CGA-N9, a peptide derived from human chromogranin A (CGA), was found to have antimicrobial activity in our previous investigation, but its mechanism of action remains unclear. Herein, the mechanism of action of CGA-N9 was investigated. We found that CGA-N9 induced the depolarization of the cell membrane and uptake of calcium ions into the cytosol and mitochondria. With the disruption of the mitochondrial membrane potential, the generation of intracellular reactive oxygen species (ROS) increased. Accordingly, we assessed apoptotic processes in Candida tropicalis cells post-treatment with CGA-N9 and found cytochrome c leakage, chromatin condensation and DNA degradation. The interaction of CGA-N9 with DNA in vitro showed that CGA-N9 did not degrade DNA but bound to DNA via an electrostatic interaction. In conclusion, CGA-N9 exhibits antifungal activity by inducing apoptosis in C. tropicalis.
Invasive fungal infections continue to appear as the size of the immunocompromised population increases [1,2]. Candida tropicalis is one of only five Candida species accounting for 92% of candidemia cases [3–5]. The number of infections caused by Candida parapsilosis and C. tropicalis in this population is increasing . Azole drugs are commonly used to treat Candida infections . However, the overuse of azole compounds in past decades has promoted the emergence of drug-resistant Candida species . Thus, novel anticandidal agents are urgently needed [8,9].
Antimicrobial peptides (AMPs), which are produced by virtually all life forms and are multidimensional defence molecules that are not easily overcome by microorganisms using single-approach resistance strategies, represent almost an inexhaustible source of potential therapeutic agents [10,11]. Chromogranin A (CGA) is a ubiquitous soluble protein in most endocrine cells and neuronal cells. The N-terminus of CGA is reported to have antimicrobial activity . Our previous research demonstrated that CGA-N9, corresponding to the N-terminal amino acid sequence from amino acids 47 to 55 of human CGA, has antimicrobial activity against Candida species, Cryptococcus neoformans and some gram-positive bacteria .
In our previous research, we found that most CGA-N9 passes through the C. tropicalis cell membrane via direct cell penetration, whereas the remainder enters through macropinocytosis and sulfate proteoglycan-mediated endocytosis, with a slight contribution from clathrin-mediated endocytosis . The intracellular activity of CGA-N9 after its internalization is still unclear. In the present study, the mechanism of CGA-N9 related to apoptosis in Candida was explored.
Materials and methods
Microorganism and materials
C. tropicalis CGMCC2.3739 (Beijing, China) was cultured in Sabouraud dextrose (SD) medium at 28°C for 24 h. CGA-N9 with N-terminal and C-terminal deprotection was synthesized by the solid-phase peptide synthesis method. Peptide purification was performed by high-performance liquid chromatography. Other chemicals used in the present study were of analytical grade and from commercial suppliers.
Depolarization of the cell membrane
The C. tropicalis cell membrane depolarization post-CGA-N9 treatment was detected by using DiSC3(5) [14,15]. Briefly, C. tropicalis cells in the mid-log phase were suspended in 5 mM HEPES to an OD600 of 0.05. A final concentration of 4 μM DiSC3(5) in DMSO was added when the absorbance was stable. The fluorescence intensity of DiSC3(5) was detected at λex/λem = 600 nm/675 nm at 30 s intervals. CGA-N9 was added to a final concentration of 3.9 μg/ml at the stable and maximal fluorescence intensity. Membrane potential disruption was determined by the change in fluorescence intensity. NaN3 was used as a positive control, and PBS (20 mM, pH 7.2) was used as a negative control. The measurements were repeated three times.
Detection of calcium ions in the cytosol and mitochondria
Changes in Ca2+ levels in the cytosol and mitochondria were detected by the membrane-penetrating Ca2+ probes Fluo-4AM and Rhod-2AM (Sigma–Aldrich, Shanghai, China), respectively [16,17]. C. tropicalis cells (5 × 106 CFU/ml) in the mid-log phase were incubated with 3.9 μg/ml CGA-N9 for 4 h to 16 h at 28°C in 4 h intervals. H2O2 (10 mM) was used as a positive control, and PBS (20 mM, pH 7.2) was used as a negative control. The cells were washed with Krebs buffer (132 mM NaCl, 4 mM KCl, 1.4 mM MgCl2, 6 mM glucose, 10 mM HEPES, 10 mM NaHCO3, 1 mM CaCl2; pH 7.2) containing 0.01% Pluronic F-127 (Sigma–Aldrich, Shanghai, China) and 1% bovine serum albumin. The cells were suspended in Krebs buffer and incubated with 5 μM Fluo-4AM or 10 μM Rhod-2AM at 28°C for 30 min. After washing three times with calcium-free Krebs buffer, the fluorescence of Fluo-4AM (λex/λem = 494 nm/516 nm) and Rhod-2AM (λex/λem = 549 nm/578 nm) was detected using a fluorescence spectrophotometer (Cary Eclipse, Australia). PBS (20 mM, pH 7.2) was used as a negative control, and H2O2 was used as a positive control. The measurements were repeated three times.
Assessment of intracellular ROS accumulation
The oxidation-sensitive fluorescent dye dihydrorhodamine-123 (DHR123) was used to detect intracellular ROS accumulation . A total of 5 × 106 CFU/ml C. tropicalis cells in the mid-log phase were incubated with 3.9 μg/ml CGA-N9 for 8 h at 28°C. H2O2 (10 mM) was used as a positive control, and PBS (20 mM, pH 7.2) was used as a negative control. The cells were stained with 5 μg/ml DHR-123 (Sigma–Aldrich, Shanghai, China) for 1 h at 28°C, avoiding light after washing with 20 mM phosphate-buffered saline (PBS; pH 7.2). The samples were analyzed by a FACSCalibur flow cytometer (BD, U.S.A.) at λex/λem = 507 nm/529 nm.
Mitochondrial membrane potential assay
The Molecular Probes JC-1 dye (Beyotime, Shanghai, China) was used to examine the dissipation of mitochondrial membrane potential in C. tropicalis cells . The fluorescence of aggregates and monomers was detected by flow cytometry. Briefly, C. tropicalis cells (5 × 106 CFU/ml) in the mid-log phase were incubated with 3.9 μg/ml CGA-N9 at 28°C for 8 h. After washing with 20 mM PBS (pH 7.2), JC-1 at a final concentration of 1× was added to the cell suspension and incubated at 28°C for 20 min. C. tropicalis cells treated with 10 mM H2O2 were used as a positive control, and those treated with PBS (20 mM, pH 7.2) were used as a negative control. The fluorescence intensity at 525 nm (FL1) or 595 nm (FL2) was recorded with a FACSCalibur flow cytometer (BD, U.S.A.). The ratio of JC-1 aggregate to monomer intensity was calculated, and the data represent the mean ± standard deviation of three independent experiments. Statistical significance was determined by Student's t-test. P values <0.05 and <0.01 indicate statistical significance.
Detection of cytochrome c release
The mitochondrial and cytosolic Cyt c content was isolated via differential velocity centrifugation . Briefly, C. tropicalis cells (5 × 106 CFU/ml in the mid-log phase) were incubated with 3.9 μg/ml CGA-N9 for 0, 4, 8, 12 and 16 h at 28°C. The incubated cells were homogeneously dispersed in medium (50 mM Tris, 2 mM ethylenediaminetetraacetic acid, 1 mM phenylmethanesulfonylfluoride; pH 7.5) and then glucose was added to a final concentration of 2%. The suspension was centrifuged for 10 min at 20 000×g, and the supernatant and cell pellet were collected. The supernatant was centrifuged at 30 000×g for 45 min to collect the resulting supernatant from the superspeed centrifugation to quantify cytosolic Cyt c. The cell pellets obtained from the above step were homogenized in Tris-EDTA buffer (50 mM Tris, 2 mM EDTA; pH 5.0) and centrifuged at 7727×g for 30 s. The resulting pellet was suspended in 2 mg/ml Tris-EDTA buffer to quantify mitochondrial Cyt c. Ascorbic acid was added to a final concentration of 500 mg/ml for 5 min to reduce the Cyt c in the cytosolic and mitochondrial samples. The absorbance at 550 nm was measured with a spectrophotometer (UV1800; AOXI, Shanghai, China) to determine the relative quantities of reduced cytoplasmic and mitochondrial Cyt c in the samples. The Cyt c content changes in the mitochondria and cytosol mediated by CGA-N9 were analyzed to estimate Cyt c release from mitochondria to the cytosol.
Detection of nuclear condensation
Nuclear condensation was assessed by using 4,6-diamidino-2-phenylindole (DAPI) staining (Sigma–Aldrich, Shanghai, China) . C. tropicalis cells (5 × 106 CFU/ml) in the mid-log phase were incubated with 3.9 μg/ml CGA-N9 for 4, 8, 12 and 16 h at 28°C. C. tropicalis cells that had not undergone CGA-N9 treatment were used as a control. After washing with PBS (10 mM, pH 7.2), the cells were stained with 50 μg/ml DAPI for 20 min in the dark. The samples were then observed using a laser scanning confocal microscope (Olympus FA100, U.S.A) at λex/λem = 358 nm/461 nm.
The terminal dexynucleotidyl transferase (TdT)-mediated dUTP nick-end labelling (TUNEL) assay was employed to detect chromosome fragmentation . A total of 5 × 106 CFU/ml C. tropicalis cells in log phase were incubated with 3.9 μg/ml CGA-N9 at 28°C for 4, 8, 12 and 16 h. After washing with PBS (10 mM, pH 7.2), the cell pellets were fixed with 3.6% paraformaldehyde for 30 min. After incubation with 0.3% Triton X-100, the cell pellets were treated with 50 μl TUNEL reaction mixture for 2.5 h at 37°C in darkness. After washing, the samples were observed using a laser scanning confocal microscope (Olympus FA100, U.S.A.) at λex/λem = 495 nm/519 nm. C. tropicalis cells treated with PBS (10 mM, pH 7.2) for 16 h were used as a negative control, and cells treated with 2.5 mM H2O2 for 16 h were used as a positive control.
The fluorescence intensity was assessed by fluorescence spectrophotometry (λex/λem = 495 nm/519 nm) to quantitate DNA labelling.
Gel retardation test
A gel retardation test was used to assess the effect of CGA-N9 on DNA structure and binding . Briefly, 100 ng/μl chromosomal DNA isolated from C. tropicalis cells was incubated with CGA-N9 at 1 × minimum inhibitory concentration (MIC), 2 × MIC, 4 × MIC, 8 × MIC, 16 × MIC, and 32 × MIC at room temperature for 12 h. The samples were subjected to gel electrophoresis at 50 V for 30 min. The gel was stained with EB. The migration of DNA was observed under ultraviolet light at a wavelength of 302 nm.
Ultraviolet spectrum assay
CGA-N9 at 3.9 μg/ml was incubated with C. tropicalis chromosomal DNA (0 ng/μl, 1.25 ng/μl, 2.5 ng/μl, 5 ng/μl, 10 ng/μl, and 20 ng/μl) for 1 h. The absorbance of the mixture was recorded by an ultraviolet spectrophotometer at wavelengths of 190–350 nm .
Data are presented as the mean ± standard deviation from three independent experiments. Statistical significance was determined by Student's t-test. P values <0.05 and <0.01 indicate statistical significance.
Cell membrane depolarization
The membrane potential-sensitive fluorescent probe 3,3′-dipropylthiadicarbocyanine iodide (DiSC3(5)) concentrates in the cytoplasmic membrane in a membrane potential-dependent manner, which results in the self-quenching of the fluorescence. Once the membrane is depolarized, the dye dissociates into the buffer solution, which causes an increase in fluorescence intensity . As shown in Figure 1, there was a sharp increase in the fluorescence intensity after the addition of DiSC3(5). The fluorescence intensity quickly decreased at the end of the DiSC3(5) addition period because of the self-quenching of the fluorescence. There was a balance of DiSC3(5) monomers and polymers. The fluorescence intensity tended to be stable until CGA-N9 depolarized the cell membrane and disturbed the balance of DiSC3(5) monomers and aggregates. The dye dissociated into the buffer, which caused an increase in the fluorescence intensity.
Membrane depolarization of C. tropicalis post-CGA-N9 treatment.
Disruption of calcium homeostasis
Cell membrane depolarization increases the cellular uptake of calcium, which disrupts cellular calcium homeostasis . The Ca2+-sensitive fluorescent dyes Fluo-4AM (cytoplasmic) and Rhod-2AM (mitochondrial) were used to measure cytosolic and mitochondrial Ca2+ levels, respectively. Compared with those of the control cells, cytosolic and mitochondrial Ca2+ levels increased in a time-dependent manner in CGA-N9-treated cells and notably increased at 8 h post-treatment with CGA-N9 (Figure 2), indicating that CGA-N9 induces an influx of Ca2+ into cells to disrupt mitochondrial and cytosolic Ca2+ homeostasis .
Effect of CGA-N9 on Ca2+ levels of C. tropicalis.
CGA-N9 induces ROS accumulation
Calcium homeostasis dysregulation causes excess ROS accumulation in cells, which induces mitochondrial dysfunction and apoptosis [27,28]. The ROS level in C. tropicalis cells post-treatment with CGA-N9 was evaluated by assessing DHR123 fluorescence intensity, whereby an increase in DHR123 intensity reflects an increase in ROS production. Compared with untreated control cells, CGA-N9-treated cells displayed a sharp increase in ROS accumulation, which peaked at 8 h post-treatment (Figure 3). These results indicate that CGA-N9 treatment-induced ROS accumulation in C. tropicalis cells. Therefore, CGA-N9 promotes apoptosis in yeast cells by inducing ROS accumulation.
ROS accumulation in C. tropicalis cells 8 h post-treatment with CGA-N9.
Mitochondrial membrane potential disruption
The dysregulation of mitochondrial Ca2+ homeostasis is now recognized as a key effect of ROS accumulation and increased mitochondrial membrane permeability in several pathologies [25,29]. The lipophilic cationic dye JC-1 exists as aggregates (red fluorescence, 595 nm) in cells with polarized mitochondria but as monomers (green fluorescence, 525 nm) in apoptotic cells with mitochondrial membrane potential disruption . A decrease in the ratio of aggregates to monomers reflects the level of mitochondrial membrane potential disruption. In this study, the fluorescence of JC-1 aggregates (FL2) and monomers (FL1) was detected by flow cytometry. The fluorescence ratios of FL2/FL1 in cells treated with CGA-N9 and H2O2 for 8 h were 1.41 and 0.85, respectively, which is lower than that (1.72) of the control cells (Figure 4). These results indicate that CGA-N9 induces mitochondrial membrane potential disruption.
Mitochondrial membrane potential disruption in C. tropicalis post-treatment with CGA-N9.
Cytochrome c release from mitochondria into the cytosol
The release of Cyt c from mitochondria into the cytosol is an important feature of mitochondrial dysfunction, which is followed by apoptosis [17,31]. The changes in the levels of cytosolic and mitochondrial Cyt c were quantified according to the recorded ultraviolet absorbance. As shown in Figure 5, the relative quantity of Cyt c in the cytoplasm increased, while the relative quantity of Cyt c in mitochondria decreased in response to CGA-N9 treatment. At 12 h after treatment of C. tropicalis cells with CGA-N9, Cyt c leakage into the cytoplasm from mitochondria reached the maximal level.
CGA-N9 induces Cyt c release from mitochondria into the cytoplasm.
C. tropicalis cells were stained with the membrane-penetrating dye DAPI, which binds to DNA . The results are shown in Figure 6. DAPI-bound chromatin DNA was distributed homogeneously at the early stage of CGA-N9 treatment. Upon incubation with CGA-N9, the DAPI fluorescence condensed in the cells, which indicates chromatin condensation in C. tropicalis cells. This result indicates a late apoptotic stage phenotype in C. tropicalis cells after treatment with CGA-N9.
Detection of nuclear DNA condensation in C. tropicalis cells treated with CGA-N9.
The degradation of nuclear DNA produces free 3′-OH. Under the catalysis of terminal deoxynucleotidyl transferase, exposed free 3′-OH produced by DNA fragmentation reacts with FITC-labelled dUTP . As shown in Figure 7, increased dUTP-FITC fluorescence in C. tropicalis cells was observed after treatment with CGA-N9 in a time-dependent manner. This result demonstrates that chromosomal DNA degraded after treatment with CGA-N9, as the dUTP-FITC conjugate binds to free the 3′-OH of the degraded DNA fragment.
Fluorescence intensity of the damaged DNA assessed by TUNEL staining.
Interaction of CGA-N9 with DNA
The interaction of CGA-N9 with chromosomal DNA was examined in vitro. DNA electrophoretic migration was performed to assess the binding of CGA-N9 to DNA. After incubation with CGA-N9, there was no DNA migration retardation, and DNA ladder bands were observed even up to a CGA-N9 concentration of 8 × MIC. However, no DNA mobility was observed on the gel when the CGA-N9 concentration was greater than 16 × MIC (Figure 8). This result demonstrates that CGA-N9 does not damage the DNA structure and inhibits its separation by gel electrophoresis only at high concentrations. We speculate that CGA-N9 binds only weakly to DNA.
Interaction of DNA with CGA-N9.
There will be a redshift of characteristic absorption if a small molecule is embedded in DNA . To investigate the direct interaction between CGA-N9 and DNA, ultraviolet spectrometry was performed to detect the spectral changes of CGA-N9. Compared with the control absorption spectrum, the absorption spectrum of CGA-N9-bound DNA increased, but a redshift was not obvious (Figure 9). Therefore, groove and intercalation binding modes of CGA-N9 with DNA were excluded. Combined with the result of the gel retardation assay, this result suggests that electrostatic interactions formed between positively charged CGA-N9 and negatively charged DNA, and the complex blocked the mobility of the DNA during gel electrophoresis.
UV absorbance spectra of CGA-N9.
Our previous research demonstrated that most CGA-N9 passes through the C. tropicalis cell membrane via direct cell penetration . Herein, the interaction of CGA-N9 and cell membrane disturbed the cell membrane permeability.
Apoptosis is a reported mechanism of many AMPs [18,35,36]. Ca2+ is a significant regulator of various cellular processes in all eukaryotic cells and a core component of the mitochondrial functional effect [25,29]. Cell membrane depolarization induces Ca2+ uptake via the plasma membrane voltage-gated Ca2+ channel (Cch1/Mid1 complex) and other unknown transporters [25,37]. The specific Ca2+ transport system in the mitochondria membrane aids Ca2+ influx into mitochondria. Ca2+ homeostasis is often regarded as the initial signal of apoptosis, and oxidative stress is closely related to excessive accumulation of Ca2+ in cells [25,29]. Conversely, overloading of Ca2+ stimulates oxidative phosphorylation of mitochondria, which increases mitochondrial activity stress; consequently, mitochondrial oxygen consumption increases, resulting in increased ROS production [29,37,38]. CGA-N9 disturbs calcium haemostasis in a time-dependent manner, which corresponds to its bactericidal kinetics .
The accumulation of Ca2+ induces an increasein intracellular ROS levels. Ca2+ and ROS are two key signalling molecules mediating apoptosis, which can cause mitochondrial dysfunction. Ca2+-induced ROS production is an early apoptotic event . In addition to damaging mitochondria, ROS can also cause DNA fragmentation, nuclear damage and dysfunction of macromolecular substances, such as proteins, leading to cell death [19,39]. Because of the lack of a Ca2+ unidirectional transporter (MCU) in the yeast mitochondrial membrane  for the rapid balance of Ca2+ between the cytoplasm and the mitochondrial matrix, Ca2+ transfers through the mitochondrial permeability transition pore (mPTP). Increased Ca2+ uptake eventually leads to the continuous opening of the mPTP and disturbed mitochondrial Ca2+ homeostasis, resulting in mitochondrial extracorporeal membrane swelling and the release of intermembrane proteins, such as Cyt c, and other mitochondrial content . In the present study, after treatment with CGA-N9, we detected ROS accumulation and mitochondrial Ca2+ homeostasis dysregulation in C. tropicalis cells.
Mitochondria are important cellular organelles. Increased cytosolic Ca2+ and ROS generation are triggering signals that lead to mitochondrial permeabilization and the release of proapoptotic factors [25,29,40]. The disruption of mitochondrial membrane potential is closely related to the production of ROS in mitochondria . Endogenous oxidative stress directly causes the opening of the mPTPs, followed by the triggering of a series of apoptotic events, such as the leakage of the apoptotic signalling molecule Cyt c and certain inducible apoptotic factors into the cytoplasm, which act on the corresponding targets and ultimately induce apoptosis [26,42–44]. In this study, the increased cytosolic Ca2+ influx and ROS accumulation after CGA-N9 treatment promoted the disruption of the mitochondrial membrane potential, which is a sign of early apoptosis .
Cyt c release from the outer surface of mitochondria into the cytoplasm is a key signalling event in the apoptosis pathway . Although certain studies have suggested that Cyt c is involved in apoptosis in yeast [16,47], it is not clearly understood whether Cyt c participates in yeast apoptosis in the same way that it participates in apoptosis in mammalian cells . No evidence has demonstrated that Cyt c can initiate the caspase activation cascade once released into the cytosol by active metacaspase in yeast .
Intracellularly accumulated ROS mainly attack the nucleic acids of chromosomes, resulting in single- or double-stranded DNA breaks [28,43,48]. DNA fragmentation and nuclear condensation are typical features of late apoptosis [41,49]. After C. tropicalis cells were treated with CGA-N9, nuclear condensation became apparent. Therefore, C. tropicalis cell death due to CGA-N9 shares many comparable features with yeast apoptosis. The in vitro investigation of the interaction of DNA with CGA-N9 showed that CGA-N9 did not damage the integrity of the DNA but bound weakly to the DNA through electrostatic attraction. Electrostatic binding, as one of the three non-covalent binding modes between DNA and small molecules, is a non-selective interaction between the DNA phosphoric acid skeleton and positive molecules [24,34,50].
In summary, CGA-N9 in C. tropicalis cells induces dysregulation of calcium homeostasis, which results in intracellular ROS accumulation. Under oxidative stress, mitochondrial membrane permeability increases, allowing an influx of Ca2+ into mitochondria and an efflux of Cyt c from mitochondria, leading to the disruption of the mitochondrial membrane potential, which is also a sign of apoptosis. ROS accumulation further induces chromatin condensation, which represents the late apoptosis. Therefore, CGA-N9 induces apoptosis in C. tropicalis by attenuating mitochondrial function.
a chromogranin A-derived fragment, consisting of the 47th to the 55th amino acids of the N-terminus
China General Microbiological Culture Collection Center
- Cyt c
the first fluorescence channel
minimum inhibitory concentration
mitochondrial permeability transition pore
reactive oxygen species
side scatter channel
Ruifang Li drafted the manuscript and participated in the design and coordination of the experiments. Chen Chen carried out the experiments. Beibei Zhang provided guidance for the experiments. Hongjuan Jing and Zichao Wang participated in the final editing of the manuscript. Chunling Wu and Pu Hao assisted Chen Chen with the confocal microscopy experiments. Yong Kuang and Minghang Yang participated in the peptide synthesis. All authors have read and approved the final manuscript.
This study was supported by the National Natural Science Foundation of China (31572264 and 31071922), the Innovative Research Team (in Science and Technology) at the University of Henan Province (19IRTSTHN008), and the National Engineering Laboratory for Wheat & Corn Further Processing, Henan University of Technology (NL2016010).
The Authors declare that there are no competing interests associated with the manuscript.