Antimicrobial peptide CGA-N12 decreases the Candida tropicalis mitochondrial membrane potential via mitochondrial permeability transition pore

Candida tropicalis is the most common non-albicans Candida isolated from patients with candidiasis [1,2], and exhibits increased levels of fluconazole resistance [3,4]. Thus, it has become urgent to find novel and effective anti-Candida agents against the drug-resistant Candida species.

Antimicrobial peptides (AMPs) are an essential component of the defense system and have shown a broad spectrum of antimicrobial activity, high safety, low drug resistance and killing activity by multiple mechanisms [5–7]. AMPs are potent candidates for clinical infections.

Chromogranin A (CGA) exists in the secretory granules of many endocrine and neuroendocrine cells [8]. Lugardon et al. found that the highly conserved N-terminus of CGA has antibacterial activity [9]. In our previous work, we found that amino acid sequence from 65th to 76th residue of the N-terminus of CGA (CGA-N12), a more potent and stable derivative of CGA that corresponds to the N-terminal Ala55-Gln76 sequence, exhibits antagonistic activity against Candida species [10]. Our previous studies clarified that CGA-N12 induces mitochondrial depolarization, and cytochrome c (Cyt c) leakage, ultimately causing mitochondria-dependent apoptosis of C. tropicalis [11]. However, the mechanism by which CGA-N12 decreases mitochondrial potential is unclear.

The mitochondrial permeability transition pore (mPTP), a complex protein pore of the inner mitochondrial membrane (IMM) and outer mitochondrial membrane (OMM), has attracted much attention in regulating mitochondrial membrane permeability [12]. From a representative diagram of the proposed models of the mPTP, known 3-dimensional (3D) structures obtained from the Protein Data Bank (http://www.rcsb.org/pdb) [13,14], Bcl-2 associated X protein (Bax), hexokinase (HK) II, outer membrane voltage-dependent anion channel (VDAC), mitochondrial creatine kinase (mtCK), peripheral benzodiazepine receptor (TSPO), adenine nucleotide transporter (ANT) and cyclophilin-D (CyP-D) formed the mammalian mPTP complex at membrane contact sites, where VDAC and ANT formed the core pore in the OMM and IMM, respectively. ANT dimers exist in two conformations, the matrix-facing (m) and cytosolic-facing (c) conformations in the IMM. The mitochondrial permeability transition activity is determined by the orientation of the nucleotide site of ANT. The m conformation of ANT favors mPTP closure, while c conformation favors mPTP opening. When ANT is inhibited with bongkrekic acid (BA), which locks the ANT in the m conformation, it opposes mPTP opening. Mitochondrial Cyp-D bound to ANT in the matrix [15]. It was identified as a key pore opening regulator through its cyclosporin A (CsA)-sensitive interaction with the pore complex [15].

mPTP regulates the permeability of the mitochondrial membrane [13,15]. Once irreversible, mPTP opening occurs, the mitochondrial matrix osmotically swells, the OMM is damaged, and the release of Cyt c, leading to cell death [13,16].

Herein, the actions of CGA-N12 on C. tropicalis mPTP complex were investigated. We analyzed the effects of CGA-N12 on mPTP by measuring the level of mitochondrial swelling, polyethylene glycol (PEG) size across opening mPTP and Cyt c release.

CGA-N12 (H-ALQGAKERAHQQ-OH) was synthesized using a solid-phase method and purified by high-performance liquid chromatography. The anti-Candida activity of CGA-N12 was assessed by the broth micro-dilution method to define the minimum inhibitory concentration (MIC), which was 75 μM [10].

C. tropicalis (CGMCC2.3739) was supplied by the China General Microbiological Culture Collection Center (Beijing, China), subcultured on to Sabouraud Dextrose Agar at 28°C for 16 h, and maintained at 4°C for short-term storage.

Mitochondria were isolated from C. tropicalis cells following differential centrifugation procedures as previously described [17]. The isolated mitochondria were suspended in buffer C (0.6 M mannitol, 20 mM HEPES (K⁺), pH 7.35, 0.05% BSA, 0.1 mM EGTA) and stored on ice. The mitochondrial protein amount was determined by the bicinchoninic acid (BCA) method using BSA as a standard.

Ultrastructural examination of isolated mitochondrial preparations was performed using transmission electron microscopy (TEM). Mitochondria pellets were fixed in a 2.5% solution of glutaraldehyde [18] and then in a 1% solution of osmium tetroxide. After dehydrating and embedding in resin, ultra-thin sections were prepared and stained with uranyl acetate followed by lead citrate. The specimens were observed by TEM (FEI Tecnai G20, FEI, U.S.A.).

The effect of CGA-N12 on the mitochondrial membrane potential was evaluated with JC-1 (Beyotime, Shanghai, China) followed the procedure provided by the manufacturer. Mitochondria (100 μg protein) in buffer C were pretreated with 1× MIC CGA-N12 for 1 min, and then added the diluted JC-1 working solution. After incubation at 28°C for 3 min, fluorescence at λex/λem = 485/590 nm was recorded with a fluorescence spectrophotometer (Hitachi F-4600, Hitachi, Japan) to detect the changes of mitochondrial membrane potential within 10 min. Mitochondria not treated with CGA-N12 were set as a negative control, and 1 mM ethanol was set as a positive control [19]. Each experiment was performed in triplicates.

Mitochondrial swelling was measured by monitoring the absorbance at 540 nm within 10 min at room temperature as described previously [20] with minor modifications. Mitochondria (400 μg/ml protein) were suspended in 1 ml of measuring solution (300 mM mannitol, 10 mM HEPES (TEA⁺), pH 7.35, 0.2 mM EGTA, 0.5 mg/ml BSA, and 15 μg of oligomycin/mg of protein) and incubated with CGA-N12 at concentrations of 0× MIC, 0.25× MIC, 0.5× MIC, 0.75× MIC and 1× MIC. Spectra were recorded on a Shimadzu UV 2450 double-beam spectrophotometer. Mitochondria not incubated with CGA-N12 were used as a negative control, and 1 mM ethanol was used as a positive control [19]. Each experiment was performed in triplicates.

In the presence of the mPTP-opening inhibitors CsA and BA, the effects of CGA-N12 on mPTP opening were detected to determine the effect of CGA-N12 on mPTP complex [21,22]. Each experiment was repeated three times.

The solute exclusion method was performed by measuring the pre-swollen mitochondria shrinkage caused by the osmotically active solute PEG to estimate the size of the mPTP [19,23]. Once swelling induced by 1× MIC CGA-N12 was stopped, shrinkage was promoted by the addition of different molecular weight PEGs and followed spectrophotometrically [19]. A total of 300 mOsm (milliliter/liter) PEG storage solutions in 10 mM HEPES (Na⁺), pH 7.35 was added into the measurement solution till 10% of the final volume to form an iso-osmotic solution. To get 300 mOsm PEG stock solutions, the concentrations of different molecular weight PEGs was: 199 mM for 0.4 kDa, 172 mM for 0.6 kDa, 136 mM for 1.0 kDa, 111 mM for 1.5 kDa PEG and 93 mM for 2.0 kDa. The osmotic pressure of these solutions was confirmed using a vapor pressure osmometer. Each experiment was repeated three times.

The effect of CGA-N12 on mPTP opening and the action on mPTP complex were further determined by the solute exclusion of mPTP in the presence of the mPTP-opening inhibitors CsA and BA. First, 1.0 kDa PEG (PEG1000) was incubated with pre-swollen mitochondria treated by CGA-N12. The absorbance at 540 nm were continuously recorded to detect the shrinkage of the mitochondria. Each experiment was performed in triplicate.

The cytosolic Cyt c contents of C. tropicalis cells post-treatment with CGA-N12 at different concentrations were measured to test the leakage of Cyt c from mitochondria according to the reported method [24]. C. tropicalis cells were incubated with 0× MIC, 0.25× MIC, 0.5× MIC, 0.75× MIC and 1× MIC CGA-N12 for 12 h at 28°C; 1 mM ethanol was set as control.

SPSS version 21.0 (IBM, U.S.A.) was used for the statistical analysis (ANOVA and Tukey’s test). All data are presented as the mean ± standard deviation. A P-value <0.05 indicated a difference, P-value <0.01 indicated significant difference, and a P-value <0.001 indicated an extremely significant difference.

Isolated mitochondrial ultrastructure was observed under TEM. As shown in Figure 1, the mitochondria which were not treated with CGA-N12 maintained their integrity, showing classical ultrastructure containing well-defined outer and inner membranes, a narrow intermembrane space and compact cristae (Figure 1A). After treatment with CGA-N12, the mitochondrial ultrastructure was damaged with the disappearance of cristae and a loss of contents, but the outer membrane kept its integrity (Figure 1B). The results indicated that CGA-N12 induced the loss of the substance in the mitochondria.

JC-1, a mitochondrial membrane potential dye, exists as aggregates in polarized mitochondria. Mitochondrial depolarization leads to the formation of JC-1 monomer [25]. We determined the fluorescence intensity of JC-1 aggregates by using a fluorescence spectrophotometer (λex/λem = 485/590 nm). The results showed that, compared with that of the control, the fluorescence intensity of mitochondria treated with CGA-N12 decreased in a time-dependent manner (Figure 2). CGA-N12 perhaps induces mitochondrial membrane potential dissipation similarly to ethanol, which could induce mPTP opening [19].

Cyt c leakage is considered an important indicator of mitochondrial permeability, which is regulated by OMM mPTP opening [26,27]. The effect of CGA-N12 on mPTP opening was determined by measuring the leakage of Cyt c. As shown in Figure 3, CGA-N12 induced the leakage of Cyt c from mitochondria in a dose-dependent manner. The results demonstrated that CGA-N12 could induce OMM mPTP opening.

Mitochondrial swelling was evaluated in light of the absorbance decrease at 540 nm (A₅₄₀) within 10 min. The effects of CGA-N12 on mitochondrial swelling are shown in Figure 4. CGA-N12 induced mitochondria swelling in a dose-dependent manner.

No shrinkage happens when PEGs could penetrate mitochondria via mPTP. The molecular weight of PEG reveals a size exclusion property of mPTP. Therefore, mPTP opening size could be estimated by the pre-swollen mitochondria contraction caused by PEG of different molecular weights. The pre-swollen mitochondria induced by CGA-N12 were treated with PEGs of various molecular weights under iso-osmotic conditions to determine their effectiveness in causing shrinkage. According to the results (Figure 5A), PEG400 (0.4 kDa) and PEG600 (0.6 kDa) readily passed through mPTP. Significant mitochondrial contraction was observed after the addition of PEG with a molecular weight more than 1.0 kDa, indicating that PEG1000 was excluded or passed through the open mPTP at a decreased rate. The contraction level induced by 1.5 kDa PEG was almost the same as that of 2.0 kDa PEG, which inferred that 1.5 kDa may be the maximum molecular mass of a PEG that can penetrate the mPTP when the yeast mPTP opens at an irreversibly high level. The relationship between the contraction levels and PEG sizes showed that the half-maximum contraction effect of swollen mitochondria induced by CGA-N12 at a 1× MIC concentration was 0.9 kDa PEG (Figure 5B).

Based on the results of solute size exclusion, PEG1000 was chosen to verify the effects of CGA-N12 on mPTP. The opening size of mPTP induced by CGA-N12 at different concentrations was determined by the swollen mitochondria contraction caused by PEG1000. The results showed that contraction degree of the swollen mitochondria increased in a concentration-dependent manner (Figure 5C). The relationship between the contraction rate and swelling rate of mitochondria treated with CGA-N12 showed that the swelling of the mitochondria induced by CGA-N12 at less than 1× MIC was nearly proportional to the shrinkage induced by PEG1000 (Figure 5D).

The effect of CGA-N12 on mPTP complex was further explored by using mPTP opening-inhibitors CsA and BA. Compared with the control, CsA treatment did not shrink mitochondrial swelling induced by CGA-N12 (Figure 6A), and there was a rapid shrinkage of the pre-swollen mitochondria caused by PEG1000 (Figure 6B). BA significantly shrank mitochondrial swelling induced by CGA-N12 (Figure 6A), and the mitochondrial shrinkage rate caused by PEG1000 was slowest (Figure 6B). The results proved that CsA could not reverse mPTP opening induced by CGA-N12, while BA could. Therefore, we speculated that CGA-N12 prevented ANT from the m conformation, while had no effect on the Cyp-D binding with CsA.

The mPTP complex, a nonselective channel across the IMM and OMM, is a key participant that controls mitochondrial fate and cell fate. mPTP opening is involved in various cellular processes, including membrane potential dissipation. It has been found that mPTP opening presents two distinct states: a low- and a high-conductance state. In the low-conductance state, the molecular weight cut-off value of mPTP is below 300 Da. Under such conditions, the mPTP opens transiently (‘lickering’), and mitochondrial swelling is absent [28,29]. In contrast, in the irreversible high-conductance state, the molecular weight cut-off value of the mPTP increased, allowing ions and solutes smaller than 1.5 kDa to passively diffuse across the IMM. Upon formation solutes ≤1.5 kDa in size can cross the IMM, resulting in organelle swelling and eventual rupture, a key event in necrotic cell death [30]. In the present study, the half-maximum contraction effect of swollen mitochondria induced by CGA-N12 at a 1×MIC concentration was 0.9 kDa PEG. PEG1000 passed through the mPTP post-treatment with CGA-N12. Therefore, CGA-N12 induced a high-conductance state in mPTP.

Cyt c in mitochondria cannot pass through the OMM unless mPTP irreversibly opens [26,27]. The mPTP opening led to a release of Cyt c [31]. In the present study, CGA-N12 induced Cyt c leakage, which indicated that CGA-N12 could open mPTP in the OMM.

Carboxyatractyloside (CAT), which stabilizes ANT in the c conformation, favors mPTP opening. While BA, which stabilizes ANT in the m conformation, favours mPTP closure [32]. In our research, the complete protection by BA against CGA-N12-induced mitochondrial swelling indicated that CGA-N12 opens mPTP by preventing ANT from the m conformation.

CyP-D controls mPTP opening by regulating its calcium sensitivity [33], while CsA is the typical inhibitor of CyP-D in mammalian cells [34]. In our research, CsA has no protective effect on the mitochondrial swelling induced by CGA-N12, although CsA is considered as a classical inhibitor of mammalian mPTP [34,35].

The authors declare that there are no competing interests associated with the manuscript.

This work was supported by the National Natural Science Foundation of China [grant numbers 31572264, 31071922]; the Innovative Research Team (in Science and Technology) at the University of Henan Province [grant number 19IRTSTHN008]; and the National Engineering Laboratory for Wheat and Corn Further Processing, Henan University of Technology [grant number NL2016010].

R.L. and J.Z. drafted the manuscript and participated in the design of the experiments. J.Z. carried out the experiments. L.H. took part in the results statistical analysis. Y.Y. and A.L. participated in the final editing of the manuscript. D.L. and M.T. assisted J.Z. with the isolation of mitochondria. Y.L. participated in the peptide synthesis. All authors have read and approved the final manuscript.

AMP

antimicrobial peptide

ANT

adenine nucleotide transporter

BA

bongkrekic acid

CGA

chromagranin A

CGA-N12

amino acid sequence from 65th to 76th residue of the N-terminus of CGA

CsA

cyclosporin A

CyP-D

cyclophilin-D

Cyt c

cytochrome c

IMM

inner mitochondrial membrane

MIC

minimum inhibitory concentration

mPTP

mitochondrial permeability transition pore

OMM

outer mitochondrial membrane

PEG

polyethylene glycol

PEG1000

1.0 kDa PEG

TEM

transmission electron microscopy

VDAC

voltage-dependent anion channel