P. murina organisms for these studies were obtained from 6 to 8-week old Bagg Albino (BALB/c) mice (Charles River Wilmington, MA). The mice were handled in strict accordance with good animal practice, as defined by the Institutional Animal Care and Use Committee at the Veterans Affairs Medical Center, Cincinnati, OH. To safeguard against environmental exposure of P. murina and other microbes, mice were housed under barrier conditions with autoclaved food, acidified water, and bedding in sterilized shoebox cages equipped with sterile microfilter lids.

Mice were immunosuppressed by adding dexamethasone (4 mg/L) to acidified drinking water and infected with P. murina organisms isolated from previously infected mice and stored in liquid nitrogen. A dose of 2 × 10⁶ organisms / 50 μL was instilled by intranasal inoculation under anesthesia. The mice were continued on immunosuppression for 6 weeks and then euthanized by CO₂ for isolation of organisms.

Lungs were removed, homogenized (gentleMACS™ Dissociator, Miltenyi Biotec, Auburn, CA), and filtered through a 40 μm pore mesh, and cells were recovered by centrifugation at 3400 rpm for 10 min. Host cells were removed by a 2^nd centrifugation at 250 rpm for 5 min. Slides were then prepared to enumerate organisms for the downstream CuSO₄ treatment, as previously described (Cushion et al., 2010).

Two experiments were conducted to evaluate the impact of copper(II) sulfate on P. murina. Experiment 1, a short-term assessment, focused on differential gene expression in response to CuSO₄ exposure. Experiment 2, a long-term study, aimed to determine the viability of P. murina under sustained CuSO₄ exposure.

During the isolation process, P. murina were purified from lung cells by washing at least 3 times with phosphate buffered saline (PBS) as well as a short incubation period with 0.85% aqueous ammonium chloride, after which they were washed again with PBS. After rinsing with PBS, the cells were centrifuged to form a pellet, reconstituted in PBS for the next washing.

P. murina cultures were prepared at 2.5 × 10⁶ cells/mL using Roswell Park Memorial Institute (RPMI) 1,640 medium (Gibco, Grand Island, NY) supplemented with 10% fetal bovine serum (FBS) (Cytiva, Marlborough, MA), 1,000 U/mL penicillin, 1,000 μg/mL streptomycin (Gibco, Grand Island, NY), 1% minimum essential medium (MEM) vitamin solution, and 1% MEM non-essential amino acid solution (Gibco, Grand Island, NY). Cultures were distributed in quadruplicate into 48-well polystyrene plates (Corning Costar, Glendale, AZ). Each well was inoculated with 0.25 mL of either media (control) or CuSO₄ solutions at final concentrations of 1, 10, and 100 μM for both experiments, and an additional concentration of 625 μM for Experiment 2. The plates were incubated at 36°C in a 5% CO₂ atmosphere.

Sampling for ATP quantification was conducted at designated time points: 30 min, 2 h, and 5 h for Experiment 1; and 1, 3, 5, and 7 days for Experiment 2. For this, 50 μL samples were extracted from each well and transferred to a Nunclon Delta Surface white 96-well plate (Nunc A/S, Roskilde, Denmark). ATP levels were quantified using an ATP-lite luciferin-luciferase assay (Perkin Elmer, Waltham, MA), with luminescence readings obtained via a Synergy HTX spectrophotometer (BioTek, Winooski, Vermont). Background luminescence was measured in control wells with blank culture medium. To confirm sample purity, microscopic examinations for microbial contamination were performed and the cultures were microscopically monitored over the sampling periods to detect any bacterial or fungal outgrowth.

Cell pellets were lysed on ice using a lysis buffer composed of 0.1% SDS, 1% digitonin, 50 mM Tris–HCl, and 50 mM NaCl (pH 7.4) for 5 min, followed by 30 s of vortexing at maximum speed. The resulting cell lysates were then filtered through a 0.45 μm filter. Protein concentrations were determined using the Bradford assay, and the lysis buffer in each sample was adjusted accordingly to standardize protein levels across all samples. These lysates underwent SEC-ICP-MS analysis to detect copper across different molecular size fractions. The separation utilized a TSKgel G3000SWxl column (7.8 × 300 mm, 10 μm particle size, Tosoh Corporation, Tokyo, Japan) connected to an Agilent 1200 HPLC system. The mobile phase consisted of 50 mM ammonium acetate and 0.5% methanol (pH 7.4), with a flow rate of 0.6 mL/min. Copper quantification in each fraction employed three 50 μL injections, calibrated against a Cu standard curve (5 to 100 ng/mL) derived from human recombinant ceruloplasmin. Total metal content was assessed in triplicate aliquots via microwave digestion, comparing SEC-derived copper concentrations to total copper levels to ascertain a 94 ± 2% recovery efficiency.

Post-copper profiling, remaining samples (approximately 20 mL, excluding 100 μL for whole-lysate proteomics and 500 μL for non-fractionated IMAC enrichment) were fractionated via multiple 100 μL HPLC injections, with column performance verified every 20 injections using standards (Bio-Rad gel filtration standard mix and ceruloplasmin for Cu signal). Offline fraction collection from SEC was timed at intervals for high (10–12 min), medium (12.5–17.5 and 17.5–20 min), and low (20–25 min) molecular weight fractions. Collected fractions were lyophilized prior to IMAC processing.

To prepare for Cu-based IMAC protein enrichment, native copper was removed from proteins using Chelex 100 resin (Bio-Rad, CA) to expose Cu-binding sites. Proteins, either from whole lysate or dried fractions, were resuspended in 2 mL of 50 mM Na₂HPO4 and 300 mM NaCl (pH 7.4), treated with 0.5 g of Chelex resin, and agitated at 4°C for 1 h, centrifuged at 5,000 g for 2 min. This process was repeated two more times to maximize copper removal. The clarified supernatant was then subjected to IMAC enrichment using the Cu-loaded IMAC resin.

An IMAC copper resin (Geno Technology, Inc., MO) was used to enrich the SEC-fractionated samples for Cu-binding proteins. The manufacturer instructions were followed, including the rinsing of the resin with milli-Q water, the conditioning of the resin with the loading buffer that contains 10 mM imidazole, loading of the sample followed by 20 min of agitation as incubation, the removal of the supernatant, washing of the resin with binding buffer, and the final step consisted on the elution of the Cu-binding proteins with 3 consecutive volumes of 50 mM Na₂HPO4, 0.3 M NaCl, 0.25 M imidazole, pH 8.0. The amount of resin was adjusted according to the Cu concentration in each fraction, with the nominal Cu concentration of 30 μmol per ml of resin as reported by the vendor. As a positive control, Cu-depleted ceruloplasmin was introduced to the sample blank, while Zn-depleted carbonic anhydrase served as the negative control in a separate sample blank. The IMAC procedure was applied to the controls in a manner identical to that of the fractionated sample. The enriched sample fractions obtained were lyophilized to dryness and re-suspended in 50 mM ammonium carbonate for future tryptic digestion. Protein identification was conducted at the University of Cincinnati proteomics core facility utilizing liquid chromatography–tandem mass spectrometry (LC–MS/MS), following the previously published protocol (Green et al., 2013). Samples submitted for proteomics analysis included the high, medium, and low molecular weight fractions derived from the uninfected host lungs, the host lungs infected with P. murina, and the purified pathogen.

P. murina organisms were resuspended in 500 μL of mirVana miRNA lysis buffer (Invitrogen, Waltham, MA) and promptly snap-frozen at −80°C for subsequent storage and transfer to a sequencing core. Directional polyA RNA sequencing (RNA-seq) was conducted at the Genomics, Epigenomics, and Sequencing Core at the University of Cincinnati, employing established protocols detailed in previous publications (Walsh et al., 2019; Rapp et al., 2020) with updates. Briefly, total RNA quality was assessed using the Bioanalyzer (Agilent, Santa Clara, CA). The NEBNext Poly(A) mRNA Magnetic Isolation Module (New England BioLabs, Ipswich, MA) was employed for the isolation of polyA RNA during library preparation, utilizing 400 ng of high-quality total RNA. The enrichment of polyA RNA was performed using the SMARTer Apollo automated NGS library prep system (Takara Bio United States, Mountain View, CA). Subsequently, the NEBNext Ultra II Directional RNA Library Prep kit (New England BioLabs) was utilized for library preparation, employing a Polymerase Chain Reaction (PCR) cycle number of 10. Following library QC and quantification via Qubit quantification (ThermoFisher, Waltham, MA), individually indexed libraries were proportionally pooled and subjected to sequencing on the NextSeq 550 Sequencer (Illumina, San Diego, CA) with a sequencing setting of SR 1 × 85 bp, generating approximately 25 million reads. Upon completion of the sequencing run, FASTQ files for subsequent data analysis were automatically generated through Illumina BaseSpace Sequence Hub.

Raw sequencing files were first preprocessed by FASTP (Chen et al., 2018) to remove or prune low quality reads, and trim adapter sequences. Then, transcript abundance was quantified by SALMON (Patro et al., 2017) using the Pneumocystis murina B123 strain’s genome as the reference (NCBI RefSeq Assembly ID: GCF_000349005). Differential gene expression analysis was conducted through DESeq2 (Love et al., 2014). Differentially expressed genes were identified based on a fold change (FC) of ≥2 and an adjusted p-value (q-value) of <0.05.

Genes of interest were selected for validation by reverse transcription-quantitative polymerase chain reaction (RT-qPCR). The oligonucleotide primers to these genes were designed based on sequences from NCBI/GenBank. Primers were designed to span an intron junction to detect amplification of genomic DNA. cDNA was synthesized from the extracted RNA using SuperScript IV VILO master mix with ezDNase enzyme (Invitrogen, Grand Island, NY) per the manufacturer’s protocol. The remaining RNA and the freshly synthesized complementary DNA (cDNA) were stored at −80°C.

Real-time PCR was performed in an Applied Biosystems 7,500 fast real-time PCR system (Life Technologies, Grand Island, NY). The reactions were performed in triplicate in a final volume of 20 μL containing 1 μL PowerUp SYBR green master mix (Life Technologies, Grand Island, NY) and 500 nM each primer pair. The reaction mixtures were initially incubated at 50°C for 2 min and 95°C for 2 min, followed by 40 cycles of 15 s at 95°C, 59°C for 15 s, and 72°C for 1 min. Fluorescence data were captured during the 72°C step. Disassociation curves were also calculated for the reactions to determine the melting temperature (Tm) of the products. The reference gene used was thymidylate synthase (TS). The threshold cycle (ΔCT) value between the validation gene and the TS gene was calculated by subtracting the average CT value of TS from the average CT value of the validation gene. ΔΔCT values were calculated by subtracting the ΔCT value of the TS gene from the ΔCT value of the validation sample.

Copper associated protein annotations were retrieved from the FungiDB database (Amos et al., 2022) for sequenced Pneumocystis species and other clinically significant and model fungi (as of August 2023). Since FungiDB transfers annotations inclusively, even from non-top homologs, an additional manual curation of Pneumocystis proteins was conducted using a homology search by BLAST against the NCBI nr database (top 250 hits with E-value<0.001, restricted to the fungi taxon) in conjunction with the UniProt and Pfam annotations. Table 1 provides a summary of the number of proteins identified using the keyword “copper.” Notably, Pneumocystis species exhibit approximately three to five times fewer copper-associated proteins compared to other fungi. Given that this study employed P. murina as the experimental model organism, Table 2 presents a list of respective proteins, derived from FungiDB annotations, either through direct gene description (second column) or annotations of homologous genes (third column), followed by the manual curation.

P. murina cell lysates and lung homogenates from both infected and uninfected mice underwent SEC-ICP-MS analysis to detect Cu. Figure 1 presents the chromatograms for Cu profiling, where the area under the curve correlates directly with the Cu concentration in each fraction. Since the total protein concentration of each cell lysate was normalized, the chromatograms allow for a direct comparison of Cu content and distribution. Notably, the Cu distribution in host cell lysates remained consistent regardless of P. murina infection. However, the Cu distribution in the isolated pathogen markedly differed, underscoring the effectiveness of the pathogen isolation procedure.

A summary of the subsequent analysis of the fractions and the identification of the IMAC enriched Cu-binding proteins via LC–MS/MS is presented in Table 3. Of note, since the pathogen cannot be cultivated, any samples with extracted and purified organisms are still contaminated with the host cells. Table 4 provides the list of P. murina proteins identified in all molecular weight factions. The complete list of the identified proteins associated with copper, both for the pathogen and the host, can be found in Supplementary File S1.

The proteomic identification through the IMAC enrichment method likely includes a range of metalloproteins typically linked with zinc, manganese, or other divalent metal ions, alongside the intended copper-binding proteins. This diversity arises from the demetallation process using Chelex 100 resin, which also extracts these metals while primarily removing Cu to generate apo-forms of the existing Cu-binding proteins. The predictability of differing affinities for Cu across various apo-forms of metalloproteins is challenging due to the IMAC resin’s nature and the attempt to limit available Cu to minimize nonspecific binding. Consequently, achieving complete selectivity for Cu-binding proteins is difficult. To enhance method specificity, optimization of incubation times, rinse volumes, and elution buffer volumes was undertaken to minimize the retention of carbonic anhydrase, a common off-target zinc protein, employing ceruloplasmin as a positive control due to its multifaceted copper-binding sites. Further refinement was evidenced by re-analyzing demetallated SEC fractions with ICP-MS prior to IMAC enrichment, revealing an 89 ± 6% reduction in the original Cu content (data not shown). This adjustment significantly improved the selectivity for ceruloplasmin over carbonic anhydrase, potentially decreasing false negatives in Cu-proteome analysis.

To gain insights into the copper sensing mechanisms in P. murina, we exposed freshly extracted organisms to copper(II) sulfate for 0.5, 2, and 5 h, followed by RNA-seq analysis. In the initial batch, extracted from 6 mice, we applied concentrations of 1 ng/mL and 100 ng/mL of CuSO₄ for comparison with untreated organisms. Notably, no statistically significant differential gene expression was observed (Supplementary Figure S1A).

In a subsequent batch, we exposed the organisms to elevated CuSO₄ concentrations: 1, 10, and 100 μM (0.1596, 1.596, and 15.96 μg/mL, respectively), at the same time points as the first batch. In this case, organisms from three out of five animals exhibited differential gene expression patterns (Supplementary Figure S1B), leading to their selection for further RNA-seq analysis.

Figure 2 depicts the temporal dynamics of gene perturbations, with notable changes occurring at the 5-h time point. The most significant perturbation was evident during treatment with 100 μM CuSO₄. Table 5 provides a summary of the genes that were statistically significantly perturbed. Comprehensive gene lists can be accessed in Supplementary File S2.

To validate some genes as potential candidates responding to copper excess in P. murina, we selected eight genes with descriptions or domain annotations related to copper or other metal ions (Table 6). Figure 3 illustrates the outcomes of RT-qPCR analysis for these genes using samples treated with 100 μM CuSO4, in comparison to control samples. Notably, Figure 3 reveals that four out of the eight selected genes exhibit statistically significant upregulation. Among these four genes, PNEG_02609 is indicative of a metal ion transporter, while PNEG_01236, PNEG_01675, and PNEG_01730 are likely transcription factors.

To assess the tolerance of P. murina to free labile copper, viability assays were conducted. Cryopreserved samples of the organisms were exposed to varying concentrations of Cu(II) sulfate for seven days, including both the concentrations employed in differential gene expression studies above and the concentration (625 μM) reported as beneficial in axenic culture media (Riebold et al., 2023). The results, depicted in Figure 4, reveal that P. murina exhibits comparable viability to untreated controls at lower CuSO₄ concentrations (1 and 10 μM). However, at higher concentrations, Cu(II) sulfate demonstrates a toxic effect, indicated by a rapid decline in the number of metabolically active organisms.

The two-way ANOVA test reveals that both individual variables (days post-treatment and copper sulfate concentration) as well as their interaction exhibit statistically significant differences in results. Further analysis through Tukey’s post-hoc test demonstrates that treatments utilizing 100 μM and 625 μM CuSO₄ significantly differ from the controls and treatments with lower concentrations. It should be noted that there is no continuous culture/growth method for any Pneumocystis species outside the host lungs. The decline in ATP in the short-term culture we used is typical for such in vitro systems.