Deletion of ppr3, ppr4, ppr6, or ppr10 results in a dramatic loss of viability

Because mitochondrial ETC dysfunction might lead to a decrease in lifespan [8], we first constructed all ppr deletion mutants except Δppr9 (ppr9 is essential for cell viability and cannot be deleted) and evaluated the viabilities of these mutants using spotting assays. Deletion of ppr1, ppr3, ppr4, ppr6, or ppr10 dramatically decreased cell viability at the stationary phase, whereas deletion of other ppr genes only modestly affected cell viability (Fig. 1A). The viabilities of Δppr2, Δppr5, Δppr7, and Δppr8 mutants were comparable to that of the wild-type strain even after prolonged growth (106 h). Δppr3, Δppr4, Δppr6, and Δppr10 mutants also exhibited severe flocculation (Fig. 1B). We also confirmed that none of these mutants could grow on media containing a nonfermentable carbon source such as galactose or glycerol (Fig. 1C), which requires respiration and therefore mitochondrial gene expression. Below, we wanted to focus on the loss of viability phenotype exhibited by Δppr3, Δppr4, Δppr6, and Δppr10 mutants. Efforts to understand the flocculation phenotype will be described elsewhere (Y. Su, J. Zhang & Y. Huang, unpublished results).

Deletion of ppr3, ppr4, ppr6, or ppr10 induces apoptosis in S. pombe

To determine whether apoptosis accounted for the loss of viability in cells deleted of ppr3, ppr4, ppr6, or ppr10, we examined nuclear morphology of the wild-type and the mutant cells using DAPI staining [20-23]. After grown for 24 h, cells were stained with DAPI and nuclei were visualized by fluorescence microscopy. As shown in Fig. 2A, most nuclei in Δppr3, Δppr4, Δppr6, and Δppr10 cells examined were fragmented, a phenotype that resembles apoptosis in multicellular organisms, whereas the nuclei of wild-type cells remained intact for the same time point. Similar results were obtained at 36 and 48 h time points (data not shown).

In mammals, exposure of PS, a phospholipid component of cell membranes, on the cell surface has been considered a characteristic feature of apoptosis and serves as a molecular cue for engulfment of dying cells by phagocytes. This apoptotic PS externalization has also been observed in S. cerevisiae and S. pombe [20, 24]. To confirm that the death of the ppr deletion mutants was occurring via apoptosis, we examined PS externalization in the mutant cells by Annexin V-FITC staining. Spheroplasts of wild-type, Δppr3, Δppr4, Δppr6, and Δppr10 cells grown to stationary phase were prepared and incubated with FITC-labeled Annexin V. Strong FITC fluorescence was observed at the periphery of all Δppr3, Δppr4, Δppr6, and Δppr10 cells examined, while no FITC fluorescence could be observed for wild-type cells, suggesting that PS is indeed exposed to the outer membrane of the mutant cells (Fig. 2B). We also examined the membrane integrity of the mutants by propidium iodine (PI) staining. In Δppr3, Δppr6, and Δppr10 cells, PI was excluded from the Annexin V-positive cells, suggesting that the membranes of these mutant cells were intact. At the same time, 54.5 ± 3.3% (mean ± SD) of Δppr4 cells could be stained by PI, suggesting increased membrane permeability, which is characteristic of cells in the later stages of apoptosis (or in necrosis). These results suggest that PS exposure as detected by Annexin V-FITC binding in Δppr3, Δppr6, and Δppr10 cells appears to be early apoptosis, whereas PS exposure in Δppr4 cells were advanced apoptosis or necrosis. Collectively, these observations suggest that the observed losses of cell viability upon deletion of ppr3, ppr4, ppr6, or ppr10 result from the induction of apoptosis.

Mitochondrial fragmentation correlates with apoptosis in mammals [25-27]. We went on to examine the mitochondrial morphology in the wild-type cells and cells depleted of ppr3, ppr4, ppr6, or ppr10 by fluorescence microscopy after MitoTracker staining. As shown in Fig. 3A, all examined cells depleted of ppr3, ppr4, ppr6, or ppr10 had fragmented mitochondria that appeared as small individual dots more or less clustered, whereas wild-type cells exhibited a branching tubular mitochondrial network for the same time point. These results suggest that deletion of ppr3, ppr4, ppr6, or ppr10 causes mitochondrial fragmentation.

Disruption of ppr3, ppr4, ppr6, or ppr10 results in ROS formation

Mitochondrial ETC dysfunction has been shown to increase the production of ROS and accumulation of ROS is a key event in triggering apoptosis [28, 29]. To determine whether deletion of ppr3, ppr4, ppr6, or ppr10 caused accumulation of ROS, cells grown for different time points were incubated with dihydrorhodamine 123 (DHR123), which can be oxidized to the fluorochrome rhodamine 123 in the presence of ROS. As shown in Fig. 3B, wild-type cells showed only a background level of fluorescence, whereas Δppr3, Δppr4, Δppr6, and Δppr10 cells showed an intense intracellular staining by DHR123.

Next, we monitored the ROS levels in wild-type cells and cells depleted of ppr3, ppr4, ppr6, or ppr10 at the different time points by flow cytometric analysis using DHR123 as probe, which undergoes oxidization by ROS, and then, fluoresces. ROS levels were significantly higher at all time points in cells depleted of ppr3, ppr4, ppr6, or ppr10 compared with wild-type cells (Fig. 3C). Among these mutants, the highest ROS levels were observed in Δppr3 cells, while the lowest ROS levels were found in Δppr4 cells. These results suggest that mitochondrial ETC dysfunction caused by deletion of ppr3, ppr4, ppr6, or ppr10 leads to mitochondrial fragmentation and ROS formation.

Transcriptomic analysis of the ppr10 deletion mutant

To determine the effects of deletion of ppr3, ppr4, ppr6, or ppr10 on global gene expression, we used high-throughput RNA sequencing (RNA-seq) to analyze the transcriptome of the Δppr10 mutant. The Δppr10 mutant was chosen as the representative because of our interest in this gene specifically. Total RNA was isolated, and the total number of reads obtained and mapped for each sample is detailed in Table 1. The expression profile of the Δppr10 mutant was compared with that of the WT strain, and only genes differently expressed more than twofold and false discovery rate (FDR) < 0.05 were subsequently analyzed. We identified a total of 146 genes whose expression was changed by twofold or greater: 100 genes were up-regulated and 46 genes were down-regulated (Table S1). We performed gene ontology (GO) biological process enrichment analysis to identify biological processes most affected by deletion of ppr10. This analysis revealed enrichment of five GO categories with P < 0.05 (Table 2): iron ion transport (GO:0006826), cellular iron ion homeostasis (GO:0006879), cell adhesion (GO:0007155), mitochondrial ATP synthesis-coupled electron transport (GO:0042775), and disaccharide metabolic process (GO:0005984).

We further examined the genes associated with enriched GO categories. Nine genes (frp1, frp2, fip1, fio1, str1, str3, sib2, mmt1, and pcl1) are associated with the enriched GO categories ‘iron ion transport’ and/or ‘iron ion homeostasis’. All of these genes, except for pcl1, were up-regulated > 2 and FDR < 0.05 in Δppr10 cells (Table 3). frp1, frp2, fip1, and fio1 mediate ferrous iron transport [30, 31]. Fe³⁺ is first reduced to Fe²⁺ by the cell surface reductase Frp1 and Frp2. Then, Fe²⁺ is transported into the cell by a two-component complex composed of the ferroxidase Fip1 and the iron permease Fio1. str1 and str3 encode membrane transporters that take up siderophore-bound iron [32]. sib2 encodes ornithine N⁵ monooxygenase that participates in ferrichrome biosynthesis [33]. mmt1 encodes a lone mitochondrial iron ion transmembrane transporter in S. pombe. pcl1, which encodes a vacuolar ferrous iron transporter and is involved in iron storage [34], is down-regulated in the mutant. In addition, we found that shu1 was up-regulated by 29-fold, with FDR < 0.05 in the mutant. Shu1 mediates iron ion uptake through Shu1-dependent heme assimilation [35] but is annotated to another GO category ‘transmembrane transport’ (GO:0055085). These results suggest a connection between mitochondrial ETC dysfunction to perturbation of iron ion transport and iron ion homeostasis.

To confirm these results, we performed qRT-PCR to analyze the expression levels of frp1, frp2, fip1, fio1, str1, str3, and shu1. Consistent with predictions from RNA-seq data, the mRNA levels of these genes were up-regulated in the Δppr10 mutant (Fig. 4). Thus, we concluded that deletion of ppr10 led to increased expression of iron uptake and/or iron homeostasis genes.

Among six genes (pfl2, pfl6, pfl7, pfl8, pfl9, and cbf12) associated with the enriched GO category ‘cell adhesion’, pfl6 pfl8, pfl9, and cbf12 were up-regulated > 2 and FDR < 0.05 in Δppr10 cells (Table 3), which is consistent with the observation that ppr10 deletion causes flocculation (Fig. 1B). pfl2, pfl6, pfl7, pfl8, and pfl9 encode flocculin, while Cbf12, which is a member of the CSL transcription factors, regulates expression of flocculin genes [36]. Besides these genes, we found that pfl1/gsf2, which encodes a flocculin but annotated to another GO category ‘flocculation’ (GO:0000128), is also up-regulated ~13-fold, with FDR < 0.05 in the mutant (Table 3). It has been shown that individual overexpression of each of flocculin genes in S. pombe is sufficient to trigger flocculation. This effect is strongest for pfl1/gsf2 [36]. Overexpression of cbf12 can also trigger flocculation through activation of pfl1/gsf2 [36].

Nine genes (qcr7, cyt1, cox6, qcr6, qcr8, qcr10, cox5, rip1, and nde2) were found in the enriched GO category ‘mitochondrial ATP synthesis-coupled electron transport’. Qcr7, Cyt1, Qcr6, Qcr8, Qcr10, and Rip1 are nuclear-encoded mitochondrial ETC complex III subunits, while Cox5 and Cox6 are nuclear-encoded mitochondrial complex IV subunits. Nde2 is a putative mitochondrial NADH dehydrogenase. All these genes were down-regulated in Δppr10 cells. In addition, we found that cox13-encoding nuclear-encoded mitochondrial complex IV subunit VIa was also down-regulated > 2-fold, with FDR < 0.05 in the mutant. However, this gene is not annotated in the GO category ‘mitochondrial ATP synthesis-coupled electron transport’.

Our analysis showed that five genes (mel1, ntp1, tps1, tps3, and agl1) associated with the enriched GO category ‘the disaccharide metabolic process’ were up-regulated > 2-fold, with FDR < 0.05 in the ppr10 mutant. Among these genes, mel1, which encodes α-galactosidase (catalyzing the hydrolysis of melibiose into glucose and galactose), was the most highly up-regulated gene (~19-fold). ntp1, tps1, and tps3 are involved in trehalose biosynthesis, which is linked to the stress response [37-39]. agl1 encodes an extracellular α-glucosidase involved in maltose utilization [40]. It is likely that increased expression of these genes may contribute to protection against oxidative stress caused by mitochondrial dysfunction.

Deletion of ppr3, ppr4, ppr6, or ppr10 causes increased sensitivity to iron toxicity

Because RNA-seq and qRT-PCR analyses revealed that expression of iron uptake and/or iron ion homeostasis genes were elevated in the Δppr10 mutant, we tested the ability of Δppr3, Δppr4, Δppr6, and Δppr10 mutants to grow on solid growth medium containing a high concentration of FeCl₃ or FeSO₄. We found that incubation of Δppr3, Δppr4, Δppr6, and Δppr10 mutants on high-iron media resulted in decreased viability (Fig. 5). These results suggest that deletion of ppr3, ppr4, ppr6, or ppr10 may perturb iron homeostasis, which may result in increased iron toxicity.

Addition of bathophenanthroline disulfonic acid (BPS) nearly restores viability to Δppr3, Δppr4, Δppr6, and Δppr10 mutants and reduces ROS levels in the mutants

RNA-seq analysis revealed that the expression levels of many genes involved in iron uptake and/or iron homeostasis have been elevated in Δppr10 cells, suggesting increased assimilation of iron in these mutants. In the presence of oxygen, excess amounts of intracellular free iron ions may catalyze the generation of more active ROS such as hydroxyl radical, which can damage cellular components and cause apoptotic cell death. Thus, we first tested whether addition of the selective iron chelator, bathophenanthroline disulfonic acid (BPS), could restore the viabilities of Δppr3, Δppr4, Δppr6, and Δppr10 mutants. We found that addition of BPS resulted in restoration of viabilities of these mutants to near wild-type (Fig. 6A). However, BPS could not restore viabilities of these mutants in the presence of FeCl₃ (Fig. 6A), confirming the specificity of iron-chelating phenotype. Addition of the EGTA could not restore the viabilities of the ppr deletion mutants, indicating that the calcium-specific chelator EGTA did not have similar effect (Fig. 6B). Taken together, these results suggest that the excess iron may be a reason for the apoptotic cell death of the mutant cells.

We next examined apoptotic markers such as nuclear fragmentation and externalization of PS in Δppr3, Δppr4, Δppr6, and Δppr10 mutants grown in the presence of BPS. Consistent with results from spotting assays, we found that 83 ± 2.2% (mean ± SD) of Δppr3, 84 ± 2.3% (mean ± SD) of Δppr4, 74 ± 3.4% (mean ± SD) of Δppr6, and 85 ± 3.9% (mean ± SD) of Δppr10 cells exhibited normal nuclear morphology upon BPS treatment (Fig. 6C). In addition, the BPS treatment led to no FITC fluorescence in many Δppr3, Δppr4, Δppr6, and Δppr10 cells (Fig. 6D). These results suggest that BPS can significantly suppress apoptotic cell death of Δppr3, Δppr4, Δppr6, and Δppr10 mutants. To determine whether increased survival of Δppr3, Δppr4, Δppr6, and Δppr10 mutants was associated with reduced levels of ROS in these mutants, we examined the intracellular levels of ROS in Δppr3, Δppr4, Δppr6, and Δppr10 cells grown in the presence of BPS. We found that BPS significantly suppressed ROS generation in these mutants (Fig. 6E,F).

Strains and media

Schizosaccharomyces pombe strains used in this study are listed in Table 4. Deletion mutants were constructed by the one-step gene replacement method. The deletion cassettes to generate the ppr1-ppr8, ppr10 null alleles were generated by cloning the 5′ and 3′ flanking sequences of ppr1-ppr8, ppr10 into pFA6a-kanMX6 [48]. These deletion cassettes were transformed into wild-type strain yHL6381 by the lithium acetate method [49]. Primers used for deletion cassette construction and other plasmid constructions are available upon request. All deletions were verified by PCR.

Schizosaccharomyces pombe strains were routinely grown in liquid-rich YEPD medium (0.5% yeast extract, 0.5% tryptone, 1% glucose, and appropriate supplements). The survival difference between the wild-type and mutants is more noticeable when cells were grown in YEPD compared to those grown in YES (0.5% yeast extract, 3% glucose, and appropriate supplements).

Cell growth assay

Cells were grown at 30 °C in liquid YEPD medium until they reached midlog phase. Cultures were then normalized to an OD₆₀₀ of 0.2 and continued to grow in the absence or presence of 200 μm BPS, 200 μm BPS, and 50 mm FeCl₃ or 5 mm EGTA. Samples were collected at the indicated time points (in hours), and normalized to an OD₆₀₀ of 0.2, serially fivefold diluted, and spotted (3 μL of dilutions) on solid-rich medium. Plates were photographed after 2 or 4 days of incubation at 30 °C.

Sensitivity to iron was also evaluated by spotting assays. Fresh colonies were grown in liquid-rich media overnight, diluted to an OD₆₀₀ of 0.2 in fresh medium and continued to grow at 30 °C for 24 h. An aliquot of each culture was taken and normalized to an OD₆₀₀ of 0.2, serially fivefold diluted, and spotted on solid-rich medium without or with 5 mm FeCl₃ or 5 mm FeSO₄. These concentrations were chosen based on previous studies [50, 51]. The plates were photographed after 2.5–5 days of incubation at 30 °C.

DAPI staining

Cells were grown at 30 °C in liquid medium overnight, diluted to an OD₆₀₀ of 0.2, and continued to grow for 24 h. Cells were washed twice with PBS (2.5 mm NaH₂PO₄, 7.5 mm Na₂HPO₄, 145 mm NaCl, pH 7.4), incubated in PBS containing 0.25 mg·mL⁻¹ 4′,6′-diamidino-2-phenylindole (DAPI; Solarbio, Beijing, China) for 5–10 min. After washing with PBS, cells were imaged using a Zeiss Axio Imager A1 microscope (Carl Zeiss, Thornwood, NY, USA) equipped with a PCO Sensicam CCD (charge-coupled-device) camera (Sensicam, PCO, Germany). Data were analyzed using MetaMorph image processing software (Universal Imaging, Downington, PA, USA).

Annexin V staining

Annexin V staining was performed with the Annexin V-FITC Apoptosis Detection Kit (Beyotime Biotechnology, Shanghai, China). After growth in liquid medium for 24 h, cells were centrifuged, then washed twice with PBS, and resuspended in 10 mm sodium phosphate buffer, pH 5.8, 1.2 m MgSO₄. Cells were incubated for 1–3 h at room temperature with 10 mg·ml⁻¹ lysing enzymes from Trichoderma harzianum (L1412; Sigma-Aldrich, St. Louis, MO, USA) and 10 mg·mL⁻¹ Yatalase (TaKaRa Shuzo, Otsu, Japan) until spheroplasts were formed. Once spheroplasts were formed, an equal amount of STC buffer (1.2 m sorbitol, 10 mm CaC1₂, 10 mm Tris/HCI, pH 7.5) was added to the cell suspension. After centrifugation at 1000 g for 10 min at 4 °C, pellets were washed with 200 μL of Annexin V-FITC-binding buffer, resuspended in 210 μL Annexin V-FITC-binding buffer containing 5 μL of Annexin V-FITC and 10 μL of PI, and incubated at room temperature in the dark for 10–20 min. After incubation, the cells were washed once in 200 μL of Annexin V-FITC-binding buffer and resuspended in 200 μL of Annexin V-FITC-binding buffer. Cells were visualized by fluorescence microscopy.

MitoTracker staining

Cells were taken after growth in liquid culture for 12 h. After washed twice with PBS, samples were incubated 5 min at 30 °C with 10 μm of MitoTracker (Life Technologies, Foster City, CA, USA) in PBS, and analyzed immediately after washing with PBS by a Zeiss Axio Imager A1 microscope equipped with a PCO Sensicam CCD (charge-coupled-device) camera.

Detection of ROS

To visualize ROS, cells were taken at the indicated time points. After washed twice with PBS, cells were incubated 30 min at 30 °C with 30 μm of DHR123 (Sigma-Aldrich) in PBS, washed with PBS, and resuspended in PBS. The resulting fluorescence was imaged using a Zeiss Axio Imager A1 microscope equipped with a PCO Sensicam CCD (charge-coupled-device) camera. To quantitate ROS levels, the fluorescent intensity in cell suspension was measured by flow cytometer (BD Accuri™ C6; BD Biosciences, San Jose, CA, USA) with 488 nm excitation and 533/30 nm emission filters. Experiments were run in quadruplicate, and at least 10 000 cells per sample were analyzed. Mean fluorescence was calculated with cflow software (BD Biosciences).

RNA isolation and RNA sequencing (RNA-seq)

Total RNA was prepared from cultures of wild-type and Δppr10 cells grown to an OD₆₀₀ of 1.0 in YES medium containing 2% galactose (which was used to deflocculate cells) using an E.Z.N.A.^® Yeast RNA Kit (OMEGA BIO-TEK, Norcross, GA, USA) according to the manufacturer's instructions. Contaminating genomic DNA in RNA was removed by treatment with RNase-free DNase (Fermentas, Hanover, MD, USA) to remove potential genomic DNA contamination. mRNAs were purified from total RNAs with oligo-dT magnetic beads. The mRNA was broken into short fragments of 200 bp, and the resulting RNA was reverse-transcribed into cDNA using random hexamers. The 3′-ends of cDNA were adenylated and 5′-ends were repaired, and adapters were ligated. The ligated fragments were amplified by PCR and sequenced on the Illumina HiSeqTM 2500 (Illumina Inc., San Diego, CA, USA) according to standard protocols. Quality-controlled reads were mapped to the S. pombe genome from NCBI using SOAPaligner/SOAP2. The reads per kilobase per million for each gene was calculated based on the length of the gene, and the read counts mapped to it. GO enrichment analysis was performed using the GO database (http://geneontology.org/) and annotations to biological process terms [52].

Quantitative real-time RT-PCR

RNA was isolated from wild-type S. pombe (yHL6381) and Δppr10 cells using the E.Z.N.A.® Yeast RNA Kit (OMEGA BIO-TEK). Contaminating genomic DNA in RNA samples was removed by treatment with RNase-free DNase (Fermentas). RNA samples were reversed transcribed with the oligo-dT₁₆ primer using RevertAid™ First Strand cDNA Sythesis Kit (Fermentas). Comparative qPCR analysis was performed using the StepOne™ Real-Time PCR system (Life Technologies) with each primer sets (Table 5). All reactions were performed in triplicate. Data analysis was performed by StepOne™ software (Life Technologies). The fold change in gene expression was calculated by using the 2^−ΔΔCT method. The C_T values were normalized against that of actin (act1) mRNA from the same preparations to give the ΔC_T values. The ΔΔC_T values were calculated by subtracting the ΔC_T vale of the wild-type sample from those of different ppr deletion mutants.