homologous integration. The linearized constructs were transformed into the A. fumigatus akuB^KU80 strain as previously described and transformants selected by resistance to hygromycin B. The transformants obtained were verified for homologous integration by PCR and also by fluorescent microscopy to visualize GFP fluorescence.

The A. fumigatus strain expressing the cbpA-egfp fusion construct was grown in GMM liquid medium as shaking cultures for 24 h at 37^°C. Cell extracts were prepared by homogenizing the mycelia using liquid nitrogen in buffer A (50 mM Tris/HCl, pH 7.5, 150 mM NaCl, 50 mM KCl, 0.01% Triton X-100, 1 mM PMSF, and 1:100 protease inhibitor cocktail). Total cell lysate was initially centrifuged at 5000 rpm to eliminate cell debris, and clarified by further centrifugation at 7000 rpm for 10 min at 4^°C. The final supernatant fraction was collected and protein content determined by Bradford assay. Total protein was normalized to contain ∼10 mg protein in the sample before GFP-Trap affinity purification (Chromotek). GFP-Trap^® resin (35 μl) was equilibrated by washing three times in 500 μl ice-cold dilution buffer (10 mM Tris-HCl pH 7.5, 150 mM NaCl, 0.5 mM EDTA, 1 mM PMSF, 1:100 Protease Inhibitory Cocktail) according to the manufacturer instructions and resuspended in 100 μl ice cold dilution buffer. The GFP-Trap^® resin suspension was then mixed with total crude cell lysate containing ∼10 mg total protein and incubated at 4^°C by gentle agitation for 2 h. The suspension was centrifuged at 2000 rpm for 10 min at 4^°C and the pelleted GFP-Trap^® resin was washed once in 500 μl of ice-cold dilution buffer and then twice with 500 μl of wash buffer (10 mM Tris-HCl pH 7.5, 350 mM NaCl, 0.5 mM EDTA, 1 mM PMSF, 1:100 Protease Inhibitory Cocktail).

Protein-bound GFP-Trap^® resins were washed three times with 200 μl of 50 mM ammonium bicarbonate, pH 8.0, and suspended in 30 μl 50 mM ammonium bicarbonate, pH 8.0, supplemented with 0.1% Rapigest SF surfactant (Waters Corp). Samples were reduced with 5 mM dithiothreitol for 30 min at 70^°C and free sulfhydryls were alkylated with 10 mM iodoacetamide for 45 min at room temperature. Proteolytic digestion was accomplished by the addition of 500 ng sequencing grade trypsin (Promega) directly to the resin, with incubation at 37^°C for 18 h. Supernatants were collected following a 2 min centrifugation at 1000 rpm, acidified to pH 2.5 with TFA, and incubated at 60^°C for 1 h to hydrolyze the remaining Rapigest surfactant. Insoluble hydrolyzed surfactant was cleared by centrifugation at 15,000 rpm for 5 min. Ninety percent (by volume) of the sample was then removed for subsequent phosphopeptide analysis, and the remaining 10% (by volume) was subjected to an unbiased protein interaction analysis.

For the phosphopeptide analysis, samples were lyophilized to dryness using vacuum centrifugation and resuspended in 65 μl 80% acetonitrile, 1% TFA. Peptides were subjected to phosphopeptide enrichment using a 10 μl GL Sciences TiO2 Spin Tip and subsequently washed with 80% acetonitrile, 1% TFA. Peptides were eluted in 50 μl 20% acetonitrile, 5% aqueous ammonia, pH 10.5, and then acidified to pH 2.5 with formic acid prior to lyophilization to dryness.

Samples were resuspended in 10 μl 2% acetonitrile, 10 mM citric acid 0.1% formic acid and subjected to chromatographic separation on a Waters NanoAquity UPLC equipped with a 1.7 μm HSS T3 C18 75 μm I.D. x250 mm reversed-phase column. Phosphopeptide enriched samples were additionally supplemented with 10 mM citric acid. The mobile phase consisted of (A) 0.1% formic acid in water and (B) 0.1% formic acid in acetonitrile. Following a 5 μl injection, peptides were trapped for 5 min on a 5 μm Symmetry C18 180 lm I.D. x20 mm column at 20 μl/min in 99.9% A.

The analytical column was connected to a fused silica PicoTip emitter (New Objective, Cambridge, MA, USA) with a 10 μm tip orifice. Non-phosphopeptide enriched samples were analyzed on a Synapt G2 QToF mass spectrometer operating in a data-dependent mode of acquisition with a precursor MS scan from m/z 400–2000 followed by three MS/MS scans at a CID energy of 30% and a dynamic exclusion of 30 s. Phosphopeptide enriched samples were analyzed on a QExactive Plus mass spectrometer with a precursor MS scan from m/z 300–1600 with r = 70,000 at m/z 200 and a target AGC setting of 1e6 ions. In a data-dependent mode of acquisition, MS/MS spectra of the 10 most abundant precursor ions were with a CID energy setting of 27 and a dynamic exclusion of 20 s was employed for previously fragmented precursor ions.

Raw LC-MS/MS data files were processed in Mascot distiller (Matrix Science) and then submitted to independent Mascot searches (Matrix Science) against a custom NCBI_Aspergillus database containing both forward and reverse entries of each protein. Search tolerances were 5 or 10 ppm for precursor ions and 0.02 or 0.04 Da for product ions using trypsin specificity with up to two missed cleavages for phosphopeptide enriched or non-phosphopeptide enriched data, respectively. Carbamidomethylation (+57.0214 Da on C) was set as a fixed modification, whereas oxidation (+15.9949 Da on M), deamidation (+0.98 Da on NQ), and phosphorylation (+79.98 Da on STY) were allowed. All searched spectra were imported into Scaffold (v4.0, Proteome Software) and scoring thresholds were set to achieve a protein false discovery rate of 0% using the PeptideProphet algorithm. Normalized spectral counts were used to estimate relative protein abundances for interaction studies. This was accomplished by adjusting the sum of the selected quantitative value for all proteins in the list within each MS sample to the average of the sums of all MS samples present in the experiment.

As a first step toward identifying the phosphorylation status of CbpA in vivo we initially utilized the A. fumigatus strain expressing cnaA-egfp fusion construct considering the fact that CbpA would co-purify with CnaA. As shown in Supplementary Table S2, purification of CnaA by GFP-Trap^® affinity purification and phosphopeptide-enrichment (see Supplemental Materials and Methods) led to the identification of six phosphorylated residues in CbpA (Ser156, Ser160, Thr205, Ser208, Ser217, Ser223). Next, in order to validate this finding and more clearly verify the phosphorylation status of A. fumigatus CbpA in vivo, a strain expressing the cbpA–egfp fusion construct under the control of its native promoter was generated in the akuB^KU80 strain. The CbpA–EGFP fusion protein was purified by GFP-Trap^® affinity purification and subjected to protease digestion, followed by phosphopeptide-enrichment and tandem mass spectrometry. As shown in Table 1, a total of nine residues (Ser8, Thr133, Ser156, Ser160, Ser208, Ser217, Ser223, Ser257, and Ser266) were identified to be phosphorylated in A. fumigatus CbpA.

Based on the location of the majority of the phosphorylated residues, they were classified into three different domains: phosphorylated domain I (PD-I), phosphorylated domain II (PD-II), and phosphorylated domain III (PD-III; Figure 1). While the phosphorylation of the FxISPPxSPP motif in the PD-I (Ser156, Ser160; Supplementary Figure S2) is well-known from mammalian and the yeast RCANs (Vega et al., 2002; Hilioti et al., 2004; Abbasi et al., 2006), the identification of three residues phosphorylated in PD-II (Ser208, Ser217, Ser223) seems to be specific to A. fumigatus as this region was not conserved in the C. neoformans Cbp1 or in the Human MCIP1 (Figure 1). In addition, phosphorylation of two serine residues (Ser257 and Ser266) in PD-III, previously designated as the CIC motif (Calcipressin Inhibitor of Calcineurin), was also noted. The calcineurin-binding PxIxIT sequence (PSVVVH) together with the ELHA sequence comprises the CIC motif and is conserved in all vertebrate RCANs/MCIPs (underlined in Figure 1; Aubareda et al., 2006; Mulero et al., 2007). Both the ELHA and PxIxIT sequences are required for binding to calcineurin with high affinity (Martínez-Martínez et al., 2009; Mulero et al., 2009). As shown in Supplementary Figure S3, the two serine residues (Ser156 and Ser160) in the FxISPPxSPP motif (PD-I) are conserved in all the filamentous fungi. The fungal orthologs, however, show less conservation to mammalian RCAN in the ELHA and PxIxIT domains (PD-III; Supplementary Figure S3). However, it is interesting to note that phosphorylation of a conserved serine residue was observed within the CIC (PD-III) motif of MCIP1 (Figure 1; serine residue shown in white color) and the protein kinase CK2 was identified as the enzyme responsible for the phosphorylation to potentiate inhibition of NFATc signaling by disrupting the calcineurin-NFATc interaction (Martínez-Høyer et al., 2013). Although we do not have conservation of the ELHA or the PSVVVH residues in the A. fumigatus CbpA CIC motif (PD-III), we do have the predicted CK2 target sequence (S/TxxE/D) within the PD-III.

While phosphorylation at the CIC motif (PD-III) was also suggested to be independent of the phosphorylation occurring at the FxISPPxSPP motif (PD-I) of the RCANs, the non-phosphorylated CIC (PD-III) caused increased NFATc (Martínez-Høyer et al., 2013). The absence of PD-II domain in the mammalian MCIP1, the non-conservation of the PD-III domain in the fungal orthologs (Supplementary Figure S3), and our recovery of PD-III phosphorylated CbpA from the wild-type strain all suggest possible differential regulation of the fungal Cbps through phosphorylation-dephosphorylation. In the future, it will be interesting to examine the effect of mutation of these residues on CbpA toward the CnaA–CrzA (ortholog of NFAT) interaction and activation.

Previous studies have suggested the dephosphorylation of RCANs by calcineurin (Fox and Heitman, 2005). In the mammalian RCANs, phosphorylation of the two serine residues within the FxISPPxSPP motif (PD-I; Ser108 and Ser112) occurs first at Ser112 through MAPK, which then triggers subsequent phosphorylation at Ser108 by GSK3β (Vega et al., 2002). The phosphorylated Ser108 can then be dephosphorylated by calcineurin. It was also previously shown that alterations in the FxISPPxSPP motif (PD-I) phosphorylation influence protein stability, with mutation of the FxISPPxSPP leading to a more stable protein (Genesca et al., 2003). Genetic evidence from S. cerevisiae suggests that phosphorylation of Ser108 by GSK3 can release the inhibitory activity of RCAN and convert the protein to an activator of calcineurin, although the exact mechanism is unknown (Hilioti et al., 2004). Also it is unknown whether the phosphorylation at the CIC motif (PD-III) is subject to regulation by calcineurin protein phosphatase activity.

Considering these data, and the rationale that absence of calcineurin may inhibit dephosphorylation of the residues that undergo calcineurin-mediated dephosphorylation, we next determined if any of the phosphorylated residues are targets for calcineurin by deleting cnaA and monitoring the phosphorylation status of CbpA. The FxISPPxSPP residues (PD-I; Ser156, Ser160; Figure 2) and Ser217 and Ser223 in the PD-II domain remained phosphorylated, in addition to the single Thr133 residue (Table 1). In addition to the N-terminal Ser8 residue, the Ser208 in the PD-II domain, and the residues in the CIC motif (PD-III; Ser257, Ser266) were not phosphorylated in the cnaA deletion background, indicating the possibility of calcineurin involvement in the phosphorylation-dephosphorylation mechanism of CbpA. Specifically, the residues in the FxISPPxSPP motif (PD-I) and the PD-II domain may be amenable for dephosphorylation by calcineurin.

Phosphorylation of the FxISPPxSPP motif (PD-I) at both serine residues in C. neoformans Cbp1 was shown to be necessary for its stability, activity, and regulation of hyphal elongation during mating (Fox and Heitman, 2005). Interestingly, the phosphorylation of the FxISPPxSPP motif (PD-I) did not mediate the binding of calcineurin to Cbp1 (Fox and Heitman, 2005). In order to investigate the importance of phosphorylation of A. fumigatus CbpA at the FxISPPxSPP motif (PD-I), we next mutated the two phosphorylated residues within the FxISPPxSPP motif to alanine and aspartic acid residues, respectively, to mimic a non-phosphorylated state (cbpA^mt-PD-I-2SA) and a constitutively phosphorylated state (cbpA^mt-PD-I-2SD). Radial growth analyses of the strains under normal growth conditions over a period of 5 days indicated a slower growth rate in the cbpA^mt-PD-I-2SA strain compared to the cbpA^mt-PD-I-2SD strain (Figure 3). Unexpectedly, the ΔcbpA strain displayed a slightly better growth rate than the cbpA^mt-PD-I-2SA strain. Based on our previous report (Pinchai et al., 2009), it is possible that the complete deletion of cbpA induces higher expression of calcineurin or inappropriate activation of calcineurin that can negatively affect growth. Although the cbpA^mt-PD-I-2SD strain grew better than the cbpA^mt-PD-I-2SA strain, its growth was comparable to the ΔcbpA strain, indicating that both dephosphorylation and continued phosphorylation of CbpA at the two serine residues in the FxISPPxSPP motif (PD-I) affected its activity. A similar observation was made with C. neoformans Cbp1-phospho mimetic expression, wherein mutation of both the serines to glutamic acid abolished filamentation during mating, a phenotype also observed with complete deletion of Cbp1 (Görlach et al., 2000). Importantly, in contrast to the FxISPPxSPP motif mutations (PD-I), the mutation of the three phosphorylated residues in the PD-II domain did not have any effect on the growth under normal conditions (Figure 1).

In mammalian cells, increased levels of RCAN1 were found to be protective against oxidative stress and suppress cell growth. It is not known whether these features of RCANs are a direct consequence of calcineurin inhibition or result from a yet unidentified action of RCAN (Leahy and Crawford, 2000). Furthermore, phosphorylation of the mammalian RCANs at the FxISPPxSPP motif (PD-I) was correlated with an increase in calcineurin-mediated transcriptional inhibition (Genesca et al., 2003). Hence, in order to determine the role of phosphorylated CbpA in counteracting oxidative stress, the respective cbpA mutant strains were assayed for growth in the presence of the oxidative stress inducer paraquat (Figure 4). Based on phenotypes observed under normal growth conditions, phosphorylation of CbpA at the FxISPPxSPP motif (PD-I) is important for proper hyphal growth, but the phosphorylation at the PD-II domain seemed unimportant for growth. Furthermore, both the cbpA^mt-PD-I-2SA and the cbpA^mt-PD-I-2SD strains were sensitive to oxidative stress induced by paraquat (Figure 4) and H₂O₂ (data not shown), indicating the importance of phosphorylation-dephosphorylation of CbpA for counteracting oxidative stress conditions. The possibility of other phosphatases regulating the dephosphorylation of CbpA cannot be precluded.

It was previously demonstrated that binding of mammalian RCANs to the calcineurin catalytic subunit (CnA) does not interfere with binding of either calmodulin (CaM) or the calcineurin regulatory subunit (CnB) to CnA (Vega et al., 2002). In fact, most data indicate that RCANs bind preferentially to active calcineurin, suggesting that CaM binding facilitates RCAN binding (Martínez-Martínez et al., 2009). However, in C. neoformans it was shown that the regulatory subunit of calcineurin is required for the binding of CBP1 to drugs that inhibit calcineurin activity (cyclosporin A or FK506) and also disrupt the CBP1/CNA interaction (Görlach et al., 2000), suggesting that the site of CBP1 interaction with CNA overlaps with that of immunophilins.

We previously generated A. fumigatus strains harboring mutations in key domains of CnaA (the PxIxIT-binding motif; CnaB binding helix; CaM-binding domain) to understand their significance for calcineurin localization and function (Juvvadi et al., 2013). All the respective cnaA mutated constructs were tagged to egfp and verified for stable expression (Juvvadi et al., 2013). Because there is no predictable PxIxIT motif in the A. fumigatus CbpA, we utilized the A. fumigatus strain expressing CnaA-NIR^mt-AAA in which the PxIxIT motif-binding residues (352-NIR-354; Asn352 Ile353 Arg354) are mutated to alanines in CnaA to determine if the binding of CbpA to CnaA requires the PxIxIT motif. As shown in Table 2, we found that mutation of the PxIxIT motif-binding NIR residues in CnaA completely abolished the binding of CbpA to CnaA. The binding of CbpA to CnaA was also found to be dependent on CnaB binding to CnaA. This was demonstrated by utilizing the A. fumigatus mutant strains expressing CnaA^mt-V371D and CnaA-THL^mt-PLS constructs wherein the key CnaB-binding residue Val371 on CnaA is mutated to aspartic acid (V371D) and three other residues (THL^mt-PLS; Thr359 His361 Leu365 mutated to Pro359 Leu361 Ser361, respectively) located close to the CnaB binding helix completely inhibited the binding of CbpA to CnaA. Moreover, treatment with FK506 also inhibited the binding of CnaA to CbpA. Contrary to these findings, mutation of the CaM binding domain residues RVF (442-RVF-444; Arg442 Val443 Phe444 to alanines) on CnaA did not affect the binding of CbpA to CnaA. The strain expressing wild-type CnaA-EGFP was used as a comparative control to determine the CbpA peptides bound to CnaA. The binding of CbpA peptides to CnaA were validated in three independent experiments by mass spectrometry (Table 2 and Supplementary Table S3). Taken together, these results indicated that CbpA binds to CnaA through a PxIxIT motif and requires the binding of CnaB to CnaA. Although we could not identify the PxIxIT motif in CbpA, it is possible that a PxIxIT-like sequence (PVDPGT) close to the newly identified phosphorylated domain II (PD II) is responsible for binding to CnaA. Future crystallization studies on the calcineurin-CbpA complex would shed light on the role of each of these domains and the functional motifs in the binding of CbpA to calcineurin. Furthermore, detailed knowledge of the molecular mechanisms governing the CbpA-calcineurin interaction may also prove useful in the rational design of future immunosuppressant drugs.

A recent study using human cell lines identified the phosphorylation of a serine residue at the CIC motif in all the RCANs (RCAN1, RCAN2, and RCAN3; Martínez-Høyer et al., 2013). Through in vitro and in vivo phosphorylation assays, CK2 was identified as the kinase phosphorylating this serine residue and this phosphorylation enhanced the disruptive potential toward the calcineurin-NFATc interaction. Interestingly, in the present study, the serine (Ser257) along with another serine residue at position 266 was identified to be phosphorylated the region spanning the CIC motif (PD-III) in A. fumigatus CbpA. In the future, it will be interesting to examine the effect of mutation of these residues on CbpA toward the CnaA-CrzA interaction and activation. Taken together, our results indicate that phosphorylation of CbpA at the FxISPPxSPP motif (SP repeat motif) is conserved in filamentous fungi and this phosphorylation is important for CbpA function in vivo. In addition, the binding of CbpA to CnaA occurs via a PxIxIT motif and the CnaB binding helix is essential for the interaction.