Identification and functional characterization of the putative members of the CTDK-1 kinase complex as regulators of growth and development in Aspergillus nidulans and Aspergillus fumigatus

| Mycology | Research Article Identification and functional characterization of the putative members of the CTDK-1 kinase complex as regulators of growth and development in Aspergillus nidulans and Aspergillus fumigatus Z. Agirrezabala,1 X. Guruceaga,2 A. Martin-Vicente,2 A. Otamendi,1 A. Fagoaga,1 J. R. Fortwendel,2 E. A. Espeso,3 O. Etxebeste1 AUTHOR AFFILIATIONS See affiliation list on p. 26. ABSTRACT Asexual spores are the main vehicle used by fungi to disperse to new niches. The Eurotiomycete Aspergillus nidulans is the main reference for the study of the genetic/molecular control of asexual development. In this species, Flb proteins control the expression of the master gene brlA, and thus, loss-of-function mutations in flb (upstream developmental activation [UDA]) genes block brlA transcription and, consequently, the production of conidiophores, the structures bearing asexual spores known as conidia. However, the aconidial phenotype of specific flb mutants, such as that of the ΔflbB strain, is reverted under salt-stress conditions. Previously, we generated a collection of second-site mutants of ΔflbB unable to conidiate on culture medium supplemented with NaH2PO4 (0.65 M). Here, we identified a Gly347Stop mutation within flpA as responsible for the FLIP57 phenotype and characterized the role of the putative cyclin FlpA and the remaining putative components of the C-terminal domain kinase-1 (CTDK-1) complex in A. nidulans and Aspergillus fumigatus. FlpA, Stk47, and FlpB are necessary (i) for timely germination, (ii) in the transition from metulae to phialides (the cells generating conidia) during conidiophore development, and (iii) for the develop ment of sexual structures (cleistothecia) in A. nidulans. The three proteins are nuclear, and the nucleoplasmic localization of Stk47 depends on the activity of FlpA, which correlates with the retention of Stk47 by FlpA in pull-down assays. Overall, this work links the putative CTDK-1 complex of aspergilli with growth and developmental control. Identification of a mutation in flpA as inhibitor of conidiation in A. nidulans and functional characterization of FlpA, Stk47 and FlpB as putative members of the C-terminal domain kinase complex CTDK-1 in A. nidulans and A. fumigatus. IMPORTANCE Aspergillus fumigatus has been included by the World Health Organiza tion in the priority list of fungal pathogens because (i) it causes 90% of invasive aspergillosis cases, with a high mortality rate, and (ii) infections are becoming increas ingly resistant to azole antifungals. A. nidulans is an opportunistic pathogen and a saprotroph which has served during the last 80 years as a reference system for filamentous fungi. Here, we characterized the role in morphogenesis and develop ment of the putative transcriptional cyclin/kinase complex CTDK-1 in both aspergilli. The null mutants of the corresponding genes showed delayed germination, aberrant conidiophore development, and inhibition of cleistothecia production. While in higher eukaryotes this complex is formed only by a cyclin and a kinase, the fungal complex would incorporate a fungal-specific third component, FlpB, which would enable the interaction between the kinase (Stk47) and the cyclin (FlpA) and may be used as a target for antifungals. November/December2023 Volume 14 Issue 6 10.1128/mbio.02452-23 1 Editor Gustavo H. Goldman, Universidade de Sao Paulo, Sao Paulo, Brazil Address correspondence to O. Etxebeste, oier[email protected]. Z. Agirrezabala and X. Guruceaga contributed equally to this article. The order of these two authors was determined alphabetically and with their consent. The authors declare no conflict of interest. See the funding table on p. 27. Received 12 September 2023 Accepted 3 October 2023 Published 9 November 2023 Copyright © 2023 Agirrezabala et al. This is an open-access article distributed under the terms of the Creative Commons Attribution 4.0 International license. Downloaded from https://journals.asm.org/journal/mbio on 03 May 2024 by 158.227.89.37. KEYWORDS filamentous fungi, Aspergillus nidulans, Aspergillus fumigatus, vegetative growth, asexual development, conidiation, sexual development, stress response, cyclin, kinase, CTDK-1, RNA polymerase II Aspergilli constitute an important genus of filamentous fungi. Taxonomically, the genus is located within the phylum Ascomycota, subphylum Pezizomycotina, and the class of Eurotiomycetes. It is composed of hundreds (~350) of species (1). Some of them are pathogens of fruits, seeds, animals, and humans. For example, Aspergillus fumigatus is the main agent causing invasive aspergillosis (2). Other aspergilli are used in industry as a source of valuable products (3, 4). Aspergillus species can reproduce sexually or asexually (5). Sexual development has been described in a subset of species of this genus, but sequencing of Aspergillus genomes has uncovered that the presence of genes coding main regulators of sex is a general trend (1). Sexual reproduction is based on meiotic cell divisions, generating spores with combinations of the genetic content of the parentals (6, 7). Asexual reproduction is based on mitosis, producing spores with an identical genetic content. This explains why sexual reproduction is mainly directed to the interchange of genetic material, while the main aim of asexual development is dispersal. Spores germinate under favorable environmental conditions. Polar growth and branching of germlings generate mature hyphae, while fusion of hyphae through anastomosis generates the mycelium, the structure specialized in substrate colonization (8). Depending on the stimulus (O2 or CO2, light/darkness, nutrient availability/starva tion, salt or osmotic stress, presence/absence of specific metabolites) (9), asexual or sexual developmental programs will be activated or repressed. Those spores disperse again to new niches, initiating new life cycles. One of the main model organisms within the genus Aspergillus is Aspergillus nidulans. Several features make this species a suitable reference organism: fast growth rates at laboratory conditions, the fact that it is not a pathogen, the possibility of carrying out sexual crosses in short periods of time, the amenability for genetic manipulation, and the availability of a vast array of standardized cellular and molecular techniques (10). Together with the Sordariomycete Neurospora crassa, A. nidulans is the main reference species in the study of developmental programs (3, 10, 11), since most of the known regulators of sexual and asexual development were identified and characterized for the first time in A. nidulans (2). Asexual spores of A. nidulans are known as conidia since they are produced by localized budding and subsequent constriction from an external sporogenous cell, called the phialide (12). Phialides and conidia are the last cell types produced in the asexual developmental cycle, giving rise to multi-cellular structures called conidiophores. Each A. nidulans conidiophore, and the same holds true for A. fumigatus, produces thousands of conidia, and each colony generates millions of conidiophores on solid culture (9, 10, 13–15). In his seminal works, Timberlake estimated that more than a thousand mRNAs increased their concentration after the induction of conidiophore development (16). To date, the number of identified and functionally characterized proteins with a role in asexual development is far from this estimation. These regulators have been classified into signal transducers, central regulators, repressors, and balancers (9, 17). There are complex functional and genetic relationships among these groups of regulators, as well as between them and regulators of other cellular processes such as polar growth, sexual development, secondary metabolism, or cell death. A simplified model describes the activity of two main pathways (18). UDA pathways are signal transduction pathways involved in the inhibition of polar growth of hyphae and the decision of whether to induce or not, depending on extracellular and intracellular stimuli, asexual development. There are at least three UDA subpathways, which are defined by flbA, flbB/flbD/flbE, and flbC, respectively. FlbB, FlbC, and FlbD are transcription factors and play an important role in the induction of the expression of brlA. In fact, brlA is the central gene in this model, since its expression is controlled by UDA-s, and then, BrlA controls the expression Research Article mBio November/December2023 Volume 14 Issue 6 10.1128/mbio.02452-23 2 Downloaded from https://journals.asm.org/journal/mbio on 03 May 2024 by 158.227.89.37. of central developmental pathway (CDP) genes, which regulate the formation of most of the cell types of the conidiophore (9, 17, 19). As mentioned above, there are known repressors of brlA expression, which act at two levels: repressors that also activate sexual development and those that repress brlA expression once asexual development has been completed (20–24). Due to their role in the control of brlA expression, deletion or loss-of-function mutations in UDA genes inhibit or delay conidiation. Mutants show an aconidial phenotype known as fluffy (14). An example is the ΔflbB mutant, in which the CDP pathway is blocked. Nevertheless, the fluffy phenotype of the null flbB mutant is reverted under specific stress conditions, such as supplementation of the standard solid Aspergillus minimal medium (AMM) with high concentrations (above 0.5 M) of NaH2PO4 (25–27). In a previous work, we took advantage of this phenotype and mutagenized through UV light conidia of a ΔflbB strain with the aim of isolating second-site mutants unable to conidiate when cultured on AMM supplemented with 0.65 M NaH2PO4 (27). These mutants were named as FLIP (fluffy in phosphate mutants). In this work, mutants FLIP57 and FLIP76 have been characterized. Identification of the mutations causing their phenotypes led to the characterization of AN10640/FlpA (fluffy in phosphate A) as a putative cyclin required in the transition from metulae to phialides during conidiophore development. Fluorescence microscopy suggested that its localization is cell cycle dependent, being located in the nucleoplasm in the interphase but not during mitosis. The same phenotype and subcellular localization were observed for its putative interaction partners AN8190/Stk47, a cyclin-dependent kinase (CDK), and AN6312/FlpB, a homolog of Schizosaccharomyces pombe Ctk3. Specific dependencies were described among these three proteins for nuclear accumulation and immunodetec tion patterns. Besides development, germination and radial growth were also remarkably affected in the null mutants of flpA, stk47, or flpB both in A. nidulans and A. fumigatus. Overall, results suggest that the activity of the putative C-terminal domain kinase-1 (CTDK-1) complex in the genus Aspergillus is required in multiple cellular processes, participating in the coordination of growth and developmental programs. RESULTS Sequencing and analysis of FLIP57 and FLIP76 genomes In a previous work, 80 second-site mutants of ΔflbB unable to conidiate on AMM supplemented with 0.65 M NaH2PO4 (FLIP mutants) were grouped into seven phenotypic groups (27). We determined that the mutant phenotype of FLIP166, which belonged to the group of FLIP mutants with a totally aconidial phenotype under phosphate stress conditions, was caused by a mutation in AN1459/pmtC (27, 28). Eight additional FLIP mutants were selected (Fig. S1), their genomic DNA extracted and pmtC sequenced. None of the FLIP mutants analyzed bore a mutation in this gene, showing that their phenotypes were not caused by mutant pmtC alleles. Genomic DNA samples of FLIP57 and FLIP76 (Fig. 1A) were sequenced and compared to the reference FGSC4 genome, the FLIP166 genome, and A. nidulans transcriptomes (see Materials and Methods; see File S1) (23, 27, 29). Four and three exonic candidate mutations were identified in FLIP57 and FLIP76, respectively (Fig. 1B). However, in FLIP76, an exonic mutation was identified in AN5717/kapI, leading to a premature stop codon (R557E+8-Stop; KapI is 1,095 amino acids long; File S1; Fig. 1B). Since we previously described that the double ΔkapI;ΔflbB mutant was completely aconidial under stress conditions induced by the addition of 0.5 M NaH2PO4 (30), we concluded that the FLIP76 phenotype was caused by the additive effect of flbB deletion and a loss-of-function mutation in kapI. Consequently, we discarded FLIP76 and continued the analysis with FLIP57. In this second mutant, out of four exonic mutations identified, AN6932 and AN7856 (File S1) were discarded because the corresponding null mutant did not show a defect in conidiation (31) or because the gene showed very low expression levels (AN7856). AN7842 was also discarded because, despite the low expression levels, RNA-seq reads suggested that it is not correctly annotated. Thus, the analysis was focused on the Research Article mBio November/December2023 Volume 14 Issue 6 10.1128/mbio.02452-23 3 Downloaded from https://journals.asm.org/journal/mbio on 03 May 2024 by 158.227.89.37. mutation found in AN10640 because (i) RNA-seq data showed that the annotation of exons and introns was correct; (ii) it showed moderate expression levels (see below); and (iii) the gene is predicted to encode a cyclin. That would link the control of asexual development with the control of the cell cycle, transcription, and/or additional cellular processes. A point mutation corresponding to a Gly347Stop truncation in AN10640/ FlpA caused the FLIP57 phenotype AN10640 is located in chromosome V (coordinates 1,323,461–1,325,516, coding strand) and is composed of five exons and four introns. The G-T mutation of FLIP57 is located FIG 1 Mutations found in FLIP57 and FLIP76. (A) Phenotypes of mutants FLIP57 and FLIP76 after 72 hours of culture at 37°C in AMM (row 1) or AMM supplemented with 0.65 M NaH2PO4 (row 2). A parental ΔflbB and the mutant FLIP166 (27) were used as controls. (B) Mutations found in the genomes of FLIP57 and FLIP76, which could be responsible for the corresponding FLIP phenotypes. Research Article mBio November/December2023 Volume 14 Issue 6 10.1128/mbio.02452-23 4 Downloaded from https://journals.asm.org/journal/mbio on 03 May 2024 by 158.227.89.37. in exon 5 and modifies the codon in the position 347 (GGA, Gly) by a stop signal (TGA, Gly347Stop) (Fig. 2A). Since the corresponding protein is predicted to be 392 amino acids long, the FLIP57 mutation would cause the truncation of the last 45 amino acids (Fig. 2A). RNA-seq data showed that, in the culture conditions and genetic backgrounds analyzed by others in A. nidulans (reviewed in reference 32), the position and extension of introns of AN10640 matched the FungiDB annotations. Furthermore, RNA-seq data, including FPKM values for wild type and the null flbB backgrounds before (VG) and after (AD, 5 hours) the induction of asexual development (23, 33), showed moderate expression levels in all conditions/backgrounds analyzed, with no remarkable induction/inhibition of AN10640 in any of the samples (Fig. 2B). The Interpro server predicted the presence of a cyclin-like domain (IPR013763, amino acids 59–263; Fig. 2A), which is described to be found in cyclins but also in transcription factor IIB and in the retinoblastoma tumor suppressor (34–36). The FungiDB database described that orthologs of AN10640 have roles in positive regulation of septation initiation signaling, regulation of phosphor ylation of RNA polymerase II C-terminal domain, and trimeric positive transcription elongation factor complex b localization. The FLIP57 mutation is located outside of this putative cyclin-like domain. To confirm that the mutation Gly347Stop in AN10640 (the gene was named as flpA, fluffy in phosphate A) was the cause of the FLIP57 phenotype, protoplasts of this strain and those of its parental ΔflbB strain were transformed with flpA::ha3x::pyrGAfum, flpA(Gly347Stop)::ha3x::pyrGAfum, flpA::gfp::pyrGAfum, or flpA(Gly347Stop)::gfp::pyrGAfum constructs (Fig. 2C). The insertion of the mutant constructs in the null flbB parental caused an inhibition of conidiation in AMM supplemented with 0.5 M NaH2PO4, while the wild-type constructs did not alter the ΔflbB phenotype. Inversely, the wild-type constructs reverted the FLIP57 phenotype under phosphate stress. No difference was observed between flpA::ha3x and flpA::gfp counterparts, suggesting that the size of the 3′-tag is not detrimental to FlpA activity. On the contrary, the mutant construct flpA(Gly347Stop)::gfp::pyrGAfum did not revert the FLIP57 phenotype under phosphate stress (Fig. 2C). These results confirmed the aforementioned hypothesis. FlpA and its putative interactors Stk47/AN8190 and FlpB/AN6312 are widely conserved in the kingdom Fungi Paolillo and colleagues carried out a deep bioinformatics and phylogenetic analysis of A. nidulans cyclins (34). They classified AN10640 within group III cyclins, as a T/K-like cyclin, as was AN4981/PchA. To determine the conservation pattern of FlpA in the fungal kingdom, we carried out BLAST analyses to identify putative ortholog sequences. Table S1 shows the taxonomy of the species corresponding to the FlpA hits analyzed, the length of the hit, score and e values, the query coverage, and the result (first sequence) of the confirmatory reverse retrieval of each hit sequence at the FungiDB database. Only hits giving AN10640/FlpA as the first sequence in this confirmatory reverse retrieval were considered. Overall, the values given in Table S1, together with the phylogenetic tree in Fig. 3A, show that FlpA is widely conserved in the kingdom Fungi. Xie and colleagues have recently determined the structure of the Schizosaccharomy ces pombe ortholog of FlpA, Ctk2, as part of the trimeric CTDK-1 (C-terminal domain [CTD] kinase I) complex (37). CTDK-1 acts as the primary RNA polymerase II CTD Ser2 kinase complex. The Swiss-Model website modeled FlpA (from His26 to Lys374; Model 7jv7.1.B was taken as the reference; sequence identity of 18.67%) (37), with higher confidence, as expected, in the cyclin-like domain than in the case of the C-terminus (Fig. 3D; Fig. S2A). Since it corresponded to a stop signal, Dynamut (38) was unable to predict the hypothetic effect of the FLIP57 mutation in the structure of AN10640. The remaining two components of the CTDK-1 complex in S. pombe are Ctk1, a cyclin-dependent kinase, and Ctk3, the latter being a Ctk1 activator and contributing to the assembly of the complex by interacting with Ctk1 and Ctk2 (37). The A. nidulans ortholog of Ctk1 is AN8190/Stk47 (39), while that of Ctk3 is AN6312/FlpB. Exons and introns of both genes were annotated correctly in the FungiDB database (not shown). Research Article mBio November/December2023 Volume 14 Issue 6 10.1128/mbio.02452-23 5 Downloaded from https://journals.asm.org/journal/mbio on 03 May 2024 by 158.227.89.37. FIG 2 A mutation within AN10640/flpA causes the FLIP57 phenotype. (A) Annotation of AN10640/flpA and analysis of the mutation causing the FLIP57 phenotype. The locus flpA was compared in FLIP57, FLIP76, and FLIP166 genetic backgrounds and with transcriptomes of a null flbB strain (RNA-seq1) and a null sltA strain (RNA-seq2). FLIP57 bore a mutation in the last exon of AN10640, which led to the substitution of the codon for Gly347 by a stop (Continued on next page) Research Article mBio November/December2023 Volume 14 Issue 6 10.1128/mbio.02452-23 6 Downloaded from https://journals.asm.org/journal/mbio on 03 May 2024 by 158.227.89.37. Table S1 includes more than 400 fungal species containing FlpA, Stk47, and FlpB (see below) orthologs, representing 22 classes within the phyla Ascomycota, Basidiomycota, and Mucoromycota (Fig. 3B and C). The Swiss-Model website modeled both proteins taking, as in the case of FlpA, PDB structure 7jv7.1 as the reference (subunits α and γ, respectively, for Stk47 and FlpB; much lower coverage in the case of Stk47, from Ser746 to Leu1052; from Glu7 to Asn234 in the case of FlpB) (Fig. 3D; Fig S2B and C). ColabFold predicted the structure of a putative complex formed by the three A. nidulans proteins (full-length versions of FlpA and FlpB were used as queries and the fragment from Pro735 to Leu1067 in the case of Stk47, based in this latter case on the Swiss-Model prediction). The model in Fig. 3E shows that FlpB would interact with both FlpA and Stk47, enabling the assembly of the complex, as has been described in S. pombe (see the structure in Fig. S2D) (37). Overall, bioinformatics results strongly suggest that CTDK-1 is a widely conserved trimeric complex in the kingdom Fungi, while in Homo sapiens, there is no Ctk3 ortholog (37) (see Discussion). Deletion of flpA causes a decrease in radial growth and conidial yield, while truncation of the last 45 amino acids of FlpA has negligible effects in a wild-type background To functionally characterize the role of FlpA in growth and development of A. nidulans, a null flpA mutant was generated, both in wild-type and ΔflbB genetic backgrounds. Wild-type protoplasts were also transformed with wild-type and mutant (Gly347Stop) flpA::ha3x::pyrGAfum constructs and also a wild-type flpA::gfp::pyrGAfum construct (see Materials and Methods). Phenotypes of selected strains were analyzed in solid AMM and AMM supplemented with 0.5 M NaH2PO4 (Fig. 4A). First, we observed that, after 72 hours of culture at 37°C, deletion of flpA caused a significant inhibition in radial extension, both in wild-type and null flbB backgrounds, and mainly in AMM supplemented with 0.5 M NaH2PO4 (Fig. 4A through C). The average growth rates from 48 to 96 hours of culture were 0.027 ± 0.004 and 0.025 ± 0.003 cm/hour for wild-type and ΔflbB parentals in medium supplemented with sodium dihydrogen phosphate. The values for the null flpA and double-null ΔflpA;ΔflbB strains were 0.012 ± 0.003 cm/hour and 0.012 ± 0.001 cm/ hour, respectively, in the same culture medium (n = 3 for each strain, P = 0.006 and 0.004 for each null mutant compared to the parental strain). Furthermore, compared to the null flbB parental, the strain that integrated the flpA(Gly347Stop)::ha3x::pyrGAfum construct in this background showed only a minor decrease in the growth rate (0.023 ± 0.004 cm/h, n = 3, P = 0.629). The strain that integrated the flpA(Gly347Stop)::ha3x::pyrGAfum construct (wild-type background) showed no inhibition of conidia production compared to the reference strain (6.17 × 107 ± 1.45 × 107 and 3.62 × 107 ± 1.55 × 107 conidia/cm2, for mutant and wild-type strains, respectively; n = 3 for each strain; mutant not shown in the graphs; P = 0.035) in AMM culture medium. In the null flpA strain, however, the inhibition was statistically significant (1.86 × 106 ± 3.23 × 105 conidia/cm2 in the null flpA strain, n = 3, P = 0.003) (Fig. 4A and D). The results obtained in the ΔflbB background confirmed the trends described for the wild-type background, with the difference that both deletion or truncation of the last 45 codons completely inhibited conidia production, probably due to the additive effect on conidiation caused by the absence of flbB (26, 30, 40). However, the C-terminus of FlpA could play a minor role in these processes, as observed in the wild-type background. FIG 2 (Continued) codon (loss of the last 45 amino acids). The identified mutation is outside the cyclin-like domain (IPR013763, amino acids 59–263). (B) FPKM or TPM values for AN10640/flpA in the multiple RNA-seq experiments available for A. nidulans (see references within reference 32). (C) Confirmation of the Gly347Stop truncation within AN10640/flpA as the mutation causing the FLIP57 phenotype. (Left) Phenotypes of FLIP57 and its parental ΔflbB strain on AMM and AMM supplemented with 0.5 M NaH2PO4. (Middle) Transformants of ΔflbB protoplasts with the synthetic DNA constructs flpA::ha3x::pyrGAfum, flpA(Gly347Stop)::ha3x::pyrGAfum, flpA::gfp::pyrGAfum or flpA(Gly347Stop)::gfp::pyrGAfum. (Right) Transformants of FLIP57 protoplasts with the synthetic DNA constructs flpA::ha3x::pyrGAfum, flpA::gfp::pyrGAfum or flpA(Gly347Stop)::gfp::pyrGAfum. Research Article mBio November/December2023 Volume 14 Issue 6 10.1128/mbio.02452-23 7 Downloaded from https://journals.asm.org/journal/mbio on 03 May 2024 by 158.227.89.37. FIG 3 Evolutionary analysis of FlpA/AN10640, Stk47/AN8190, and FlpB/AN6312. Phylogenetic trees (maximum-likelihood method and the JTT matrix, with 50 replicates in each case) corresponding to the putative orthologs of FlpA (A), Stk47 (B) and FlpB (C) included in Table S3. Coverage (bar graphs in the trees) and score (shape plots) values for each hit compared to A. nidulans queries are included. The color key indicates which fungal class each ortholog belongs to. (Continued on next page) Research Article mBio November/December2023 Volume 14 Issue 6 10.1128/mbio.02452-23 8 Downloaded from https://journals.asm.org/journal/mbio on 03 May 2024 by 158.227.89.37. Null mutants of stk47 and flpB show the same phenotype as the null flpA strain Since FlpA, Stk47, and FlpB may be the components of the CTDK-1 complex in A. nidulans, single-null mutants of stk47 and flpB were also generated, and their phenotypes were compared to that of the ΔflpA strain after 72 hours of culture at 37°C in solid AMM or AMM supplemented with 0.5 M NaH2PO4 (Fig. 4E and F). Different combinations of double-null mutants were also generated by transformation of protoplasts of the single-null mutants with the corresponding deletion cassettes. The three single-null mutants showed a highly similar phenotype. We observed a slight increase in colony diameter and production of conidia only in the case of the null flpB strain. While the reference wild-type strain produced 2.67 × 107 ± 0.82 × 107 conidia/cm2, ΔflpA, Δstk47, and ΔflpB strains produced, respectively, 5.94 × 105 ± 3.80 × 105, 5.63 × 105 ± 2.11 × 105, and 1.02 × 106 ± 2.26 × 105 conidia/cm2 (P = 0.032 in the comparison between the wild-type and the null flpA strains, P = 0.908 and 0.172 when ΔflpA was compared to Δstk47 and ΔflpB strains, P = 0.063 when null stk47 and null flpB strains were compared; n = 3 for each strain). Of note is the phenotypic difference between null stk47 strains when A. fumigatus pyrG or riboB genes were used as selection markers (Fig. 4E and F). Conidia production increased from 5.63 × 105 ± 2.11 × 105 conidia/cm2 in the Δstk47::pyrG strain to 1.38 × 107 ± 2.02 × 106 conidia/cm2 in the Δstk47::riboB strain (P = 0.008, n = 3 for each strain). Double-null mutants displayed a highly similar phenotype compared to their corresponding parental strains, and no additive effects in conidia production were observed after deletion of the second gene (Fig. 4E and F). The double-null ΔflpA;ΔflpB (generated after transformation of ΔflpA protoplasts with the construct for flpB deletion) produced 1.07 × 106 ± 2.75× 105 conidia/cm2 (P = 0.190 compared to its parental strain, n = 3 for each strain). Double-null mutants Δstk47;ΔflpA and Δstk47;ΔflpB (both generated by transformation of protoplast of the Δstk47::riboB strain) produced 1.36 × 107 ± 2.04 × 106 and 2.08 × 107 ± 1.34 × 106 conidia/cm2 (P = 0.914 and 0.008 when compared to their parental strain). Finally, the ability to generate cleistothecia by the single-null flpA, stk47, and flpB mutants was assessed qualitatively. Figure 4G clearly shows that the mutants do not produce cleistothecia in our culture conditions (7 days of culture at 37°C in solid AMM). Overall, these results strongly suggest that FlpA, Stk47, and FlpB are necessary for both morphogenesis and sexual or asexual developmental processes. The phenotypic similarities among single-null and double-null mutants also suggest that the three proteins function in the same pathway, in correlation with their hypothetic participation in the formation of the CTDK-1 complex. ΔflpA, Δstk47, and ΔflpB strains generate aberrant conidiophores with a deficient metula-to-phialide transition The null flbB mutant fails to induce conidiophore development, while deletion of both flpA and flbB has an additive inhibitory effect on conidia production. Thus, we hypothe sized that FlpA could have a role different from that carried out by FlbB. While the latter is necessary for the induction of asexual development, FlpA may play a role in the correct and timely generation of the cell types that form conidiophores (foot-cell, stalk, vesicle, metulae, phialides, and conidia) (13). To verify that, we carried out a phenotypic characterization of the ΔflpA strain in submerged culture and under nitrogen FIG 3 (Continued) (D) Predicted three-dimensional structures for FlpA, Stk47, and FlpB, modeled by Swiss-Model. The models cover the regions from His26 to Lys374 in FlpA, from Ser746 to His1052 in Stk47, and from Glu7 to Asn234 in FlpB, and are based on PDB structure 7jv7.1, which corresponds to the CTDK-1 complex of S. pombe and is shown in Fig. S2D (37). The alignments between FlpA, Stk47, or FlpB and the reference structure can be seen in Fig. S2A–C. (E) Predicted three-dimensional structure for a hypothetic complex formed by FlpA, Stk47, and FlpB, modeled by ColabFold. Full-length sequences of FlpA and FlpB were used, together with an N-terminally truncated form of Stk47 covering residues from Pro735 to Leu1067. Research Article mBio November/December2023 Volume 14 Issue 6 10.1128/mbio.02452-23 9 Downloaded from https://journals.asm.org/journal/mbio on 03 May 2024 by 158.227.89.37. antifungal compounds and for virulence in a mouse model of invasive aspergillosis. The three single-null mutants showed a clear growth and developmental phenotype (Fig. 8A), with the difference that the radius of ΔflpB colonies in medium supplemented with 0.5 M NaH2PO4 is slightly bigger than those of ΔflpA and Δstk47 colonies (Fig. 8B). In all cases, the decrease in colony radii of single-null mutants compared with the reference wild-type strain was statistically significant (P < 0.05 in the three comparisons). We also observed a delay in conidial germination in the three single-null mutants (Fig. 8C), which may partially explain the radial growth phenotypes described in Fig. 8A and B. To test for impacts on growth in host-niche stress, we next examined the ability of the mutants to grow under oxidative, iron limitation, hypoxic, and cell wall stresses. All strains showed similar growth patterns in response to oxidative stress and under oxygenor iron-limiting conditions (Fig. S6). However, we found that when Δstk47, ΔflpA, and ΔflpB conidia were inoculated on medium supplemented with Congo red (CR) or calcofluor white (CFW), all mutants showed reduced radial growth compared to the reference wild-type strain (Fig. 8D). To confirm that the observed hypersuscept ibility phenotype of the mutants to these cell wall stressors is not a result of a ger mination defect, we decided to test the ability of mature hyphae to develop under these conditions. Thus, we grew the same strains in minimal medium for 36 hours and transferred mycelial pellets of the same size to AMM agar plates supplemented with CR or CFW (80 µg/mL). After 72 hours of culture, CR caused a 48.48 ± 5.01 % growth reduction in the reference strain, when compared to minimal medium (Fig. 8E and F). Growth reduction was of 64.34 ± 4.33 %, 63.79 ± 3.69 %, and 69.42 ± 2.96% for Δstk47, ΔflpA, and ΔflpB mutants, respectively (P < 0.05 in all cases). The addition of CFW caused a 28.66 ± 4.33% growth reduction in the reference strain, while it was much greater in the single-null mutants, mainly in ΔflpA (84.14 ± 2.13%) and ΔflpB (97.26 ± 1.25%) (P < 0.05 in all cases). To see if this susceptibility would also be evident in response to cell wall-targeting antifungal drugs, we next tested susceptibility of the mutants to the echinocandin caspofungin, which inhibits β-glucan synthesis. Using a strip diffusion assay, we found the wild-type and ΔflpA caspofungin minimum effective concentration (MEC) was 0.125 µg/mL, and the ΔflpB and Δstk47 displayed a caspofungin MEC of 0.25 µg/mL (Fig. S7). In addition, all strains displayed similar susceptibilities to membrane-targeting antifungals, with a voriconazole MIC ( range of 0.125–0.25 µg/mL and posaconazole MIC of 0.25 µg/mL. Therefore, loss of stk47, flpB or flpA does not impact antifungal susceptibility. The hypersusceptibility of the three mutant strains to CR and CFW may suggest that the organization of the cell wall, and subsequently PAMP exposure, could be altered by deletion of flpB, flpA, or stk47. Altered PAMP exposure could subsequently lead to changes in host recognition during infection. To check if deletion of any gene alters host recognition, we next challenged THP-1 monocytes with conidia of the fungal strains and assayed inflammasome activation using interleukin (IL)-1β release as an endpoint (44). However, only a slightly lower release of IL-1β in the single-null mutants compared to FIG 7 (Continued) the green square, the lighter band increases its intensity in ΔflpA and ΔflpB backgrounds. (C) Treatment of crude protein extracts of Stk47::HA3x with λ phosphatase (λPP) and λPP plus sodium orthovanadate, the latter as an inhibitor of phosphatase activity. Gels stained with Bio-Safe Coomassie (Bio-Rad) are shown as loading controls in panels A–C. (D) Pull-down assay using the GFP-Trap resin of Chromotek. Resin samples were sequentially incubated (see Materials and Methods) with 3 mg of a crude protein extract of a strain expressing FlpA::GFP and after washing of the non-retained fraction, with 3 mg of a second crude protein extract of a strain expressing Stk47::HA3x. (E) Pull-down assays carried out by sequentially incubating the GFP-trap resin with 3 mg of a crude protein extract of a strain expressing FlpB::GFP (wild-type or ΔflpA background), and then with 3 mg of a second crude protein extract of a strain expressing FlpA::HA3x or Stk47::HA3x (wild-type or ΔflpA background). Strain-free gels (Bio-Rad) were used for protein electrophoresis in panels D and E. The image in row 1 was obtained as loading control before protein transference to polyvinylidene difluoride (PVDF) membranes. NR, non-retained; R, retained fraction; TE, total extract. Research Article mBio November/December2023 Volume 14 Issue 6 10.1128/mbio.02452-23 16 Downloaded from https://journals.asm.org/journal/mbio on 03 May 2024 by 158.227.89.37. FIG 8 Phenotype of single-null mutants of flpA, stk47, and flpB in A. fumigatus. (A) Phenotypes of wild-type and single-null flpA, stk47, and flpB strains of A. fumigatus after 72 hours of culture at 37°C in AMM or AMM supplemented with 0.5 M NaH2PO4. Diameter of plates is 5.5 cm. (B) Diameter of the colonies in panel A after 24, 48, and 72 hours of culture under the same conditions (***P < 0.001, compared to the reference ΔakuB-pyrG + strain; n = 3 for each strain). (Continued on next page) Research Article mBio November/December2023 Volume 14 Issue 6 10.1128/mbio.02452-23 17 Downloaded from https://journals.asm.org/journal/mbio on 03 May 2024 by 158.227.89.37. the reference wild-type strain was detected, and these differences were not statistically significant (P > 0.05 in the three comparisons; Fig. 8G). Therefore, we concluded that there are no significant differences in the ability of the mutants to trigger the inflammasome. Considering the growth, cell wall, and developmental phenotypes, we also analyzed the virulence of the A. fumigatus single-null mutants in a chemotherapeutic model of invasive aspergillosis. We observed that, after 12 days of infection, 50% of those animals challenged with the knockouts were alive, compared to only a 20% of mice infected with the parental strain. However, due to the limited sample size, the mortality rates between strains were not statistically significant (Fig. 8H). Taken together with the in vitro stress assay data, it is unlikely that flpB, flpA, or stk47 play major roles in support of A. fumigatus virulence. DISCUSSION Among the different types of kinases, which are differentiated based on the sequence of the kinase domain, cyclin-dependent kinases, or CDKs, are serine/threonine kinases whose activity depends on the regulatory role carried out by a cyclin (45). Cyclins were named this way, first, because their protein levels fluctuated in a cyclical fashion during the cell cycle and, second, because they were defined by the presence of a cyclin box domain (see references within reference 46). Cyclins can be divided into two main groups, cell-cycle or canonical cyclins and transcriptional cyclins (46). The term transcriptional cyclins refers to their role in the regulation of RNA polymerase activity during transcription initiation and elongation. Thus, the role of cyclins goes beyond the control of events directly linked to the cell cycle. Furthermore, members of both groups of cyclins play cellular functions independently of a partner CDK. Similarly, CDKs also control transcription, metabolism, and cell differentiation (47). Paolillo and coworkers identified, in the genomes of A. nidulans and A. fumigatus, 15 genes encoding cyclins that represent the three phylogenetic groups: group I or cell cycle cyclins (three in each genome), group II (seven in each genome) , and group III (five in each genome), which mainly include transcriptional cyclins (34). On the other hand, De Souza and Osmani characterized the phenotypes of the single-null mutants of all the A. nidulans genes encoding kinases, totaling 128 (39). The null mutant of AN8190/stk47 was described to show a strong growth defect and sensitivity to stress conditions triggered by sodium chloride. Our groups recently completed the systematic disruption of all putative protein kinase encoding genes in A. fumigatus (44). Interest ingly, the A. fumigatus stk47 disruption mutant displayed only wild-type phenotypes (our unpublished results), which is in contrast to the A. fumigatus Δstk47 results reported here. However, the previously published work describing these results identified additional kinase disruption mutants that also displayed differential results upon disruption or deletion (44). These inconsistencies are presumed to be caused by continued transcrip tional read through, and subsequent expression of truncated kinase mutants, in at least some of the disruption mutants contained within the A. fumigatus library. In the analysis of the set of fluffy in phosphate mutants of our A. nidulans collection, we identified a Gly347Stop substitution in AN10640/flpA as the mutation causing the FLIP57 phenotype. FIG 8 (Continued) (C) Germination rate for each strain 8, 10, or 12 hours after inoculation (n = 3 for each strain at each culture time). (D) Phenotypes of the strains after 48 hours of culture on AMM supplemented with 40 or 80 µg/mL of CR or CFW. Conidia were inoculated at four different concentrations (104, 103, 102, or 10 conidia). Diameter of plates: 9 cm. (E) Phenotypes (72 hours at 37°C) of the same strains after the transference of mycelial pellets grown in liquid minimal medium for 36 hours to plates containing minimal medium (control) or medium supplemented with 80 µg/mL of CR or CFW. Pictures are representative of biological triplicates. (F) Bar graphs representing the percentage of growth reduction of each strain cultured in the presence of CR or CFW compared to minimal medium. Analysis was carried out by one-way analysis of variance with Dunnet’s multiple comparison test. (Top) P ≤ 0.0045; (Bottom) P ≤ 0.0006. (G) Quantification of release of interleukin-1β by THP-1 monocytes after co-culturing with conidia (multiplicity of infection: 10 conidia per THP-1 cell). (H) Percent survival of female CD-1 (n = 10 mice per group) to infection by the wild-type or the A. fumigatus single-null mutants of flpA, stk47, and flpB. Survival was recorded daily, and the Kaplan-Meier curves were compared using log-rank tests in GraphPad Prism (v.9.2.0). CFW, calcofluor white; CR, Congo red. ** P < 0.01; *** P < 0.001; **** P < 0.0001 Research Article mBio November/December2023 Volume 14 Issue 6 10.1128/mbio.02452-23 18 Downloaded from https://journals.asm.org/journal/mbio on 03 May 2024 by 158.227.89.37. Since FlpA is the ortholog of S. pombe cyclin Ctk2/Lsc1, its activity was linked to that of the CDK Stk47 and the hypothetic formation of an A. nidulans CTDK-1 complex (37, 43). In humans, CDK12-Cyclin K and CDK13-Cyclin K complexes would be the counterparts of the S. pombe Ctk1/Ctk2 complex (37). Nevertheless, the structure of the S. pombe CTDK-1 complex differs from the human counterparts in that it adds a third member, Ctk3, which is not conserved in Homo sapiens. Ctk3 enables the formation of the S. pombe CTDK-1 complex by interacting with both the cyclin and the kinase. This configuration may be conserved in the kingdom Fungi based on (i) the wide conservation patterns of the three components, (ii) the ColabFold prediction, (3) the highly similar phenotypes of the three single-null mutants in A. nidulans and A. fumigatus, and (4) the interaction patterns described in the pull-down assays carried out in this work. The NLStradamus algorithm predicted the presence of a NLS sequence in AnStk47 (1,119 amino acids) but not in AnFlpA (392 amino acids) or AnFlpB (240 amino acids). However, our results showed that, in A. nidulans, nuclear accumulation of AnStk47 was dependent on AnFlpA activity, while nuclear accumulation of AnFlpA or AnFlpB was not dependent on AnStk47. This behavior is opposite to what was described in S. pombe. In fission yeast, Lsk1/Ctk1 was required for nuclear localization of Lsc1/Ctk2, but not inversely (43). Regarding the phosphorylation pattern of AnStk47, our results are not conclusive. The use of λ phosphatase did not modify the immunodetection-band pattern of AnStk47, and thus, we were not able to confirm that the Stk47 band pair altered in the ΔflpA and ΔflpB backgrounds corresponds to a (poly)phosphorylated form of the kinase. Previously, preliminary phosphopeptide-detection analyses by our group (unpublished) detected multiple phosphorylated peptides for AnStk47, suggest ing that this kinase may be polyphosphorylated at Ser173, Ser177, Ser221, Ser498, Ser499, Ser505, Ser720, Thr1105, Ser1106, Ser1111, and Ser1112. In addition, the same analysis identified a phosphopeptide phosphorylated both at Ser2 and Ser5 of at least the heptad located at positions 1684–1690 (total length: 1745 aa) of the large subunit of RNA polymerase complex II, AnRpb1. Phosphopeptides for AnFlpA or AnFlpB were not detected. In the phosphopeptide-detection experiments carried out in this work, we were able to detect just a single phosphopeptide corresponding to the hypothetic phosphorylation of Ser720 of AnStk47 (no phosphopeptides for AnRpb1, AnFlpA, or AnFlpB were detected). Interestingly, this phosphopeptide was not detected in the null flpA strain, but, unfortunately, we have not been able to reproduce this result. Dephoure and colleagues described that the probability of observing site-determining ions is hampered by biases in peptide fragmentation that favor breakage at certain points and disfavor it at others (48). The same authors also stated that the type of protease used in sample processing (trypsin was used in this study; see Materials and Methods) may limit phosphorylation-site identification because not all phosphorylated residues are located in regions that will generate peptides detectable by mass spectrometry upon cleavage by a single protease (48). Future experiments will have to determine if FlpA is the cyclin or one of the cyclins for Stk47, if both proteins are required for phosphorylation of Rpb1, and if phosphorylation of Stk47 is a prerequisite for its nuclear import and/or retention. Deletion of flpA, stk47, or flpB has pleiotropic effects on colony formation both in A. nidulans and A. fumigatus. Compared to the reference wild-type strain, (i) germination is delayed; (ii) the radius of the colony is decreased; (iii) aberrant conidiophores with a misscheduled traisition from metulae to phialides are generated; and (iv) cleistothecia are not produced. If these three genes encode the components of the CTDK-1 complex, then these phenotypic traits are probably a consequence of an altered phosphorylation pattern of Rpb1. It has been described that phosphorylation of serine 2 of the C-terminal heptad repeats of Rpb1 is not an essential feature of general transcription in fission yeast but rather is critical for certain biological responses, such as sexual development (49). For example, Ser2 phosphorylation is critical during sexual differentiation of S. pombe for the induction of ste11 transcription (49, 50), which encodes an HMG domain mating-type transcription factor. In correlation with these observations, the single-null mutants of flpA, stk47 and flpB do not develop cleistothecia, in A. nidulans, in the conditions tested. Research Article mBio November/December2023 Volume 14 Issue 6 10.1128/mbio.02452-23 19 Downloaded from https://journals.asm.org/journal/mbio on 03 May 2024 by 158.227.89.37. In the case of asexual development, the aberrant conidiophore structures described in this work are probably related to germination and growth defects, rather than (i) misscheduled or incomplete septation processes, as was described in S. pombe (43), or (ii) a direct role of these three proteins in the control of the expression patterns of UDA or CDP genes. The involvement of genes encoding regulators of germination and hyphal polarity in the formation of conidiophores has been previously described [see, e.g., the cases of cdc42 and rac1 orthologs of A. nidulans (51)]. The role of transcriptional cyclins and kinases in A. nidulans conidiation was previously described by the group of Fischer et al. (52, 53). PtkA, a Cdk9 kinase, interacts with cyclins PchA, PclA and PclB, and also with the kinase PipA, modulating its interaction partners and activity to control morphogenesis and development. Cdk9 is supposed to phosphorylate Ser2 residues in the CTD of Rpb1 and enable transcript elongation (54). Thus, the phosphorylation status of the CTD domain of the large subunit of RNA polymerase II is regulated by multiple cyclin-kinase complexes, which would modify their configuration, depending on the morphogenetic/developmental stage and the cell type. If confirmed in the future for FlpA and Stk47, such a mechanistic flexibility would enable, on the one hand, coordination of multiple pathways and accurate integration of the corresponding signals and, on the other hand, modification of the affinity of RNA polymerase II complex for target promoters. MATERIALS AND METHODS DNA sequencing Genomic DNA of mutants FLIP57 and FLIP76 (Table S3) was extracted following standard procedures and using a phenol:chloroform:isoamyl alcohol mixture (25:24:1 vol/vol, PanReac Applichem) (55). An1459/pmtC was amplified from these genomic samples using oligonucleotides PmtC-seq1 and PmtC-GSP2 and sequenced using oligonucleoti des PmtC-seq1, PmtC-seq4, PmtC-Up, and PmtC-GSP2 (Table S4) to discard that the phenotypes of FLIP57 or FLIP76 were caused by mutations in pmtC, as occurred with FLIP166 (27). After integrity, purity and quantity checks (1.5% agarose electrophoresis and Qubit fluorimeter analyses), DNA samples were sequenced at Stabvida (Caparica, Portugal) using an Illumina Novaseq platform and 150-bp paired-end sequencing reads. The samples generated 16,607,684 (2,507 Mbp) and 16,279,736 sequence reads (2,458 Mbp), respectively, resulting in a theoretical average depth coverage of 83× and 82×, assuming a genome size of 30 Mpb. Bioinformatics Reads were mapped using BWA-MEM (56) at the Galaxy platform (https://usegalaxy.org/), and the SAM files generated were converted to BAM counterparts (SAM-to-BAM). SNPs compared to the reference A. nidulans genome (A_nidulans_FGSC_A4_version_s10-m02r03_chromosomes.fasta) were identified using Snippy. Results for strains FLIP57 and FLIP76 were compared also with those obtained for the mutant FLIP166 (27). Exonic mutations present in FLIP57 or FLIP76 but absent in the genomes of FLIP166 and the reference genome were selected as candidates. The Integrative Genomics Viewer software (57) was used to visualize and analyze DNA-seq and RNA-seq reads and to compare them with reference genomes. Gene and protein sequence analyses were carried out in the FungiDB database (58). The presence of putative functional domains in protein sequences was predicted with InterPro (59), while the National Center for Biotechnology Information website (https://blast.ncbi.nlm.nih.gov/Blast.cgi) was used to BLAST query sequences. Clustal Omega was used to align protein sequences (60). The .msf file generated was used to visualize the alignments with Genedoc, and the clustal file generated was imported into MEGAX (61) in order to generate phylogenetic trees, which were edited using iTOL (62). Structure homology models for FlpA, Stk47, and FlpB were built with Swiss Model (63) Research Article mBio November/December2023 Volume 14 Issue 6 10.1128/mbio.02452-23 20 Downloaded from https://journals.asm.org/journal/mbio on 03 May 2024 by 158.227.89.37. and ColabFold (64). The presence of nuclear localization signals was predicted using NLStradamus and SeqNLS algorithms (65, 66). Finally, GO analyses were carried out at the ShinyGO website (v.0.61) (67). Strains, oligonucleotides, and culture conditions Aspergillus nidulans and A. fumigatus strains used in this study are listed in Table S3, while oligonucleotides can be seen in Table S4. Parental strains FLIP57 and FLIP76 were previously generated in a mutagenesis screening of a null flbB (BD177) strain (27). A. nidulans strains were cultivated in adequately supplemented liquid or solid AMM using glucose (2%) and ammonium tartrate (5 mM) as the sources of carbon and nitrogen, respectively (68, 69). A medium containing 25 g/L corn steep liquor (Sigma-Aldrich) and sucrose (0.09 M) as the carbon source was used as the fermentation medium (Aspergillus fermentation medium [AFM]) to culture samples for protein extraction (70). A. fumigatus strains were routinely cultured on AMM agar plates and NaH2PO4 (0.5–0.65 M) was added to AMM to induce salt-stress conditions. To assess sensitivity of A. fumigatus strains to CR or CFW, 10 µL of 10-fold serial dilutions of conidia from each strain (104, 103, 102, and 10 conidia) were spotted on Petri plates filled with AMM supplemented with 40 or 80 µg/mL of each compound (71). Plates were incubated for 48 and 72 hours at 37°C. Colony growth of the mutants was compared to that of the reference strain over the inoculum dilution range. To evaluate the susceptibility to CR and CFW using mature hyphae, 100 mL of AMM broth was inoculated with 5 × 107 conidia and incubated for 36 hours at 37°C and 250 rpm. Then, mycelial pellets of each strain were isolated and placed on the center of Petri plates containing minimal medium supplemented with 80 µg/mL of CR or CFW. Control plates contained no drug (minimal medium). The plates were incubated at 37°C and diameters were measured after 72 hours. The ability of the A. fumigatus strains to respond to host-stress conditions was evaluated with the following assays: to determine the ability to grow in conditions of iron starvation, 103 conidia were point inoculated in the center of AMM plates or AMM lacking iron, and colony morphologies were observed after 48 hours at 37°C. To evaluate the growth in low oxygen environments, AMM plates containing 103 conidia of each strain were incubated in a hypoxia chamber at 37°C for 48 hours. The susceptibility to oxidative stress was evaluated as previously described (72), and growth inhibition halos were measured after 48 hours of incubation. Susceptibility of the strains to posaconazole, voriconazole, and caspofungin was determined using gradient diffusion strips (Liofilchem) in RPMI agar plates. Results were read after 24 or 48 hours of growth at 37°C. All the experiments were performed in biological triplicates. The procedures developed by Tilburn and colleagues, or Szewczyk and colleagues, were used to obtain and transform protoplasts of A. nidulans, and select transformants on selective (lacking uridine and uracil, on the one hand, or pyridoxine, on the other hand) regeneration medium (regeneration minimal medium [RMM]: AMM supplemen ted with 1 M sucrose) (73, 74). Protoplasts of A. fumigatus strain ΔakuB-pyrG+ were obtained and transformed using the procedure described by Yelton and colleagues (75). AfflpA (AFUB_007390, Afu1g07020), Afstk47 (AFUB_051670; Afu5g03160), and AfflpB (AFUB_027820, Afu2g12130) were deleted using a CRISPR-Cas9 gene editing procedure previously described by us (see below how DNA cassettes were generated) (76). Selection of transformants was carried out based on hygromycin resistance and correct integration of the DNA constructs was confirmed by diagnostic-PCR using the oligo nucleotide pairs shown in Table S4. The control ΔakuB-pyrG + strain was constructed previously by replacing the mutant pyrG locus of the A. fumigatus KU80ΔpyrG strain with the functional A. parasiticus pyrG homolog (71). For fluorescence microscopy analyses, conidiospores of the strains of interest were incubated for approximately 18 hours at 25°C in Ibidi µ-Dishes or Falcon multi-well plates containing, respectively, 2.0 or 1.5 mL of supplemented watch minimal medium (WMM) (77). Nuclei were visualized by adding to the cultures two drops of NucBlueTM live cell Research Article mBio November/December2023 Volume 14 Issue 6 10.1128/mbio.02452-23 21 Downloaded from https://journals.asm.org/journal/mbio on 03 May 2024 by 158.227.89.37. stain readyProbesTM reagent (DAPI, Invitrogen) plus triton X100 (PanReac AppliChem, 0.05%) and by incubating 15–20 minutes at room temperature (RT) before fluorescence microscopy analysis. To quantify conidia production in A. nidulans, asexual spores were point inoculated and cultured for 72 hours at 37°C. Conidia were collected in Tween 20 (0.02%), diluted (if necessary), and the total amount of conidia was determined using a Thoma cell counter. The number of conidia was divided by the area of the colony. Microsoft Excel was used to determine statistically significant differences in conidia production among strains and to draw the corresponding column bar graphs. Radii of A. nidulans colonies and their radial growth rates were determined by measuring and by comparing the diameter of the colonies after 48, 72, and 96 hours of culture in adequately supplemented AMM media. Conidia of A. fumigatus were harvested from 5-day-old cultures grown on AMM. Colony morphologies of A. fumigatus mutants were assessed by spotting 5 µL containing 1,000 conidia onto the center of AMM plates (5.5 cm) and incubation for 72 hours at 37°C. Diameter of colonies were measured after 24, 48, and 72 hours of culture at 37°C. The phenotypes of ΔflpA, Δstk47, and ΔflpB mutants of A. nidulans under nitrogen starvation conditions were assessed as follows (25, 26, 41, 78). First, 106 conidia per mL of the mutants and their parental wild-type strain were inoculated in Erlenmeyer flasks filled with 25 mL of adequately supplemented liquid AMM. After 18 hours of culture at 37°C and 200 rpm, mycelia were filtered using Miracloth paper and inoculated in liquid AMM lacking a nitrogen source. After additional 20 hours of culture, phenotypes were analyzed using an Imaging Source DFK23UP031 digital camera coupled to a Nikon Eclipse E100 microscope. Hypothetic germination defects of single-null mutants were assessed by inoculating 5 × 104 (A. nidulans) conidia per mL in Erlenmeyer flasks filled with adequately supple mented liquid AMM, and by culturing them at 37°C and 200 rpm. For the evaluation of germination rates in A. fumigatus, 10,000 conidia from each strain were inoculated into AMM broth, poured over sterile coverslips, and incubated at 37°C in static conditions. After 6 or 8 (A. nidulans), or 8, 10, or 12 (A. fumigatus) hours of culture, the germinated conidia were counted from a minimum number of 100 conidia of each strain and time point. At least two replicates were analyzed per strain. The corresponding graphs were drawn with Microsoft Excel or GraphPad Prism (v.9.2.0). Distribution of septa in A. nidulans hyphae was analyzed using fluorescence brightener 28 (CFW). Briefly, conidia of the strains of interest were inoculated in Falcon multi-well plates containing adequately supplemented WMM and incubated for approximately 18 hours at room temperature. Then, the culture medium was replaced by fresh WMM containing 0.01% CFW. Samples were incubated at room temperature for 5 minutes. Finally, the culture medium was replaced twice (plus additional 5 minutes of incubation) before observation using a Nikon Eclipse Ci manual upright microscope (see below; phase contrast and GFP images were processed). Septum-to-septum distances (a minimum of 90 for each strain in three biological replicates) were measured using ImageJ (https://imagej.nih.gov/ij/), and the corresponding scatter plot was generated using GraphPad Prism (v.5.01). Generation of DNA cassettes for transformation The fusion-PCR technique was used to generate DNA constructs for transformation of A. nidulans protoplasts (79). For 3′-end gene tagging (wild-type or mutant versions), three DNA fragments were amplified and fused: first, 1.5 kb of the 3′-end of the coding region was amplified using oligonucleotides GSP1 and GSP2 (Table S4); second, gfp, mCherry, or ha3x tags plus the selection marker (pyrGAfum or pyroAAfum of A. fumigatus) were amplified using oligonucleotides GFP1 and GFP2 (2.6 or 1.9 kb, respectively, when gfp::pyrGAfum, gfp::pyroAAfum, or mCherry::pyroAAfum, and ha3x::pyrGAfum were amplified); third, 1.5 kb of the 3′-UTR region of the gene (oligonucleotides GSP3 and GSP4). The fusion-PCR reaction was carried out with an aliquot of each fragment and oligonucleotides GSP1 and GSP4. Research Article mBio November/December2023 Volume 14 Issue 6 10.1128/mbio.02452-23 22 Downloaded from https://journals.asm.org/journal/mbio on 03 May 2024 by 158.227.89.37. To generate the constructs for A. nidulans gene deletion, the following three fragments were amplified and fused. First, 1.5 kb of the promoter region of the gene was amplified using oligonucleotides PP1 and PP2. Second, the selection markers pyrGAfum or pyroAAfum were amplified with oligonucleotides SMP1 and GFP2. Third, 1.5 kb of the 3′-UTR region of the gene was amplified with oligonucleotides GSP3 and GSP4. The fusion-PCR reaction was carried out with oligonucleotides PP1 and GSP4. To constitu tively express FlpA::GFP or FlpA::HA3x chimeras under the control of the “mini” version of the promoter gpdAp (42), five fragments were amplified and fused (27, 80): first, 1.5 kb of the promoter using oligonucleotides PP1 and PP2′-ATG; second, the gpdAp promoter with oligonucleotides gpdAUp and gpdADw; third, the coding region of flpA was amplified with oligonucleotides geneSP and GSP2; fourth, the tag plus the selection marker (GFP1 and GFP2); and finally, 1.5 kb of the 3′-UTR region with oligonucleotides GSP3 and GSP4. The fusion of all those five fragments was done using oligonucleoti des PP1 and GSP4. After transformation of A. nidulans protoplasts and genomic DNA extraction, correct recombination of the constructs was confirmed by diagnostic PCR using oligonucleotides sPP1 and sGSP4. For A. fumigatus genetic engineering, hygromycin resistance repair templates, containing 40-bp micro-homology regions at both 5′- and 3′-ends of Afstk47, AfflpA or AfflpB, were amplified from plasmid pJMR2 (81) and utilized in a CRISPR/Cas9-mediated transformation to generate complete gene deletions (82). Analysis of cytokine release in THP-1 cells Measurements of IL1-β released by THP-1 cells was carried out following procedures described previously (44, 83, 84). THP-1 cells were cultured in RPMI-1640 medium containing HEPES (25 mM) supplemented with heat-inactivated fetal bovine serum (10%), penicillin and streptomycin (100 U/mL), and 100 µg/mL normocin, and assayed for viability by exclusionary trypan blue straining. Plating was carried out at a density of 5 × 104 cells per well (96-well plates) in the same medium but lacking normocin. Phorbol 12-myristate 13-acetate at a final concentration of 100 nM was added, and cells were incubated for 24 hours to differentiate to a macrophage phenotype. Supernatants were discarded and replaced with 180 µL of fresh RPMI (without phenol red) and 20 µL of distilled water containing 5 × 105 A. fumigatus conidia (multiplicity of infection, 10:1) of the single nulls or the reference strain. Cells were co-cultured for 16 hours at 37°C and 95% humidity. After incubation time, we collected the supernatants and measured the IL-1β concentrations using IL-1β enzyme-linked immunosorbent assay (ELISA) kits (Invitrogen) and following manufacturer instructions. Virulence assays in A. fumigatus All animal experiments were carried out according to the protocols approved by the University of Tennessee Institutional Animal Care and Use Committee. Groups of 10, 6-week-old female CD-1 mice (Charles Rivers Laboratories) were immunosuppressed with 150 mg/kg of cyclophosphamide (neutropenic model) on day −3 and subsequently every third day starting again on day +1 (85) and an additional subcutaneous injection of triamcinolone acetonide on day −1. On the day of the infection, mice were slightly anesthetized with 3.5% isoflurane and inoculated intranasally with 106 freshly harvested conidia contained in 35 µL of pyrogen-free saline solution. Mice were monitored at least twice a day and were given food and water ad libitum, and those animals showing severe signs of infection were humanely euthanized by anoxia with CO2. Mortality was monitored for 12 days. GraphPad Prism v.9.2.0 was used to plot survival data on a Kaplan-Meier curve and to carry out log-rank (Mantel-Cox) tests. Fluorescence microscopy Subcellular localization of FlpA, Stk47, and FlpB was analyzed using a Zeiss Axio Observer Z1 inverted microscope as previously described (27, 86) or a Nikon Eclipse Research Article mBio November/December2023 Volume 14 Issue 6 10.1128/mbio.02452-23 23 Downloaded from https://journals.asm.org/journal/mbio on 03 May 2024 by 158.227.89.37. Ci manual upright microscope. The former was equipped with a ×63 Plan Apochromat 1.4 oil immersion lens, and filters 38 (excitation at 470 nm and emission at 525 nm) and 43 (excitation at 545 nm and emission at 605 nm). The latter included 40× and 100× (oil) Nikon CFI Plan Fluor lenses, and Semrock filters DAPI-5060C (excitation at 377 nm and emission at 447 nm), GFP-4050B (excitation at 466 nm and emission at 525 nm), and mCherry-C (excitation at 562 nm and emission at 641 nm). ImageJ (https:// imagej.nih.gov/ij/) (U.S. National Institutes of Health, Bethesda, MA, USA) was used to process fluorescence microscopy and phase contrast images. Protein extraction and immunodetection For protein extraction, conidia of the strains of interest were collected in Tween 20 (4 mL, 0.02%) and washed twice by centrifugation at 4,000 rpm for 10 minutes. By default, inocula (106 conidia/mL) were cultured in AFM (or AMM for phosphopeptide enrichment analyses) for 15 hours at 37°C and 200 rpm (87). Mycelia were filtered, frozen, lyophilized, homogenized, and resuspended in 1 mL per sample of A-40 buffer (25 mM HEPES, pH = 7.5, 50 mM KCl, 5 mM MgCl2, 0.1 M EDTA, 10% glycerol, and 0.5 mM dithiothreitol (DTT), supplemented with a protease inhibitor cocktail from Roche) or Trap buffer (10 mM HEPES, pH = 7.5, 1 mM EDTA, 0.1% NP-40, 150 mM NaCl, 0.5 mM DTT, plus a protease inhibitor cocktail from Roche; used in pull-down experiments with the GFP-Trap resin of Proteintech). After incubation at 4°C during 90 minutes with gentle end-over-end mixing, samples were centrifuged at 4°C and 13,000 rpm for 30 minutes. Supernatants were transferred to new 1.5 mL tubes, and protein concentrations were determined by the Bradford assay. Two hundred micrograms of protein was precipitated per sample with trichloroacetic acid (TCA) and purified with ethanol/ether 1:1 and 1:3 mixes, respectively. Finally, precipitates were resuspended in 80 µL of SDS-PAGE loading buffer (62.5 mM Tris-HCl, pH = 6.8, 2% SDS, 5% β-mercaptoethanol, 6 M urea, and 0.05% bromophenol blue), and the integrity of the samples was assessed by polyacrylamide gel electrophore sis. Alternatively, the procedure described by Hervás-Aguilar and Peñalva was followed (88). Mycelial samples (approximately 6 mg of lyophilized powder) were lysed using 1 mL of alkaline lysis buffer (0.2 M NaOH and 0.2% β-mercaptoethanol), and proteins were precipitated by adding 7.5% TCA. Samples were centrifuged and the supernatants were discarded, resuspending the pellets in 100 µL Tris-base (1 M) and 200 µL of SDS-PAGE loading buffer. Aliquots of 10 µL were then loaded and separated in 10% polyacrylamide gels. Samples of a hundred micrograms of protein from the crude extracts of the strains of interest were incubated with λPP (total volume of 100 µL per reaction; New England Biolabs) for 20 minutes at 30°C (89). Sodium orthovanadate (10 mM) was used as a λPP inhibitor when necessary. After incubation, proteins were precipitated as described above and resuspended in 50 µL of SDS-PAGE loading buffer. The integrity of samples was checked by polyacrylamide gel electrophoresis. For immunodetection, protein samples were separated in Any kD Mini-PROTEAN TGX precast protein gels (Bio-Rad; stain-free gels were also used in specific experiments) before electro-transference (Trans-blot Turbo transfer system, Bio-Rad) to nitrocellulose filters. GFPor HA3x-tagged proteins were detected using α-GFP (1:1,000; Roche) or α-HA3x (1:1,000; Santa Cruz Biotechnology) mouse monoclonal antibody cocktails. Peroxidase-conjugated α-mouse (1:2,500; Jackson Immuno-Research Laboratories) was used as secondary antibody. Peroxidase activity was induced using the Clarity Western ECL substrate (Bio-Rad). Chemiluminescence was detected using a Chemidoc + XRS system (Bio-Rad). Phosphopeptide enrichment Aliquots of approximately 300 µg of the crude protein extracts were precipitated with acetone and resuspended in a mixture of urea (8 M) and ammonium bicarbonate Research Article mBio November/December2023 Volume 14 Issue 6 10.1128/mbio.02452-23 24 Downloaded from https://journals.asm.org/journal/mbio on 03 May 2024 by 158.227.89.37. (50 mM). Proteins were reduced with dithiotreitol (10 mM) for 30 minutes at RT, alkylated with iodoacetamide (50 mM) in the dark for 30 minutes (RT), and digested with 15 µg of trypsin overnight at 37°C. C18 SEP-PAK columns were used for desalting of peptide mixtures. Samples were dried at RT using a speedvac concentrator. For TiO2 phosphopeptide enrichment, slurry of Titansphere TiO (5 µm) beads (cat. no.: 5020–75,000) was prepared in buffer C (300 mg/mL lactic acid; 53% acetonitrile, ACN; 0.07% trifluoroacetic acid, TFA) at 25 mg/mL. Peptides pellets were resuspended in 600 µL of buffer C, and 72 µL of the titanium slurry was added to each sample. After incubation for 15–30 minutes at RT with end-over-end rotation, samples were centrifuged for spinning down the beads. Pellets were resuspended in 150 µL of buffer C and transferred of top of the C8 disks stage tips (90). Buffer C was removed at moderate speed (10–30 µL/min) using a syringe, and another 150 µL of buffer C was added on top of the C8/TiO2 stage tip and passed it through at the same speed for washing. Additional washing was carried out with 100 µL of buffer B (80% ACN, 0.1% TFA) at the same conditions. For sample elution, 2 × 50 µL of buffer D (0.5% NH4OH) was applied, collecting each eluate in 100 µL of TFA (2%). Samples were dried in a speedvac concentrator and reconstituted in TFA (0.1%) for desalting using C 18 Zip-Tip columns. For mass spectrometry analysis, fractions of 1/10 from each phosphopeptideenriched sample were analyzed by nano-LC-MS/MS. Peptides were trapped onto a AcclaimPepMap 100 C18 (2 cm) precolumn (ThermoScientific), eluted onto a Acclaim PepMap 100 C18 column (inner diameter 75 µm, 50 cm long, 2-µm particle size; ThermoScientific) and separated using a 180-minute gradient (0%–21% buffer B for 60 minutes, 21%–35% buffer B for 100 minutes, 95% buffer B for 10 minutes, and 0% buffer B for 10 minutes; buffer A: 0.1% formic acid, 2% ACN; buffer B: 0.1% formic acid in ACN) at a flow-rate of 250 nL/min on a nanoEasy HPLC (Proxeon) coupled to a nanoelectrospray ion source (ThermoScientific). A Q-Exactive mass spectrometer (ThermoScientific) in the positive ion mode was used for mass spectra acquisition. Full-scan MS spectra (m/z 400–1,500) were acquired in the Orbitrap at a resolution of 70,000 at m/z 200, and the 10 most intense ions were selected for tanden mass spectrometry (MS/MS). Fragmentation was carried out with a normalized collision energy of 27 eV. MS/MS scans were acquired with a starting mass of m/z 200; automatic gain control (AGC) target was 2e5, resolution of 17,500 (at m/z 200), intensity threshold of 8e3, isolation window of 2.0 m/z units, and maximum IT was 100 ms. Charge state screening was enabled to reject unassigned, singly charged, and equal or more than seven protonated ions. A dynamic exclusion time of 20 seconds was used to discriminate against previously selected ions. Mass spectra *.raw files were searched against the A. nidulans fasta database (10,720 protein entries) using Sequest search engine through Proteome Discoverer (v.1.4.0.288, ThermoScientific). Search parameters included a maximum of two missed cleavages allowed, carbamidomethylation of Cys residues as a fixed modification and N-terminal acetylation, C-terminal oxidation, and Ser, Thr, and Tyr phosphorylation as variable modifications. Precursor and fragment mass tolerance was set to 10 ppm and 0.02 Da, respectively. Node 3 phosphoRS was used as scoring algorithm. This node evaluates statistical confidence of location of phosphorylation sites. Identified peptides were validated using Percolator algorithm with a q-value threshold of ≤0.01 (91). Pull-down assays Pierce Anti-HA agarose beads (50 µL per sample, ThermoScientific) were centrifuged 5–10 seconds at 12,000 g and washed with a 1:1 vol of Tris buffered saline (TBS, ThermoScientific) by centrifugation at the same conditions. After discarding the liquid, 2 mg of crude protein lysate was added. Samples were incubated for 90 minutes at 4°C with gentle end-over-end mixing. After pelleting the resin with a 5to 10-second pulse at 12,000 g, supernatants were collected, and 200 mg of protein was precipitated with TCA. This fraction was saved for analysis of binding efficiency (non-retained, NR, Research Article mBio November/December2023 Volume 14 Issue 6 10.1128/mbio.02452-23 25 Downloaded from https://journals.asm.org/journal/mbio on 03 May 2024 by 158.227.89.37.