If not otherwise stated, the A. fumigatus strain used was A1160, termed wild type (wt) here, and derived mutant strains Δmac1 (lacking Mac1), ΔaceA (lacking Ace1), and ΔcrpA (lacking CrpA), which have been described previously (Cai et al., 2017; Cai et al., 2018). Furthermore, A. fumigatus strains Afs35 (a Ku70 lacking a derivative of the clinical isolate D141) (Krappmann et al., 2006), the clinical isolate Af293 (Nierman et al., 2005), Afs77 (a Ku70 lacking a derivative of the clinical isolate ATCC46645), and the Afs77-derived ΔcccA (Gsaller et al., 2012) and ΔsidA (Schrettl et al., 2004) mutant strains were used. The strains were grown at 37°C either on solid complex media (CM) or on/in solid/liquid Aspergillus minimal media (AMM) according to Pontecorvo et al. (1953) with 0.03 mM FeSO₄ as Fe source (unless otherwise noted), 1% glucose as carbon source, and the nitrogen source described in the respective experiment. The Cu (CuSO₄) concentration used is described in the respective experiments. For limitation of Cu or Fe, addition of the respective metal was omitted. CM contained 1% glucose, 2 g/l peptone (Carl Roth GmbH + Co. KG, Karlsruhe, S.T.U, Germany), 1 g/l casamino acids (Sigma-Aldrich Chemical Co., St. Louis, MO, USA), 1 g/l yeast extract (Lab M Limited, Bury, Lancs, UK), and trace elements according to Pontecorvo et al. (1953) but without Cu and Fe. Amino acid (AA) supplements are described in the respective experiment; if not noted otherwise, AAs were used in the L-configuration and were not denominated “L.” The nitrogen sources used were 20 mM ammonium (ammonium tartrate dibasic, (NH₄)₂C₄H₄O₆), 20 mM Gln, 20 mM nitrate (sodium nitrate, NaNO₃), 20 mM nitrite (sodium nitrite, NaNO₂), and 20 mM urea. For plate growth assays, 1 × 10⁴ conidia were point-inoculated; AMM plates were incubated for 48 h at 37°C, and CM plates were incubated for 30 h at 37°C. For culturing in liquid medium, 100 ml AMM in 0.5-l Erlenmeyer flasks inoculated with 10⁶/ml conidia was shaken at 200 rpm at 37°C for 24 h. Bovine serum albumin and bovine pancreatic RNase A were from Sigma-Aldrich Chemical Co., St. Louis, MO, USA.

The mycelia from liquid cultures were harvested by filtration, washed with distilled water, and freeze-dried to determine the dry weight of the biomass. For determination of the total cellular Fe content, 50 mg of freeze-dried mycelia was decomposed in closed polytetrafluorethylene vessels containing 2 ml of HNO₃ and 0.5 ml of hydrogen peroxide using a high-performance microwave digestion unit (MARS 6, CEM Microwave Technology, Buckingham, UK). Appropriate dilutions were made with distilled water, and the total contents in Cu and Fe were determined by graphite furnace atomic absorption spectrometry (Zeeman GF95Z M6 AAS, Thermo Fisher Scientific, Waltham, MA, USA) according to standard methods.

Total RNA was isolated according to the TRI Reagent (Sigma-Aldrich) method using peqGOLD PhaseTrap reaction tubes (PEQLAB, Erlangen, Germany). Formaldehyde-containing agarose gels were used to separate 10 μg of total RNA before blotting onto Hybond-N+ membranes (Amersham Biosciences, Amersham, UK) and hybridization with digoxigenin (Roche Diagnostics GmbH, Mannheim, Germany)-labeled probes. The digoxigenin-labeled hybridization probes used in this study were generated by PCR using primers 5′-ATGCGAACGAACATTGTCCC and 5′-CCAGCGGAAATGAGAAGATTCA for crpA (AfuA_3G12740), 5′-ATGGATCATATGAGCCAC and 5′-CTACCCGCAGCATTTG for ctrC (AfuA_2G03730), 5′-AAGCCGAGAAAAAGGGGG and 5′-AACCCGATGAA GCCCAG for mirB (AfuA_3G03640), 5′-ATATGTTCCTCGTGCCGTTC and 5′-CCTCAGTGAACTCCATCTC for tubA (Afu1g10910), 5′-GGAGCAGCTCGATCGCCAT and 5′-AGTGTATGCCACCATCGTTG for atm1 (AFUA_6G12870), and 5′-CCCGTCTTCCACCTGCTG and 5′-GCATCAACAGCGCTGACCTT for atm1 (AFUA_4G04318).

We observed that the Cu resistance of A. fumigatus wt is higher with 20 mM Gln compared to 20 mM nitrate as nitrogen source ( Figure 1A ): A. fumigatus A1160/Δku80, which is termed wild type (wt) here, displayed similar radial growth on solid media in the presence of 0.1 and 0.01 mM Cu as well as under Cu limitation with Gln as nitrogen source, while its radial growth decreased with increasing Cu concentration with nitrate as nitrogen source. Notably, Cu limitation is reflected by the yellow color of conidia, as biosynthesis of the green conidial pigment is dependent on a Cu-requiring laccase (Tsai et al., 1999). These results suggested that either glutamine protects against Cu toxicity and/or that nitrate enhances Cu toxicity. To test the latter hypotheses, we analyzed the impact of different nitrogen sources on Cu resistance of A. fumigatus ( Figure 1B ). To increase the sensitivity of the growth assay, we used not only A. fumigatus wt but also mutant strains that have increased susceptibility to Cu-mediated toxicity due to lacking transcriptional activation of Cu detoxification (ΔaceA, lacking the transcription factor AceA) or cellular Cu export (ΔcrpA, lacking the Cu exporter CrpA), respectively (Raffa et al., 2019). The ΔaceA and ΔcrpA mutant strains were able to grow in the presence of 0.1 mM Cu when using the nitrogen source Gln or Gln in combination with nitrate or nitrite. In contrast, the ΔaceA and ΔcrpA mutant strains were unable to grow in the presence of 0.1 mM Cu with nitrate, nitrite, ammonium, or urea as nitrogen source. Moreover, ΔaceA and ΔcrpA displayed significantly decreased growth compared to wt in the presence of 0.01 mM Cu in the absence of Gln. Taken together, these data indicate that the presence of Gln protects against Cu toxicity. In contrast to the other nitrogen sources used, the growth of wt decreased with nitrate and nitrite in the presence of 0.1 mM Cu compared to 0.01 mM Cu, which indicates that nitrate and nitrite might decrease Cu resistance.

As Gln is an AA and as some AAs are well known for providing the ligands for metal binding in proteins (Cao et al., 2017), we analyzed the impact of supplementation with different proteinogenic AAs in a concentration of 1 mM on Cu resistance of A. fumigatus ( Figure 2 ). Therefore, A. fumigatus wt and ΔcrpA strains were grown on AMM plates containing 0.1 mM Cu and ammonium as nitrogen source, which impedes the growth of ΔcrpA ( Figure 1 ). Only supplementation with 1 mM asparagine (Asn), aspartate (Asp), histidine (His), serine (Ser), threonine (Thr), or tyrosine (Tyr) rescued the growth of the ΔcrpA mutant strain, indicating that these six AAs have a higher Cu-detoxifying capacity compared to the other tested AAs such as Gln ( Figure 2A ). In order to identify possible differences in the Cu detoxification activity of these six AAs, we analyzed the growth of A. fumigatus wt and ΔcrpA strains on media containing ammonium as nitrogen source, 1 mM of the respective AA, and different Cu concentrations up to 1 mM ( Figure 2B ). Supplementation with Asn, Asp, Ser, Thr, and Tyr rescued the growth of the ΔcrpA mutant strain up to 0.1 mM Cu and with His up to 0.5 mM Cu demonstrating that His has the strongest protecting activity against Cu. Notably, the growth of wt was significantly decreased in the presence of 1 mM Cu compared to low Cu concentrations and exclusively His supplementation improved the radial growth of wt at this Cu concentration ( Figure 2B ).

In a next step, we compare the Cu-detoxifying activity of Gln and His. Determination of the AA concentration that is required to permit the growth of ΔcrpA in AMM with ammonium as nitrogen source and 0.1 mM Cu yielded 5 mM Gln and 0.2 mM His ( Figure 3A ), which underlines the difference of a low- and high-protecting AA. Moreover, determination of the Cu detoxification capacity of 20 mM Gln, the standard nitrogen source used in our laboratory, yielded 0.5 mM Cu ( Figure 3B ).

Taken together, His showed the highest Cu-detoxifying capacity with an about twofold molar excess, i.e., supplementation with 1 mM His allowed the growth of ΔcrpA in the presence of a maximum of 0.5 mM Cu ( Figure 2B ) and 0.2 mM His permitted the growth of ΔcrpA in the presence of 0.1 mM Cu, respectively ( Figure 3A ). Asn, Asp, Ser, Thr, and Tyr detoxified Cu in an approximately 10-fold molar excess, i.e., 1 mM supplementation with either of these AAs allowed the growth of ΔcrpA in the presence of a maximum of 0.1 mM Cu ( Figure 2B ). Gln detoxified Cu in an approximately 40–50-fold molar excess, i.e., 5 mM Gln was required to allow the growth of ΔcrpA in the presence of 0.1 mM Cu ( Figure 3A ) and 20 mM Gln allowed the growth of ΔcrpA in the presence of a maximum of 0.5 mM Cu ( Figure 3B ).

Due to the Cu-detoxifying activity of AAs, we analyzed the Cu resistance of A. fumigatus wt and ΔcrpA strains in complex medium (CM), which contains 0.2% peptone, 0.1% casamino acids, and 0.1% yeast extract. This medium allowed the growth of ΔcrpA in the presence of a maximum of 0.5 mM Cu ( Figure 4 ), which demonstrates its Cu-detoxifying activity. Each of the ingredients added to AMM with ammonium as nitrogen source improved the Cu resistance of ΔcrpA, i.e., it allowed the growth of this mutant in the presence of up to 0.2 mM Cu ( Figure 4 ). As these CM ingredients are not chemically defined components including AAs and peptides, we assayed in a next step the impact of a defined protein using bovine serum albumin (BSA). BSA supplementation to a final concentration of 0.1 mM also increased the Cu resistance of ΔcrpA and allowed the growth of this mutant in the presence of up to 0.2 mM Cu ( Figure 4 ). BSA consists of 583 AA residues, and consequently 0.1 mM BSA corresponds to 58.3 mM peptidic AAs. Therefore, at the level of AA molarity, BSA has a lower protecting activity compared to free AAs as 20 mM of the weakly protecting AA Gln allowed the growth of ΔcrpA in the presence of up to 0.5 mM Cu ( Figure 3B ). Similar to BSA, supplementation with 0.47 mM RNase A, which consists of 124 AA residues and therefore corresponds to 0.1 mM BSA with respect to peptidic AA, allowed the growth of ΔcrpA in the presence of up to 0.2 mM Cu, which underlines that the Cu resistance-promoting effect is a protein effect rather than being specific on the type of protein. Notably, the influence of proteins on Cu resistance might be influenced by proteolytic degradation of the proteins, but in the experiments conducted, proteolysis is expected to be low due to the presence of primary nitrogen (ammonium) and carbon (glucose) sources, which usually repress the expression of protease-encoding genes (Bergmann et al., 2009; Shemesh et al., 2017).

To investigate the mechanism of how His supplementation protects A. fumigatus ΔcrpA against Cu toxicity, we compared the protecting activity of His (L-His) and its non-proteinogenic stereoisomer D-His. As shown in Figure 5A , supplementation with L- and D-configurations of His resulted in similar Cu resistance, which indicates that the mode of protection conferred by His supplementation does not involve metabolization of His. Therefore, we hypothesized that His protects against Cu toxicity by extracellular chelation of Cu. In line, His supplementation was found to hamper the growth of the A. fumigatus Δmac1 mutant strain in a concentration-dependent manner under Cu-limiting conditions as well as in the presence of 0.02 mM Cu ( Figure 5B ). These data indicate that the presence of His negatively affects Cu uptake in the absence of high-affinity Cu uptake as Mac1 is essential for the activation of high-affinity Cu uptake (Raffa et al., 2019). In agreement, the growth of wt that is capable of high-affinity Cu uptake was not affected by His supplementation ( Figure 5B ).

To further investigate the impact of AA supplementation on metal homeostasis, we analyzed the cellular contents in Cu and Fe ( Table 1 ). Therefore, A. fumigatus wt and ΔcrpA strains were grown in liquid AMM containing 0.005 mM Cu and 20 mM ammonium as nitrogen source without AAs or supplemented with either 5 mM Gln or 1 mM His. The biomass production of wt was similar in all three growth media. In agreement with the Cu susceptibility of ΔcrpA and the different Cu-detoxifying capacities of His and Gln, biomass of ΔcrpA was only 22% of that of the wt in AMM without AAs, increased to 74% with Gln supplementation, and reached the wt level with His supplementation ( Table 1 ). The cellular Cu content of wt was similar in AMM without AAs and with Gln but was approximately halved in the presence of His. In agreement with defective Cu export, the cellular Cu content of ΔcrpA was increased about five-fold in the absence of AAs. Supplementation with His and Gln decreased the cellular Cu content to about that of the wt grown under the same condition ( Table 1 ). Taken together, these data indicate that supplementation with AAs such as His and Gln protects A. fumigatus against Cu toxicity by impeding its cellular uptake, most likely via complexation of this metal. Remarkably, the about five-fold increase in cellular Cu of ΔcrpA was accompanied by an approximately 10-fold increased Fe content ( Table 1 ), which indicates a link between Cu and Fe homeostasis.

The experiments discussed above demonstrate the protective effects of AAs against Cu toxicity mainly for the highly Cu-susceptible ΔcrpA mutant strain. As shown in Figure 6A , His supplementation also increases the radial growth of different A. fumigatus strains including the genetic background strain of the ΔcrpA mutant, A1160, termed wt here. To further investigate the role of His in protection against Cu toxicity, we analyzed the expression of Cu-detoxifying ABC transporter-encoding crpA and high-affinity Cu uptake permease ctrC by Northern blot analysis ( Figure 6B ). During Cu sufficiency, neither crpA nor ctrC was expressed. A short-term confrontation (45 min) of such mycelia with 0.2 mM Cu caused an induction of crpA reflecting Cu detoxification. In contrast, a short-term confrontation with His-complexed Cu (the same amount of Cu was preincubated with a 10-fold excess of His before addition to the mycelia) did not induce crpA. The expression of ctrC was detected in neither of these mycelia, which confirms repression of high-affinity Cu uptake in agreement with Cu sufficiency. These data indicate that complexation of Cu by His blocks Cu uptake in the absence of high-affinity Cu uptake. During Cu limitation, the expression of crpA was repressed and ctrC was induced ( Figure 6B ). A short-term confrontation with Cu caused the induction of crpA and a decrease in the ctrC transcript level. In comparison, a short-term confrontation with His-complexed Cu induced crpA to the same degree as non-complexed Cu but repressed ctrC significantly more strongly. The similar induction of crpA by His-complexed and non-complexed Cu indicates that His complexation does not block Cu uptake by high-affinity systems. Moreover, the stronger repression of ctrC by His-complexed Cu compared to non-complexed Cu might indicate an even higher bioavailability for high-affinity uptake systems of His-complexed Cu compared to Cu alone. In agreement with complexation of Cu by AA, we observed that AA form blue-colored complexes with Cu; Figure 6C shows exemplary His, Thr, Asn, and Gln. His displayed the most intensive color formation with Cu, which is consistent with the strongest protecting activity observed. Notably, there was no significant difference in color formation between Thr, Asn, and Gln, which display different protecting activities at a lower level (see above).

Based on the concomitant increase in cellular Cu and Fe contents in A. fumigatus ΔcrpA ( Table 1 ), we hypothesized that Fe is required to counteract Cu toxicity. Indeed, we found that the Cu resistance of A. fumigatus ΔcrpA increases with increasing Fe availability ( Figure 7A ); i.e., in the presence of 0.02 mM Cu, the radial growth of ΔcrpA was lowest during Fe limitation and increased with the degree of Fe supplementation. Furthermore, supplementation with ferricrocin-chelated Fe improved Cu resistance ( Figure 7A ). As Fe chelated by siderophores such as ferricrocin are taken up exclusively by siderophore-specific transporters (Aguiar et al., 2021; Misslinger et al., 2021), these data exclude the possibility that the positive impact of Fe supplementation on Cu resistance of ΔcrpA is the sole consequence of competition of Fe with Cu for uptake by low-affinity metal transporters. In line, a ΔsidA mutant, which lacks siderophore biosynthesis and consequently displays decreased Fe acquisition (Schrettl et al., 2004; Misslinger et al., 2021), as shown by lack of growth under low Fe availability, shows decreased Cu resistance when grown with either 0.1 mM Fe or 0.001 mM ferricrocin ( Figure 7B ). The ferricrocin experiment again excludes the possibility that the effects seen are based on competition of Cu and Fe for uptake. Moreover, a short-term confrontation with Cu was found to induce a higher crpA expression during Fe starvation compared to Fe sufficiency ( Figure 7C ), which supports higher Cu toxicity under Fe limitation. To further investigate the role of Fe in Cu resistance, we analyzed the expression of cmtA (also termed crd2) and atm1. cmtA encodes a putative metallothionein previously implicated in the Cu resistance of A. fumigatus (Cai et al., 2018); atm1 encodes a mitochondrial ABC transporter that links mitochondrial and cytosolic Fe–sulfur cluster biosynthesis and that has recently been shown to play a role in Cu toxicity in Cryptococcus neoformans and S. cerevisiae (Garcia-Santamarina et al., 2017). The expression of cmtA was found to be repressed under Fe starvation compared to Fe sufficiency and showed a slight upregulation in response to a short-term confrontation with Cu during Fe sufficiency ( Figure 7C ). In contrast, atm1 displayed a downregulation during Fe sufficiency compared to Fe starvation without a response to a short-term confrontation with Cu ( Figure 7C ).

To investigate if there is also an effect of Cu availability on Fe resistance of A. fumigatus, we employed the ΔcccA mutant that shows increased Fe toxicity due to the lack of a vacuolar transporter mediating vacuolar Fe deposition (Gsaller et al., 2012). As shown in Figure 8 , Fe resistance of the ΔcccA mutant increased with increasing Cu availability (8 mM Fe). Under low Fe availability (0.03 mM Fe), the ΔcccA mutant lacked a growth defect and displayed wt-like Cu susceptibility. Taken together, these data demonstrate the importance of metal homeostasis, i.e., balanced cellular metal contents.