Novel Self-Assembled Micelles Based on Cholesterol-Modified Antimicrobial Peptide (DP7) for Safe and Effective Systemic Administration in Animal Models of Bacterial Infection

Owing to their broad-spectrum antibacterial properties, multitarget effects, and low drug resistance, antimicrobial peptides (AMPs) have played critical roles in the clinical therapy of drug-resistant bacterial infections. However, the potential hazard of hemolysis following systemic administration has greatly limited their application.

ABSTRACT

Owing to their broad-spectrum antibacterial properties, multitarget effects, and low drug resistance, antimicrobial peptides (AMPs) have played critical roles in the clinical therapy of drug-resistant bacterial infections. However, the potential hazard of hemolysis following systemic administration has greatly limited their application. Here, we developed a novel AMP derivative, DP7-C, by modifying a formerly identified highly active AMP (DP7) with cholesterol to form an amphiphilic conjugate. The prepared DP7-C easily self-assembled into stable nanomicelles in aqueous solution. The DP7-C micelles showed lower hemolytic activity than their unconjugated counterparts toward human red blood cells and a maximum tolerated dose of 80 mg/kg of body weight in mice via intravenous injection, thus demonstrating improved safety. Moreover, by eliciting specific immunomodulatory activities in immune cells, the DP7-C micelles exerted distinct therapeutic effects in zebrafish and mouse models of infection. In conclusion, DP7-C micelles may be an excellent candidate for the treatment of bacterial infections in the clinic.

INTRODUCTION

Antimicrobial peptides (AMPs) are important contributors to the natural defense system against microbial infections in multicellular organisms (1–4). Currently, numerous natural AMPs and synthetic analogues have been developed for either the preclinical or clinical trial stage (5–9). However, most of these analogues are restricted to topical application only because hemolytic side effects may occur following systemic administration, which is a great limitation to their use (see Table S1 in the supplemental material) (10–21). To overcome this issue, various formulations encapsulating AMPs, such as liposomes and microgels, have been developed, and these liposomes and microgels pose fewer risks (10, 11). However, it is widely recognized that the organic solvents involved in the preparation process may be cytotoxic to the human body. In addition, the manufacturing methods are not simple enough for further production and application. Hence, there has been a great demand to develop novel strategies with high safety and efficacy for the systemic administration of AMPs.

In our previous studies, we developed a novel 12-amino-acid cationic and hydrophilic antimicrobial peptide, DP7, using an amino acid-based activity prediction method based on HH2, an antimicrobial peptide from bovine neutrophils (22). This newly discovered AMP demonstrated reduced hemolytic activity, excellent antibacterial effects in combination with antibiotic effects, and increased antimicrobial activity both in vitro and in vivo (22). Further morphological studies suggested that its antibacterial effect is a result of disruption of the outer membrane of bacteria, such as Staphylococcus aureus, with positive charges (22). However, a high dose of DP7 caused hemolysis and vascular irritation when administered intravenously (i.v.) because it disrupted the membranes of red blood cells. This deficiency limits the wide application of DP7. Although several efforts have been made to optimize this method, including packaging DP7 into liposomes, and ideal antimicrobial effects were obtained, the stability and safety of the DP7 formulation are still important issues (23).

Cholesterol (Chol) is one of the most commonly used hydrophobic compounds, with a planar multicycle unit and a flexible aliphatic chain. This unique property provides cholesterol with the ability to form nanostructures and promotes the interaction between cholesterol and amphiphilic lipid membranes with high binding capabilities (24, 25). In addition, cholesterol has been used to modify chemical drugs to promote drug delivery and reduce toxicity (24–28). Thus, given the hydrophilicity of DP7 and the unique properties of cholesterol, we hypothesized that an amphiphilic polymer could be synthesized by conjugating DP7 with cholesterol, with further self-assembly into micelles, producing a systemic injectable formulation with high safety and antimicrobial efficacy. To our knowledge, there have been few investigations regarding cholesterol-modified AMPs and their nanoformulations.

In the current study, we synthesized a conjugate of cholesterol and DP7 (DP7-C). Interestingly, in vitro antibacterial and hemolysis assays demonstrated that both the hemolytic and antibiotic capacities of DP7-C were significantly decreased. However, the DP7-C micelles exerted potent therapeutic benefits in multiple in vivo models of systemic infectious disease and considerable safety via intravenous injection. Investigation into the molecular mechanism further suggested that DP7-C could regulate immune responses in conjunction with their direct antibacterial activities. Our study demonstrates that the newly developed DP7-C possesses good antibacterial efficacy and does not produce obvious side effects following systemic administration, which suggest that DP7-C could be an excellent candidate for the treatment of bacterial infections in the clinic.

RESULTS

In vitro studies of self-assembled DP7-C homogeneous micelles.

DP7-C and AntpHD43-58 (penetratin) conjugated to cholesterol (penetratin-C) were synthesized using standard solid-phase peptide synthesis (SPPS) protocols (initiated with RinkMBHA resin that was loaded with lysine) on a CSBio 136XT peptide synthesis instrument (Fig. 1) (21). After removal of the 9-fluorenylmethoxy carbonyl (Fmoc) and coupling of the subsequent amino acids, the monocholesteryl ester of succinic acid was linked to a penetrating peptide attached to the RinkMBHA resin. The Chol-peptide conjugate was segregated from the resin with trifluoroacetic acid (TFA)-triisopropylsilane (TIS)-H₂O using the Fmoc peptide synthesis method to produce a crude product. The Chol-peptides were purified, using high-performance liquid chromatography (HPLC), to 99% purity, and the molecular weights were confirmed by mass spectrometry (Fig. 1).

FIG 1.

Synthesis route and structure of DP7-C. The synthesis route of penetratin-C is the same as that of DP7-C. L, lysine; AA, amino acid; suc, sucrose; HBTU, 2-(1H-benzotriazole-1-yl)-1,1,3,3-tetramethylaminium hexafluorophosphate; DIPEA, N,N-diisopropylethylamin.

As shown in Fig. S1 in the supplemental material, the increasing concentrations of DP7-C resulted in a substantial increase in the intensity ratio over a certain range of concentrations, which suggested that pyrene probes were incorporated into the hydrophobic core upon micelle formation. The critical micelle concentrations (CMC) were determined from the crossover point at the low end of the concentration range. It is well-known that plotting the intensity ratios of the first (374 nm) to the third (384 nm) vibronic peaks of pyrene (the I1/I3 ratio) as a function of the total surfactant concentration manifests as a typical sigmoidal decrease close to the CMC (29). Below the CMC, the pyrene I1/I3 ratio value corresponded to a polar environment, whereas the pyrene I1/I3 ratio declined rapidly when the surfactant concentration increased, which indicated that the pyrene was encapsulated into a more hydrophobic environment (29). When the DP7-C concentration was greater than the CMC, the pyrene I1/I3 ratio plateaued at a roughly constant value because of the incorporation of the probe into the hydrophobic region of the micelles. As shown in Fig. S1, the CMC of DP7-C in water was approximately 3.5 μg/ml and the CMC of DP7-C in culture medium was approximately 50 μg/ml, which suggested that DP7-C could self-assemble into micelles at a therapeutic concentration. Moreover, a long-time molecular dynamics (MD) simulation was performed to investigate the self-assembly behavior of DP7-C. To analyze the dynamic changes of DP7-C assembly, 10 DP7-C molecules were simulated as representative aggregates of the DP7-C micelle. The chemical structures of the aggregates before and after molecular simulation were compared, and the structures of the aggregates with or without solute molecules are shown in Fig. S2. The computational result was in agreement with our experimental result, and DP7-C easily formed stable aggregates.

Particle size, zeta potential, and morphological characteristics of DP7-C.

The DP7-C micelles were prepared in a one-step self-assembly method without using any carriers, organic solvents, or surfactants, which was thus a convenient and scalable preparation method. The average particle size of the prepared DP7-C micelles was 36.06 ± 1.5 nm, with a polydispersity index (PDI) of 0.176 and a zeta potential of 43.8 ± 0.27 mV (Fig. 2A and B). Based on the particle size distribution spectrum shown in Fig. 2A, the DP7-C micelle particle size was narrowly distributed. A scanning electron microscopy (SEM) image and an atomic force microscopy (AFM) image of DP7-C micelles are presented in Fig. 2C and D, respectively, and show that the DP7-C micelles were of a spherical shape in aqueous solution. The diameter of the DP7-C micelles observed by transmission electron microscopy (TEM) and AFM was in agreement with the results of particle size analysis, which demonstrated that the prepared DP7-C micelles were stable and could be well dispersed in aqueous solution.

FIG 2.

Characterization of DP7-C micelles. (A) Particle size distribution (d, diameter); (B) zeta potential; (C) TEM image of DP7-C micelles; (D) atomic force microscopy (AFM) of DP7-C micelles.

The toxicity of DP7 was significantly higher than that of DP7-C.

The toxicity of DP7 and DP7-C micelles was examined via intravenous injection in mice. In brief, 6- to 8-week-old BALB/c female mice were administered 1, 2, 4, 8, 10, 15, 20, 40, or 80 mg/kg of body weight of either DP7 peptide or DP7-C micelles. We observed that all of the mice in the DP7 group treated with a dose of 20 mg/kg died within 10 min. However, the mice treated with DP7-C micelles were still alive at 144 h postinjection, even when they were treated with a dose of 80 mg/kg (Fig. 3A). The major organs of mice treated with 10-mg/ml DP7 or DP7-C micelles were removed for histopathologic analysis. The hematoxylin-eosin (H&E)-stained sections (Fig. 3E) showed that treatment with DP7 induced pulmonary hemorrhage and liver bleeding, whereas treatment with DP7-C micelles showed no obvious toxicity to major organs.

FIG 3.

In vivo safety evaluation of intravenous administration of DP7 and DP7-C. (A) Comparison of the survival rate by intravenous injection of DP7 and DP7-C at different doses. Mice were intravenously treated with 10 mg/kg DP7 and DP7-C for 4 h and then sacrificed. (B) Concentration of total bilirubin (TBIL) in blood collected from mice administered PBS, DP7, or DP7-C. (C and D) Results of routine mouse blood test. WBC, total number of white blood cells; RBC, total number of red blood cells; RDW, platelet volume distribution width; MPV, mean platelet volume; MCH, average hemoglobin content of red blood cells; HGB, hemoglobin concentration; HCT, hematocrit; MCV, average volume of red blood cells; MCHC, erythrocyte mean hemoglobin concentration; PLT, total number of platelets. (E) H&E staining of sections of major organs from mice after treatment with different samples. Arrows, liver and lung bleeding points. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

DP7-C micelles showed low hemolytic effects compared to DP7 both in vivo and in vitro.

In vivo hemolysis assays were performed using 6- to 8-week-old BALB/c female mice, and blood was collected after administration of DP7 (10 mg/kg) or DP7-C micelles (10 mg/kg). The total bilirubin (TBIL) concentration, as an indicator of in vivo hemolysis, was measured. After administration of free DP7, the TBIL concentration significantly increased to 5.03 ± 0.86 μmol/liter, while the concentrations in the control and DP7-C micelle-treated groups were 1.4 ± 0.4 μmol/liter (Fig. 3B). This demonstrates a serious in vivo hemolysis reaction. In contrast, the TBIL concentration in the DP7-C micelle-treated group showed no significant difference from that in the phosphate-buffered saline (PBS)-treated group. In addition, the blood routine in the DP7-C micelle-treated group showed no significant difference from that in the PBS-treated group (Fig. 3C and D), while DP7-treated mice showed significant decreases in the total number of white blood cells, total number of red blood cells, mean platelet volume, hematocrit, and average volume of red blood cells and significant increases in the hemoglobin concentration and the total number of platelets. These data indicate that when DP7 was loaded with cholesterol, the in vivo hemolysis reaction was eliminated.

In vitro hemolysis assays were performed using human erythrocytes collected from venous blood red blood cells (peripheral blood mononuclear cells [PBMCs]). As shown in Fig. 4A, we observed that the DP7-C micelles demonstrated reduced hemolytic properties against human erythrocytes compared to those of DP7 (Fig. 4A). At the highest concentration tested of 1.2 mg/ml, the hemolytic ratio for the DP7-C micelle group was less than 30%, whereas the ratio for the DP7 group was greater than 70%. These results indicate that DP7-C micelles reduced the systemic toxicity of DP7, as observed by reduced human erythrocyte disruption. The PBS-treated group (Fig. 4B, vial a), the positive-control group (Fig. 4B, vial b), the DP7-treated group (Fig. 4B, vial c), and the DP7-C-treated group (Fig. 4B, vial d) were compared under similar conditions. It was apparent that the hemolytic activity of the DP7-C micelles was greatly decreased compared to that of DP7. Interestingly, the erythrocytes in the DP7-C micelle-treated group were suspended but not disrupted. These results suggest that the hemolysis and toxicity of DP7 might derive from its activity as a disruptor of cell membranes. The DP7-C micelles could also bind to the surface of human erythrocytes and stabilize their suspension in the buffer, but this interaction did not disrupt the structure or function of the erythrocytic cell membrane.

FIG 4.

In vitro hemolytic activity of DP7 and DP7-C micelles. (A) The release of hemoglobin was monitored to estimate the degree of erythrocyte lysis caused by the peptide. The x coordinate is the concentration of DP7 or DP7-C in the serial dilution series, and the y coordinate is the degree of hemolysis relative to that of the total hemolysis control (hemolysis assay). (B) Visualization of hemolysis of erythrocytes under different conditions: PBS (a), positive control (b), DP7 (c), and DP7-C (d).

DP7-C micelles effectively suppressed bacterial infection in vivo.

To test the antimicrobial effect of DP7-C micelles in vivo, we used the ATCC 10145-green fluorescent protein (GFP)-infected zebrafish model, which provided a rapid, convenient, and visually evaluable model for the determination of antimicrobial activity. First, Pseudomonas aeruginosa strain ATCC 10145-GFP was administered intraperitoneally (i.p.) into zebrafish embryos. Treatment of the infected zebrafish embryos with DP7-C micelles potently inhibited the amplification of ATCC 10145-GFP in a time-dependent manner, with the amplification of ATCC 10145-GFP being continuously blocked over 18 h after a single dose (Fig. 5A and B). Concurrently, we observed that the normal saline-treated zebrafish larvae had a continuous and significant increase in the level of green fluorescence, while the fluorescence in the DP7-C-treated larvae increased slowly. A correlation between the fluorescence signal intensity and the number of bacterial CFU was observed (Fig. S3).

FIG 5.

In vivo antimicrobial activity of DP7-C micelles. (A) Efficacy of DP7-C micelles in the zebrafish model with PAO1-GFP i.p. infection. (B) Quantification of fluorescence density in each group. (C, D) Efficacy of DP7-C micelles in the mouse model of S. aureus i.p. infection. The horizontal bars show standard deviations. **, P < 0.01 (DP7-C micelles versus normal saline control). (E, F) Survival rate (E) and weight (F) of mice in the mouse model of i.v. S. aureus infection. NS, normal saline; d, number of days; VAN, vancomycin.

DP7-C micelles were compared with vancomycin in a staphylococcal infectious mouse model using i.p. administration of the methicillin-resistant S. aureus (MRSA) strain ATCC 33591. In this study, we used penetratin conjugated to cholesterol (penetratin-C) as the negative control. Penetratin-C was reported in previous research (30). It has a secondary structure, isoelectric point (pI), and micelle potential similar to those of DP7-C. Here, we used it for the negative control to determine that the in vivo immune regulation and antibacterial effects were not caused by the peptide conjugated with cholesterol or by a micelle. The synthesis method was similar to that for DP7-C. As shown in Fig. 5C and D, at a given therapeutic dosage, the average numbers of CFU of S. aureus in both the DP7-C micelle-treated group and the vancomycin-treated group were significantly lower than those in the saline-treated group and the penetratin-C-treated group. The cell-penetrating peptide penetratin was used as a negative control. As shown in Fig. 5D, penetratin-C did not exert antibacterial activity in the murine abdominal infectious model. In addition, the average number of CFU in the vancomycin-treated group was slightly lower than that in the DP7-C micelle-treated group, but this difference was not statistically significant. As indicated in Fig. 5E and F, DP7-C also showed protection in a mouse bloodstream infection model. These results suggest that DP7-C micelles possess satisfactory antimicrobial activity against murine infection.

Potential molecular mechanism of DP7-C micelles.

Based on the aforementioned results, the MIC of the DP7-C micelles was >1,024 μg/ml against Pseudomonas aeruginosa, S. aureus, and Escherichia coli in vitro. However, DP7-C micelles at 1 mg/ml showed effective antibacterial activity in vivo.

After administration of DP7-C micelles or penetratin-C, the myeloid cell population found in the peritoneal cavity changed dramatically. There was an increase in monocytes (Gr1⁺ F4/80⁺; Fig. 6A, upper gate in the right panel, and C). The percentage of monocytes within the leukocyte population that was isolated from the peritoneal cavity is shown in Fig. 6A, C, and E. Our data also showed that macrophages (CD11b⁺ F4/80⁺, Fig. 6B, upper gate in the right panel) and neutrophils (Gr1⁺F4/80⁻, Fig. 6A, lower gate in the right panel) were unchanged. These results indicate that DP7-C micelles are effective in recruiting monocytes.

FIG 6.

Antibacterial mechanism of DP7-C micelles. (A, B) Increasing accumulation of monocytes in the abdominal cavity after intraperitoneal administration of DP7-C micelles or penetratin-C. (C, D) The percentage of monocytes (Gr1⁺ F4/80⁺) in the leukocyte population isolated from the peritoneal cavity is shown. The horizontal bars indicate standard deviations. (E) Proportion of various immune cells. (F) Gene expression differences in cytokines in mouse PBMCs after DP7-C micelle or penetratin-C administration. (G) Antiendotoxin activity of DP7-C micelles. Mouse PBMCs were stimulated with LPS (10 ng/ml) in the presence or absence of DP7-C micelles (200 mg/ml) or penetratin-C (200 mg/ml) for 4 h, and the cells were assessed for cytokine expression by qPCR. The results show the percent inhibition of LPS-induced TNF-α by DP7-C micelles. The results are the mean ± SD from three independent donors. P values were determined by Student's t test. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

We examined the expression of a number of genes at the mRNA level using real-time quantitative PCR (qPCR). When stimulated by DP7-C micelles alone, there were considerable differences in some genes in mouse PBMCs (Fig. 6F). Specifically, interleukin-1β (IL-1β), macrophage colony-stimulating factor (M-CSF), and tumor necrosis factor alpha (TNF-α) were upregulated approximately 5-fold; IL-6 was upregulated approximately 4-fold; and monocyte chemotactic protein 1 was upregulated approximately 3-fold. Thus, treatment with DP7-C micelles resulted in the upregulation of elements of innate immunity, such as cytokines/chemokines, in mouse PBMCs. The altered cell trafficking and cytokine/chemokine levels in mouse models indicated that the immunomodulatory effects of DP7-C micelles are functional in vivo. Since DP7-C micelles upregulated elements of innate immunity, we sought to determine whether the micelles could potentially cause harmful inflammation by assessing their impact on lipopolysaccharide (LPS)-induced TNF-α production in primary human and mouse cells. The results showed that DP7-C micelles may counterbalance immunoreactions rather than either stimulate or suppress them (Fig. 6G). Thus, the DP7-C micelles significantly stimulated defensive immune reactions and exerted an efficient antimicrobial capacity in vivo.

DISCUSSION

During antibacterial therapy, the increase in drug-resistant bacteria poses a serious threat to public health (31–34). Although AMPs have demonstrated broad-spectrum antibacterial activity and efficacious bactericidal effects on multidrug-resistant bacteria, their intrinsic cytotoxicity has impaired their clinical potential (35). As a result, developing antibacterial drugs with high levels of efficacy and safety for clinical application has become a critical goal. Important efforts have been made to diminish the cytotoxicity of AMPs by either adjusting the peptide sequence or introducing chemical modifications (22, 36–38). In addition, nanoformulations of AMPs have also resulted in enhanced antimicrobial activity and decreased toxicity (23, 39).

In our previous study, we packaged DP7, a previously reported broad-spectrum AMP with bacterial outer membrane disruption activities, into liposomes and loaded them with azithromycin for antibacterial experiments. The content of DP7 was only 125 to 500 μg/ml, and the study mainly focused on the combined antibacterial effects of azithromycin and DP7 (23). This prepared neutral liposome was monodispersed with a mean particle size of 100 nm. However, the size of this liposome was not very uniform, and the stability decreased over time. In the current study, in an effort to develop an intravenous injectable AMP, DP7 was modified with hydrophobic cholesterol, and DP7-C easily self-assembled into micelles in aqueous solution with a nanoscale size and a zeta potential distribution. The micelles had a mean particle size of 46 nm and a zeta potential of 43.8 mV. This novel strategy greatly reduced the hemolytic effects of DP7 both in vivo and in vitro. For DP7-C micelles, the bacteriostatic effect in vivo was more focused on the bacteriostatic effect of DP7-C itself and its immune regulation effect. Compared to the DP7-modified liposome, DP7-C micelles demonstrated several other advantages, including a simple, enhanced antibacterial effect and safety.

In the zebrafish Pseudomonas aeruginosa infection model and mouse MRSA abdominal infection model, high doses of DP7-C micelles showed therapeutic effects comparable to those of vancomycin. In addition, no obvious changes in behavior, feeding, blood routine, blood biochemistry, or morphology of the primary organs were observed after systemic administration, suggesting a high degree of safety. The coagulation caused by DP7 is mainly due to its strong positive charge, which destroys the red blood cells after intravenous administration and leads to the death of the mice because of hemolysis. While the cholesterol-modified DP7 (DP7-C) significantly decreased the rate of hemolysis, the mice survived after the intravenous injection of 80 mg/kg. However, the zeta potential showed that DP7-C still had a positive charge, so we assume that when the larger dose was used, DP7-C could still cause damage to red blood cells, resulting in coagulation, and this effect might be dose dependent, as for DP7. Therefore, from the perspective of application, DP7-C micelles have the advantages of simple preparation, a significant antibacterial effect, and ideal in vivo safety. Thus, all of these qualities indicate that DP7-C could serve as a potential candidate for the clinical treatment of bacterial infections. Perhaps in future research we will test DP7-C directly with antibiotics in order to achieve simpler drug preparation methods and better therapeutic effects.

It is interesting that although a low level of antimicrobial activity was observed in vitro, DP7-C micelles demonstrated potent therapeutic effects in models of systemic infection in zebrafish and mice. This may be because DP7-C micelles played a role in immune regulation in vivo or because DP7-C might be degraded into DP7 by a specific enzyme in the process of absorption in vivo. To further clarify the mechanism, the immune-regulatory effects of DP7-C micelles were tested. The results showed that DP7-C micelles were effective in recruiting monocytes. Recent studies have suggested that monocytes (Gr1⁺ F4/80⁺) are capable of protecting mice against lethal challenge (40, 41). As one of the primary phagocytes of the innate defense system, monocytes play a crucial role in the clearance of invading pathogens. In addition, based on the results, cytokines were increased after DP7-C micelle treatment and were further increased after LPS stimulation. With the combination of LPS and DP7-C micelles, the excessive increase in cytokines was reduced. This demonstrates that DP7-C has the ability to balance the immune response and not to only activate or inhibit the immune response. Collectively, DP7-C had the ability not only to stimulate innate immunity (such as cytokines) and induce the expression of IL-6 and M-CSF but also to reduce the harmful inflammation induced by bacteria. We assume that a possible mechanism is that under normal circumstances, bacterial signature molecules can activate the innate immune response, and the activated innate immune response leads to the activation of some signaling pathways. The activated pathways can result in a rapid proinflammatory response, such as the upregulation of the proinflammatory cytokine TNF-α and some chemokines, and a more moderate anti-inflammatory response that eventually dampens the proinflammatory response. Therefore, immune cells are enriched at the site of infection, relieving the infection, but this is accompanied by a potentially harmful inflammatory response. In the presence of DP7-C, some activated signaling pathways may be altered and others may be maintained. Changes in these signaling pathways may result in the reduced production of proinflammatory cytokines or, in some cases, enhanced chemokine production and anti-inflammatory responses. Therefore, cell recruitment and effector mechanisms, especially the balance of monocytes and neutrophils, are changed, thereby effectively controlling infections without increasing potentially harmful inflammation.

In conclusion, in this study, we have synthesized a conjugate of cholesterol and DP7 (DP7-C) which could spontaneously form micelles in water. Compared with the results for DP7, the survival rate of mice after intravenous injection of DP7-C was significantly increased, and liver bleeding and pulmonary hemorrhage were not observed. Although the in vitro antibacterial and hemolytic assays demonstrated that both the hemolytic and antibiotic capacities of DP7-C were significantly decreased compared to those of DP7, the DP7-C micelles exerted potent therapeutic benefits in in vivo infection models and considerable safety via intravenous injection at high doses. The molecular mechanism study further suggested that DP7-C could regulate immune responses in conjunction with their direct antibacterial activities. Our study demonstrates that newly developed DP7-C possesses good antibacterial efficacy and does not exhibit obvious side effects following systemic administration, findings which suggest that DP7-C could be an excellent candidate for the treatment of bacterial infections in the clinic.

MATERIALS AND METHODS

Materials and animals.

Two strains of Pseudomonas aeruginosa (ATCC 10145 and ATCC 10145-GFP), two strains of S. aureus (ATCC 25923 and ATCC 33591), and an Escherichia coli strain (ATCC 25922) were purchased from the American Type Culture Collection (Rockville, MD). In all experiments, ATCC 25923 and ATCC 33591 were cultured using Mueller-Hinton broth and Mueller-Hinton agar media, and ATCC 10145, ATCC 10145-GFP, and ATCC 25922 were cultured using LB medium. All media and agar were purchased from Qingdao Hope Bio-Technology Co., Ltd. (Shandong, China).

Female BALB/c mice weighing 22 ± 2 g and female C57BL/6J mice weighing 17 ± 1 g were purchased from Vital River Laboratory Animal Technology Co. Ltd. (Beijing, China). Mice were housed at a controlled temperature of 22 to 24°C with 12-h light and 12-h dark cycles. All the mice were kept in quarantine for 1 week before treatment. This experimental program was approved by the West China Hospital Review Committee, and humane care of the animals was conducted in accordance with the Code of Medical Ethics of the World Medical Association.

Preparation and characterization of DP7-C micelles.

DP7-C and penetratin-C were synthesized by GL Biochem (Shanghai, Ltd.). DP7-C micelles were prepared by a self-assembly method. Briefly, 5 mg of DP7-C was directly added into 1 ml aqueous solution at 37°C. Because of its amphiphilic properties, DP7-C self-assembled into micelles without using any additives. The prepared DP7-C micelles were lyophilized or stored at 4°C.

The particle size distribution and zeta potential of the DP7-C micelles were measured using a Malvern ZetaSizer Nano-ZS Zen3600 system (Malvern Instruments, UK). Measurements were performed at 25°C after equilibration for 2 min. All results are the means from three experiments, and all data are expressed as the mean ± standard deviation (SD). The morphological characteristics of DP7-C micelles were examined by using SEM (JSM-7500F scanning electron microscope; FEI) and atomic force microscopy (AFM; NSK Ltd., Tokyo, Japan). A series of 2-fold dilutions of DP7-C micelles was diluted with sterile water and placed on a copper grid covered with nitrocellulose.

Determination of CMC.

Because Kalyanasundaram and Thomas proved that the characteristic dependence of the fluorescence vibrational fine structure of pyrene could be used to determine the critical micelle concentration (CMC) of surfactant solutions, the pyrene I1/I3 ratio method has become one of the most popular procedures for the determination of this important parameter in micelle systems (42). Fluorescence spectra were recorded on a PerkinElmer LS55 fluorescence spectrophotometer, and pyrene was used as a fluorescence probe. Samples for fluorescence measurement were prepared, and the concentrations of DP7-C were as follows: 0.0001, 0.001, 0.01, 0.025, 0.05, 0.1, 0.125, 0.25, 0.50, and 1.0 mg/ml. The pyrene concentration in the aqueous solutions was 6.0 × 10⁻⁷ mol/liter. To measure the pyrene excitation and emission spectra, the slit widths for the excitation spectra were maintained at 8 nm, and the slit widths for the emission sides were 2.5 nm. The emission spectra were recorded between 373 nm and 384 nm with an excitation wavelength at 334 nm. The intensity ratios of the first (374 nm) to the third (384 nm) vibronic peaks (I1/I3) were plotted as a function of the DP7-C concentration. The CMC value of DP7-C was extrapolated from the intersection of the tangent to the curve at the inflection with the horizontal tangent through the points at low concentrations.

In vitro hemolysis assay.

Human erythrocytes were collected from venous blood samples from healthy volunteers, from which erythrocytes were harvested by centrifugation for 10 min at 400 × g, and the erythrocytes were washed three times in PBS (pH 7.0). A 20% (vol/vol) suspension of human erythrocytes was prepared in PBS. The suspension was diluted 1:20 in PBS, and 100 μl was added to a 96-well plate containing 200, 400, 600, 800, 1,000, 1,200, and 1,600 μg/ml DP7 or DP7-C micelles. The plates were incubated at 37°C for 1 h and then centrifuged for 10 min at 900 × g. Next, we transferred 160 μl of the supernatant to another 96-well plate, and the absorbance at 405 nm was measured with a microplate reader (Multiskan MK3; Thermo). The percent hemolysis was calculated according to the absorbance at 405 nm. This experimental program was approved by the West China Hospital Review Committee.

Observation of survival rate of intravenous DP7 and DP7-C.

BALB/c female mice weighing 22 to 24 g were used in the study. The mice were intravenously injected with DP7 (1 mg/kg, 2 mg/kg, 4 mg/kg, 8 mg/kg, 10 mg/kg, 15 mg/kg, 20 mg/kg) or DP7-C (20 mg/kg, 40 mg/kg, 60 mg/kg, 80 mg/kg) (n = 10) and observed for 1 week. The survival rate was recorded.

In vivo hemolysis assay.

Six- to 8-week-old BALB/c female mice were i.v. injected with DP7 or DP7-C at a dose of 10 mg/kg. At 4 h postinjection, whole blood and serum were collected for routine blood and blood biochemical analysis. The mice were then sacrificed, and the main organs were removed for histopathology analyses.

In vivo antimicrobial activity assay.

Wild-type AB zebrafish were obtained from the State Key Laboratory of Biotherapy at Sichuan University and were bred and maintained normally (temperature, 28°C; light cycle, 14 h on and 10 h off; pH 7.2 to 7.4). Zebrafish eggs were obtained and bleached according to protocols described in the zebrafish manual (46). The eggs were then kept in dishes containing water supplemented with 0.3 mg/ml of methylene blue, and from 1 h postfertilization onwards, 0.003% 1-phenyl-2-thiourea was added to inhibit melanin synthesis in the zebrafish embryos. Embryos were reared at 28°C according to the desired speed of development. The zebrafish larvae were inoculated by i.p. injection of a mixture of ATCC 10145-GFP (PAO1-GFP) and either 1.0 mg · ml⁻¹ DP7-C or normal saline. There was a positive correlation between the observed green fluorescence and the number of PAO1-GFP bacteria. Zebrafish larvae were then raised in a 6-well flat-bottom tissue culture plate. Concurrently, we observed that the normal saline-treated zebrafish larvae had a continuous increase in the level of green fluorescence, while that in the DP7-C-treated zebrafish larvae decreased. Zebrafish larvae were imaged under a fluorescence microscope (Olympus).

The in vivo antibacterial activity of DP7-C micelles was also studied in a mouse abdominal infection model using vancomycin as a positive control and penetratin-C as a negative control. The mice were injected with 0.5 ml of an inoculum of Staphylococcus aureus (ATCC 33591; 3 × 10⁸ CFU/ml) via intraperitoneal injection. After 1 h, different doses of DP7-C micelles (0.2 mg/kg, 1 mg/kg, 2 mg/kg), penetratin-C (2 mg/kg), or vancomycin (5 mg/kg) were administered intravenously. The bacterial burden was determined at 24 h.

Experimental murine model of bloodstream infection.

A well-characterized murine model of bloodstream infection was employed to study the pharmacodynamics profiles of DP7-C (43). Briefly, C57BL/6J female mice (body weight, 17 g ± 1 g) were infected intravenously with 1 × 10⁸ CFU of logarithmic-growth-phase MRSA (recovered on defibrinated sheep blood agar [Oxoid, UK]). At 1 h prior to infection, the mice were treated with DP7-C at 1 or 2 mg/kg, and the mice in the control group were injected with the same volume of saline. Vancomycin at 10 mg/kg served as the positive control. Survival and weight changes were observed every 24 h for 10 days.

Preparation of primary mouse PBMCs and quantitative real-time PCR (qPCR).

Whole blood from 6- to 8-week-old BALB/c female mice was collected into an anticoagulation tube with EDTA. The cells were separated by centrifugation and collected after two washes with RPMI 1640. The cell pellet was resuspended at 5 × 10⁶ cells/ml in RPMI 1640 with 10% fetal bovine serum and 1% penicillin-streptomycin. The cell suspension was added to a 6-well flat-bottom tissue culture plate and stimulated with DP7-C micelles (0.20 mg/ml) in the presence or absence of LPS (10 ng/ml; Sigma) for 4 h at 37°C. When the combination of DP7-C and LPS was used, cells were pretreated for 45 min with DP7-C micelles. Total RNA was extracted, and gene expression profiles were analyzed by real-time PCR. The expression of each gene was normalized to that of β-actin, and fold changes in expression (y axis) are presented as the level of gene expression relative to that in unstimulated cells, determined using the comparative threshold cycle method (44, 45).

Flow cytometry.

For flow cytometry analysis, anti-mouse F4/80-allophycocyanin, anti-mouse Gr1-phycoerythrin (PE), and anti-mouse CD11b-PE were purchased from BD Bioscience. BALB/c female mice (6 to 8 weeks old) were given 0.2 ml sterile saline, 1 mg/ml DP7-C micelles, or penetratin-C via an intraperitoneal injection. After 24 h, leukocytes were isolated from the peritoneal irrigation fluid and were pretreated on ice for 20 min with a monoclonal antibody to block nonspecific binding to Fcγ receptors. Thereafter, the fluorescence-conjugated antibodies were added to cells and the cells were incubated for 30 min on ice. Then, the cells were washed and fixed with 0.2% paraformaldehyde for 10 min. Using flow cytometry, the numbers of macrophages (F4/80⁺ CD11b⁺), neutrophils (Gr1⁺ F4/80⁻), and monocytes (Gr1⁺ F4/80⁺) were calculated. Analysis of stained cells was performed with a FACSCanto flow cytometer (BD Biosciences). Isotype controls were used to set appropriate gates. Data were analyzed using FACSDiva (BD Bioscience) and FlowJo (version 6.4.7; TreeStar Inc., Ashland, OR) software. For all samples, approximately 200,000 cells were analyzed.