Antimicrobial peptides have broad-spectrum killing activities against bacteria, enveloped viruses, fungi and several parasites via cell membrane permeation and exhibit primarily immunomodulatory and anti-infective functions in their interactions with host cells. However, the mechanism underlying their anti-inflammatory activity remains to be elucidated. L-K6, an analog of temporin-1CEb isolated from the skin secretion of Rana chensinensis, has demonstrated a wide range of antimicrobial activities against gram-negative and gram-positive bacteria. In this study, the potent anti-inflammatory mechanism of L-K6 and its analogs in lipopolysaccharide (LPS)-stimulated human macrophage U937 cells were evaluated. We found that L-K6 suppressed the expression of inflammatory factors by two downstream signaling components in the MyD88-dependent pathway, including the mitogen-activated protein kinases (MAPKs) and the NF (nuclear factor)-κB signaling pathway, but its analog L-K5, which had the same amino acid sequence as L-K6 but no Lys residue at the –COOH terminal, only inhibited the phosphorylation of I-κB and NF-κB. Importantly, L-K6 and L-K5 were actively taken up by U937 cells through an independent cell membrane disruption mechanism and were eventually localized to the perinuclear region. The L-K6 uptake process was mediated by endocytosis, but L-K5 was specifically taken up by U937 cells via TLR4 endocytosis. Our results demonstrated that L-K6 can neutralize LPS and diassociate LPS micelles to inhibit LPS from triggering the proinflammatory signaling pathway, and by partially inhibiting inflammatory responses by the intracellular target. However, L-K5 may mainly inhibit proinflammatory responses by intracellular reporters to modulate the NF-κB signaling pathway.
Inflammatory responses induced by bacteria, during the course of bacterial infections, are beneficial to combating pathogenesis; however, uncontrolled responses can lead to shock, multiple organ dysfunction, and even death. In these infections, lipopolysaccharide (LPS), an important component of the outer membrane of Gram-negative bacteria, is the main activator of host defense mechanisms. Aggregated LPS initially binds with liposaccharide-binding protein (LBP) and the primary LPS receptor CD14; then, LPS-LBP–CD14 complexes are transduced by another membrane protein, Toll-like receptor 4 (TLR4), after which they trigger proinflammatory signaling pathways and induce cytokine secretions of tumor necrosis factor alpha (TNF-α) and interleukins (ILs), etc. [1–4]. The increasing incidence of conventional antibiotic resistance and the generation of multi-resistant bacterial strains have recently made the treatment of inflammation complicated, representing a serious challenge to human health and medicine [5,6]. Therefore, there is a need to develop new types of therapeutic agents to prevent the activation of inﬂammatory cells. Compared with conventional antibiotics, some antimicrobial peptides (AMPs) may kill bacteria, but they also simultaneously neutralize released LPS and are receiving increased attention as potential therapeutic candidates in infectious disease treatment.
AMPs are typically short amino acid chains with a positive net charge that exhibit broad-spectrum antimicrobial activities, killing both gram-negative and gram-positive bacteria, viruses and fungi [7–9]. They can rapidly kill invading pathogens through initial interactions with the negatively charged outer and/or inner membranes of bacteria, whereas their amphipathic character causes membrane permeabilization and disruption, which induces leakage of cytoplasmic components, and consequently, death of the microorganism [10,11]. Notably, the evolution of pathogen resistance to AMPs may be a very slow process resulting from a fundamental change in membrane composition. To date, widespread resistance has not been reported . Many AMPs not only play important roles in host defense against pathogen infections but also form valuable molecules in the innate immune system [13–15]. Some cationic host defense peptides, including human cathelicidin LL-37, bovine indolicidin and small synthetic cationic peptides, are known to reduce LPS-induced inflammatory responses by neutralizing and binding LPS [16–18]. Based on these properties, AMPs may be the most promising agents to solve the multi-drug resistance problem and systemic treatment of infectious diseases.
In a previous study, we isolated, purified and characterized a natural 12-residue AMP temporin-1CEb from the skin secretions of Chinese brown frog, Rana chensinensis . Temporin-1CEb shows antimicrobial activity against gram-positive bacteria, but its therapeutic potential is limited by its hemolysis. Subsequently, eight temporin-1CEb analogs were designed and synthesized to maintain their amino acid sequence but decrease its hemolytic activity and increase its cationicity and amphipathicity by substituting neutral and non-polar amino acid residues on the hydrophilic face of the α-helix with five or six lysine residues . Of all the peptides, L-K6, L-K6V1 and L-K6V2 have the same net positive charge (+7), but their hydrophobicity values (H) decrease from 11.6 to 10.1, and L-K5, with the same amino acid sequence as L-K6 but no Lys residue at the –COOH terminal, has higher H (12.3) than L-K6. L-K6 had the best potential as an antimicrobial agent because its antimicrobial activity against both gram-positive and gram-negative bacteria was substantial, i.e. it demonstrated an ability to completely kill Escherichia coli and Staphylococcus aureus at 4×MIC in 60 min, but its hemolytic activity was negligible (LD50 values >1000 µM). L-K6 has an α-helix in 50% trifluoroethanol/water and 30 mM SDS solutions and can quickly kill bacteria by rapidly inducing membrane depolarization, resulting in the disruption and disability of E. coli and S. aureus plasma membranes . Nevertheless, the anti-inflammatory activity and detailed cellular process of L-K6 and its analogs in human macrophage cells are still far from being fully understood. The aim of this study was to investigate the anti-inflammatory effects of L-K6 and demonstrate its possible mechanism of action in inflammation in human macrophages and mice. Its anti-inflammatory properties were established by investigating the suppression of TNF-α, IL-6, MIP-1 and MCP-1 cytokine release in LPS-stimulated U937 cells. The mechanism of action was examined with LPS-stimulated U937 cells.
Material and methods
Peptides were synthesized using standard Fmoc solid phase technology (GL Biochem Ltd, Shanghai, China), and purified to near homogeneity (95%) by reverse-phase high-performance liquid chromatography using a Vydac 218TP1022 C-18 column (2.2 cm × 25 cm; Separations Group, Hesperia, CA, U.S.A.) equilibrated with acetonitrile/water/trifluoroacetic acid. The relative mass of the peptide was determined using MALDI-TOF MS (Shimadzu, Japan). FITC-labeling peptides were synthesized by manually conjugated at the N-terminus of peptide with a green fluorescent probe FITC during synthesis. Amino acid sequences and physical characteristics for peptides are shown in Table 1.
Cell culture and cytotoxicity assay
The human macrophage cell line U937 was obtained from the Cell Bank of the Chinese Academy of Sciences Type Culture Collection Committee (Shanghai, China). U937 cells were cultured in RPMI-1640 supplemented with 10% fetal calf serum at 37°C in a humidified atmosphere of 5% CO2. Peptide cytotoxicity against mammalian cells was determined using 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assays. Cells were subcultured at a density of 105/well in 6-well dishes for 18 h and then washed with fresh medium. Approximately 5 × 105 cells/well were added to a 96-well plate and treated with different concentrations of peptides (0, 10, 20, 40, 60 and 80 µM). After incubation for 24 h to adhere the cells, 10 µl of 5 mg/ml MTT solution was added to each well; the plate was incubated for additional 4 h at 37°C, and 150 µl of DMSO was added to dissolve the purple-blue MTT formazan precipitant. The absorbance was read using a microplate reader at 490 nm. The percentage of growth inhibition was evaluated using an MTT assay to count the number of viable cells. Experiments were run in triplicate, and the results are expressed as a percentage of inhibition in viable cells.
Cytokine production in U937 cells
LPS from E. coli (Sigma–Aldrich, Inc., Shanghai, China) was used to induce an inflammatory reaction in U937 cells. Cells were cultured overnight at a density of 1 × 106 cells/well in 24-well plates to permit adherence and washed with fresh medium. The cells were treated with 20 µM peptides in the presence of 10 µg/ml LPS for 24 h. Supernatants of the U937 cultures were collected, and the production of TNF-α, IL-6, MCP-1 and MIP-1 was quantified using a QuantiCyto human ELISA kit (Neobioscience, Shanghai, China) according to the manufacturer's instructions. Cells treated with medium alone and with LPS alone were taken as negative controls for basal cytokine levels and positive controls for LPS-induced cytokine levels, respectively. The data were analyzed based on three independent experiments in each group.
In another experiment, U937 cells were first treated with 20 µM L-K6 and its analogs for 1 h and washed three times with phosphate-buffered saline (PBS) to remove the peptides in the medium, then incubated with LPS (10 µg/ml) for 24 h to evaluate the effect of intracellular L-K6 and its analogs on LPS-induced inflammatory response in U937 cells. Production of TNF-α, IL-6, MCP-1 and MIP-1 in the supernatants was measured using an ELISA kit as described above.
Quantitative real-time PCR
U937 cells were plated in 6-well plates (5 × 105 cells/well) and cultured overnight. The cells were treated 20 µM of peptides in the presence of 10 µg/ml LPS for 24 h, and then cells were collected from the wells and washed once with PBS. Total RNA was extracted with RNAiso reagent (TaKaRa, Dalian, China), and 2000 ng of total RNA was reverse-transcribed into cDNA using a Super Script™ III kit (Invitrogen, Shanghai, China). Real-time PCR was performed using the ABI Prism 7500 Fast sequence detection system (Applied Biosystems) and the following specific primers: TNF-α, 5′-TGCTTGTTCCTCAGCCTCTT -3′ (forward primer) and 5′-CAGAGGGCTGATTAGAGAGAGGT -3′ (reverse primer); IL-6, 5′-GCCAGAGCTGTGCAGATGAG-3′ (forward primer) and 5′-TCAGCAGGCTGGCATTTG-3′ (reverse primer); IL-8, 5′-TTCAGAGACAGCAGAGCACACA-3′ (forward primer) and 5′-TTCACACAGAGCTGCAGAAATC-3′ (reverse primer); MCP-1, 5′-AGCAGCAAGTGTCCCAAAGA-3′ (forward primer) and 5′-GGTGGTCCATGGAATCCTGA-3′ (reverse primer);MIP-1, 5′-CTGCTCAGAATCATGCAGGTC-3′ (forward primer) and 5′-ACTGGCTGCTCGTCTCAAAG-3′ (reverse primer); MyD88, 5′-GCGTTTCGATGCCTTCATCT-3′ (forward primer) and 5′-CATCAGAGACAACCACCACCAT-3′ (reverse primer); IκB-α, 5′-GACAGCACAGAGATGGTGAAATC-3′ (forward primer) and 5′-AGGGACCGGGCAGAACTC-3′ (reverse primer); NF-κB, 5′-CAGACGAGTGTGTGGTGAGCTTTC-3′ (forward primer) and 5′-CCCACTGTCCACCAGCAGAT-3′; GAPDH (internal control), 5′-CTTCTACAATGAGCTGCGT-3′ (forward primer) and 5′-TCATGAGGTAGTCGTCAG-3′ (reverse primer). The reaction conditions were as follows: 95°C for 30 s, followed by 40 cycles of 95°C for 5 s, 55°C for 30 s and 95°C for 30 s and a final extension step of 72°C for 30 s. The real-time PCR data were analyzed and the Ct values were calculated according to the instructions of the manufacturer. The cells treated with medium alone, and with LPS alone were taken as negative controls and positive controls, representing the stimulated and unstimulated gene expression levels, respectively. The data were analyzed from three independent experiments in each group.
Western blot analysis
U937 cells were seeded into 6-well plates (5 × 105 cells/well) and cultured overnight. The cells were treated with 20 µM peptides in the presence of LPS (10 µg/ml) and incubated at 37°C under 5% CO2 for 24 h. Cells were harvested and washed twice with ice-cold PBS. Cytoplasmic and nuclear fractions were prepared from the collected cells using an extraction kit (NE-PERTM; Pierce, IL, U.S.A.), according to the manufacturer's instructions. Protein concentrations were determined by Bradford assay. Proteins were separated by 10% SDS–polyacrylamide gel electrophoresis (SDS–PAGE) and electroblotted onto polyvinylidene fluoride (PVDF) membrane (Millipore, Beijing, China). After blocking with TBST (20 mM Tris pH 7.5, 150 mM NaCl, 0.1% Tween 20) containing 5% skim milk for 2 h at room temperature, the membranes were washed with 0.1% Tween 20 in PBS (PBS-T), then were incubated with primary antibodies against JNK, phosphorylated JNK, ERK, phosphorylated ERK, p38, phosphorylated p38, MyD88, IκBα, phosphorylated IκBα, NF-κB and phosphorylated NF-κB, followed by incubation with horseradish peroxidase-conjugated anti-rabbit IgG. Protein–antibody blots were detected using an Azure c500 imaging system (Azure Biosystems, U.S.A.). The cells treated with medium alone and with LPS alone were taken as negative controls of the unstimulated LPS-induced protein expression levels and positive controls of the stimulated LPS-induced protein expression levels, respectively. The data were analyzed based on three independent experiments in each group.
Dissociation of LPS aggregates
The ability of peptides to dissociate LPS micelles was assayed by employing FITC-labeled LPS (FITC-LPS), as reported previously . FITC-LPS (0.5 µg/ml) was treated with increasing concentrations of peptides (0–20 µM) at 37°C for 60 min. The basal fluorescence of FITC-LPS was recorded at an excitation of 480 nm and an emission of 515 nm using a Varioskan Flash Microplate Reader (Thermo Scientific Co., Beijing, China). The fluorescence data of both sodium phosphate buffer and peptides alone at 515 nm were taken for background subtractions. Polymyxin B was selected as a positive control. The dissociation of LPS micelles was recorded as an increase in fluorescence intensity with the addition of increasing concentrations of peptides.
Dynamic light scattering studies
The dissociation of LPS micelles by peptides was studied by DLS (dynamic light scattering) measurements performed on a Zetasizer Nano ZS (Malvern Instruments, U.K.) equipped with a 633-nm HeNe laser. LPS was solubilized in 10 mM sodium phosphate buffer (pH 7.0), extensively vortexed, sonicated at 60°C for 30 min, and subjected to 3–4 temperature cycles between 20 and 60°C. After the lipid suspension was incubated at 4°C for at least 12 h, LPS aggregates were prepared. For DLS measurements, the distribution of various sizes of LPS micelles in the absence and presence of 2 µM peptides was determined using 1 µM LPS. The change in this distribution was observed after the addition of peptides at different concentrations. Scattering data were analyzed using the CONTIN method provided with the instrument.
Direct binding of peptide with LPS as assessed by isothermal titration calorimetry
Peptide binding with LPS micelles was determined using ITC (isothermal titration calorimetry) on a GE MicroCal iTC 200 (MicrocalInc, MA, U.S.A.) [26,27]. E. coli 0111: B4 LPS and peptides were dissolved in 10 mM phosphate buffer (pH 6.0), vortexed for 10 min and sonicated for 5 min. Approximately 200 µl of LPS at a concentration of 50 µM was loaded into the sample cell, and the reference cell was filled with buffer. Approximately 2 µl of peptide (0.6 mM) was injected into the reaction cell in every 2 min. The sample cell was stirred continuously at 300 rpm. The data were collected and analyzed with a MicroCal Origin 5.0. The first data point (3-ml injection) was not considered for analysis. The best fit was obtained with one binding site equation for the whole process. The association constant (Ka) and enthalpy change (ΔH) were directly obtained from ITC profiles. The dissociation constant was calculated from Kd = 1/Ka. The change of free energy (ΔG) and change of entropy (ΔS) were calculated using fundamental equations of thermodynamics, ΔG = −RT lnKa and ΔS = (ΔH − ΔG)/T, respectively.
LPS neutralization assay (TAL assay)
The ability of peptide to neutralize LPS was determined using a commercially available chromogenic Tachypleus Amebocyte Lysate (TAL) kit (Xiamen Horseshoe Crab Inc., China). The experiments were strictly performed according to the protocols recommended by the manufacturer's instructions. The peptides were prepared in endotoxin-free water provided with the kit. Peptides at concentrations of 20.0, 10.0, 5.0 and 2.5 µM were incubated with 1.0 endotoxin unit (EU) in a total volume of 100 µl in a 96-well plate for 60 min at 37°C. Approximately 100 µl of TAL reagent was added to the peptide-EU mixture and further incubated for 15 min, followed by the addition of 100 µl of chromogenic substrate. After an incubation of 10 min, the reaction was terminated by the addition of 25% acetic acid, and the release of colored product was measured spectrophotometrically at 545 nm. The percentage of LPS binding was calculated as follows: [(ODLPS − ODpeptide)/ODLPS] × 100%. Polymyxin B was selected as a positive control. All assays were repeated twice, and average values are reported.
Measurement of binding of FITC-LPS to U937 cells using FACS
Approximately 105 U937 cells were incubated with 10 µg/ml FITC-LPS in the presence or absence of peptide at 37°C under 5% CO2 for 24 h. The cells were collected by centrifugation at 1000 rpm for 5 min at 4°C after being extensively washed twice with ice-cold PBS to remove the unbound LPS from the samples. FITC-LPS binding to the cells was monitored by measuring the mean fluorescence of 10 000 cells for each sample with a BD Biosciences FACS Caliber flow cytometer.
Circular dichroism spectroscopy in LPS micelles
The CD (circular dichroism) spectrum of the peptides was measured using a Jasco-810 CD spectrophotometer (Jasco, Victoria, BC, Canada) in a 1-mm path length quartz optical cell. The spectra were scanned at wavelengths of 190–250 nm at 0.5-nm intervals with a scan rate of 20 nm/min at 25°C. Approximately 0.25 mg/ml peptide was dissolved in 10 mM SP buffer at pH 7.4 in the presence or absence of 0.1% (0.22 mM) LPS. The CD spectra of SP buffer or LPS solutions without the peptide were used as baseline spectra. The percentage of α-helical structure was calculated by the DichroWeb software. The spectra of three consecutive scans were averaged.
L-K6 uptake analysis by confocal laser scanning microscopy
To facilitate the monitoring of the cellular uptake of the peptide, U937 cells (1 × 106 cells/well) were seeded in 12-well glass-bottom Nunc plates and left to adhere to the coverslips in complete DMEM overnight at 37°C under 5% CO2. Cells were stimulated with 20 µM of FITC-labeled peptide at 37°C for 30 min. Approximately 10 µg/ml of LPS was added at the same time as FITC-labeled peptide or 1 h prior to the addition of FITC-labeled peptide to study the effect of LPS on cellular peptide uptake. After gently washing the cells with PBS three times to remove free FITC-labeled peptide, the cells were fixed using 4% paraformaldehyde and were permeabilized using 0.1% Triton X-100 in PBS. Cell membranes were stained with Cell Mask™ red fluorescent tracker. Coverslips were viewed using a confocal laser scanning microscope (LSM 710, Carl Zeiss, Germany) with excitation/emission wavelengths of 480/500 nm. The data were evaluated from three independent experiments in each group.
Quantitative analysis of L-K6 uptake by flow cytometric analysis
To quantify the cellular uptake of the peptide, U937 cells (1 × 106 cells) were seeded in 12-well plates and cultured overnight at 37°C and 5% CO2. The following day, U937 cells were incubated with 20 µM FITC-labeled peptide at 37°C for 30 min, and the free FITC-labeled peptide was then removed by washing the cells three times with PBS. For experiments at low temperatures, cells were maintained at 4°C for 15 min before the addition of peptide and were kept at 4°C throughout the experiment. Approximately 10 µg/ml of LPS was added at the same time as FITC-labeled peptide or 1 h prior to the addition of FITC-labeled peptide to the study the effect of LPS on cellular peptide uptake. After the cells were harvested using 0.25% trypsin–ethylenediamine tetra acetic acid for 15 min, which also helped to remove excess FITC-labeled peptides bound to the outer membrane and reduce the occurrence of artifacts, the FITC-labeled peptides were analyzed using a FACS Calibur flow cytometer (Becton–Dickinson, CA, U.S.A.). L-K6 uptake was determined by fluorescence intensity. The data were analyzed based on three independent experiments in each group.
The binding of L-K5 or L-K6 with mCD14 was determined by using FITC-labeled anti-CD14 mAb. The U937 cells (1 × 105 cells) were pre-incubated in the presence or absence of the peptide (L-K5 or L-K6) for 1 h, and then FITC-labeled anti-CD14 mAb was added and incubated for 30 min in dark environment. The fluorescence intensity of FITC-labeled anti-CD14 mAb was determined by a FACS Calibur flow cytometer to estimate the binding of L-K5 or L-K6 with CD14 on cell surface.
For the experiment of endocytosis of TLR4, the cells (1 × 105 cells/well) were treated with 10 and 100 mg/ml TAK242, an inhibitor of TLR4 (List Biological Laboratories Inc., Campbell, CA, U.S.A.) for 2 h prior to the addition of 20 and 40 µM FITC-labeled peptide. The cells were incubated with FITC-labeled peptides for 30 min at 37°C under 5% CO2. The stained cells were washed with flow cytometry buffer and the FITC fluorescence intensity was analyzed using a FACS Calibur flow cytometer. The data were analyzed based on three independent experiments in each group.
Cell membrane depolarization assay
Cell membrane depolarization was measured using the anionic fluorescent dye bis-[1,3-dibutylbarbituric acid] trimethine oxonol (DiBAC4), as described previously . Then, 2 µM DiBAC4 was subjected to U937 cells for time scanning by a fluorescence spectrophotometer (Varioskan Flash, Thermo Scientific, Dalian, China) with laser excitation at 488 nm and emission at 518 nm. When the fluorescence intensity was stable, the cells were treated with 20 µM peptide or sterile deionized water (negative control) or TritonX-100 (positive control). Membrane depolarization was monitored by consecutively observing changes in the intensity of fluorescence emission of the membrane potential DiBAC4 dye. The assay was repeated three times.
Experimental wound infection and treatment
Female Kunming mice (Dalian Medical University, China) weighing between 19 and 21 g were used in the experiments. The animal experiments were carried out in accordance with the approved guidelines of the ‘Principles of Laboratory Animal Care’ (NIH publication #85-23, revised 1985) and were approved by the Animal Welfare and Research Ethics Committee of Dalian Medical University (Approval Number: SYXK2004–0029). The procedures used to produce the wound infections have been described previously . Briefly, mice were anesthetized with an intraperitoneal injection of sodium pentobarbital (30 mg/kg), and then a full-thickness infectious wound was created by excising the selected dorsal skin and infecting it with the bacterial solution of S. epidermidis, S. aureus and P. aeruginosa on the wound. After the day of infection, the wounds were treated with 1 or 2×MIC L-K6 twice daily, and the entire treatment lasted for 6 days. Physiological saline was used as a negative control, mupirocin ointment was used as a positive control. Three or 6 days after treatment with L-K6, six mice from each group were sacrificed, and the infected sections of skin were obtained aseptically, weighed and homogenized in distilled water. The homogenate was then serially diluted, bacterial CFU counts were obtained and statistical analyses were performed.
Measurement of hydroxyproline content
The procedures of hydroxyproline content measurements have been described previously . Full-thickness skins from the wound were rinsed in ice-cold PBS (10 mM, pH 7.4) and then homogenized in 1 ml of PBS on ice. The homogenates were centrifuged for 5 min at 5000g after subjected to two freeze–thaw cycles. Hydroxyproline levels in the supernatant were measured according to the protocol supplied with the CSB-E08839 m ELISA Assay kit for Hydroxyproline (CUSABIO, Shanghai, China).
Results are generally expressed as means standard errors from three independent experiments. The paired Student t-test was used to test for signiﬁcance.
Cytotoxicity of peptides to U937 cells
In our previous work, L-K6 and its analogs exhibited high, broad-spectrum antimicrobial and anticancer activities [20,28]. Thus, prior to investigating the anti-inflammatory potential of the peptides, the cytotoxicity of peptides on U937 cells was evaluated by examining cell viability using the mitochondrial reduction in MTT. As shown in Supplementary Figure S1 (Supplementary Results), L-K5, L-K6V1 and L-K6V2 had low cytotoxicity to human macrophage U937 cell lines; even at a peptide concentration of 80 µM (cell viability >80%), the concentration could completely kill bacteria and tumor cells [20,28]. Conversely, L-K6 showed cytotoxicity to U937 cells; when it was administered at a concentration of 20 µM, cell viability was greater than 80%. As the concentration of L-K6 increased, cell viability decreased. Therefore, subsequent experiments were conducted at a concentration of 20 µM.
Peptide inhibition of the LPS-induced inflammatory response
LPS-treated U937 cells were used as a model of inflammatory processes to test the effects of peptides on the inflammatory response. As shown in Figure 1, LPS treatment dramatically induced cellular production of TNF-α, IL-6, MCP-1 and MIP-1. Compared with the LPS treatment, L-K6 and its analogs significantly inhibited the production of TNF-α by 54.3–72.6%, IL-6 by 58.8–81.0%, MCP-1 by 45.4–75.9% and MIP-1 by 48.2–80.5%. L-K6 exhibited the greatest effects on the expression of LPS-induced inflammatory cytokines, including TNF-α, IL-6, MCP-1 and MIP-1, with inhibition of 72.6, 81.0, 75.9 and 80.5%, respectively.
Production of TNF-α (A), IL-6 (B), MCP-1 (C) and MIP-1 (D) in LPS-stimulated U937 cells.
We next evaluated the mRNA expression of these cytokines in macrophages using RT-qPCR. Consistent with expectations, the expression of TNF-α, IL-6, IL-8, MCP-1 and MIP-1 increased upon treatment with 10 µg/ml LPS, but L-K6 and its analogs inhibited 45.0–80% of the mRNA levels (Figure 2). The expression of TNF-α, IL-6, IL-8, MCP-1 and MIP-1 mRNA was suppressed by 39–73.5%, 52.1–69.2%, 63.6–82.8%, 72.8–95.1% and 47.7–70.9%, respectively, compared with levels in the LPS treatment. L-K6 and L-K6V2 had the greatest effects on the expression inhibition of LPS-induced inflammatory cytokines, including TNF-α, IL-6, MCP-1, IL-8 and MIP-1.
Evaluation of gene expressions of TNF-α (A), IL-6 (B), IL-8 (C), MCP-1 (D) and MIP-1 (E) in LPS-stimulated U937 cells.
L-K6 suppressed MyD88-dependent mitogen-activated protein kinase pathways and subsequent NF-κB translocation in U937 cells
MyD88 is the first intracellular signal protein in the LPS-induced inflammatory response. We first investigated the effects of L-K6 and L-K5 on the activation of MyD88 by Western blotting. As shown in Figure 3A, the expression of MyD88 was inhibited 85.6 and 66.0% in L-K6- and L-K5-treated U937 cells compared with the protein levels in nontreated cells, respectively. The results suggested that L-K6 and L-K5 inhibited LPS-induced inflammatory action by a MyD88-dependent signaling pathway. Then, the effects of the two peptides on MAPK (mitogen-activated protein kinase) activation were detected by Western blotting. As shown in Figure 3B, three types of MAPKs are expressed in U937 cells, including ERK, JNK and p38. After U937 cells were treated with 10 µg/ml LPS, the phosphorylation of ERK, JNK and p38 significantly increased; however, L-K6 treatment significantly reduced phosphorylated p38, ERK and JNK levels compared with the protein levels in LPS-treated cells, with suppressed rates of 37.0, 76.0 and 86.7%, respectively. However, L-K5 failed to suppress the expression of phosphorylated ERK and JNK in cells treated with both LPS and peptide (Figure 3B).
Effect of L-K6 and L-K5 on the MyD88-dependent signaling pathway in LPS-stimulated U937 cells.
Furthermore, we examined the phosphorylation of IκB and NF-κB, the key signal proteins in the inflammatory pathway. The Western blot results showed that the phosphorylation of IκB was suppressed by 58.0 and 33.9%, respectively, in L-K6- and L-K5-treated cells and that the phosphorylation of NF-κB was suppressed by 50.1 and 55.3%, respectively, in L-K6- and L-K5-treated cells compared with the protein levels in LPS-treated cells alone (Figure 3C). Therefore, L-K6 and L-K5 also inhibit the production of TNF-α, IL-6, MCP-1 and MIP-1 by MyD88-dependent NF-κB activation. Thus, the data suggest that L-K6 prevented LPS from inducing MAPK pathways and subsequent NF-κB translocation; however, L-K5 inhibited LPS-induced NF-κB activation by inhibiting the phosphorylation of IκB in U937 cells.
L-K6 and L-K5 exhibit the interactions with LPS
L-K6 dissociates FITC-labeled LPS aggregation
LPS-stimulated proinflammatory responses depend on the physical state of LPS, and some cationic AMPs have been found to be able to dissociate the aggregated state of LPS . We assayed peptide-induced disaggregation of LPS by employing FITC-labeled LPS or DLS assay to study the effects of L-K6 and L-K5 on the physical state of LPS. As shown in Figure 4A, we found that L-K6 was able to dissociate FITC-LPS aggregates in a fluorometric study in which dequenching of the aggregated fluorescence of LPS was recorded. However, L-K5 as examined here was less active than L-K6 at inducing disaggregation in LPS aggregates, suggesting that the loss of a Lys residue at the –COOH terminal significantly impaired L-K6-induced disaggregation of LPS oligomers. The DLS results further demonstrated the above results. In the absence of peptides, LPS micelles has two distributions with 100 nm- and 7500 nm-sized aggregates. As expected, the addition of L-K6 delineated dissociation of larger size aggregates of LPS into smaller size aggregates ranging from 100 to 1000 nm, meanwhile the smaller ones are absent, indicating some reorganization of this LPS micelles (Figure 4B). L-K5 as examined here was less active than L-K6.
Dissociation of LPS micelles by L-K6 and L-K5.
L-K6 possesses greater LPS neutralization ability than L-K5
In our previous study, we confirmed that L-K6 is able to bind with LPS . We quantitatively analyze the interaction between L-K6 and L-K5 and LPS by determining binding affinities in ITC experiments. ITC is a method to determine the enthalpy change caused by peptide–LPS binding. Figure 5A shows a representative thermogram (top) and integrated heats of binding between the peptide and LPS (bottom) at 25°C. The result showed that the interaction between the two peptides and LPS was primarily an exothermic reaction as deduced from the negative enthalpy changes after each titration step at 25°C; furthermore, it can be deduced that electrostatic binding is involved in LPS–peptide complex formation. Binding saturation was observed when the peptide:LPS molar ratio was over 1.2 for L-K6 after 25 min and over 2.5 for L-K5 after 40 min. The result indicated that L-K5 had a relatively low-binding ability to LPS than L-K6. The thermodynamic parameters of the interaction between peptides and LPS were estimated using ITC measurements (Table 2). L-K5 interacted with LPS with a dissociation constant Kd of 8.84 µM, which was much higher than the Kd of L-K6 (0.35 µM), indicating a comparatively much weaker binding of L-K5 to LPS. The stoichiometry of binding, N value, is 1.39 for L-K5 and 0.97 for L-K6, both values are close to 1, which indicated that 1 molar peptide molecules bind to 1 molar LPS. The positive entropy ΔS (8.44 kcal mol−1 deg−1) and negative enthalpy ΔH (−6.576 kcal mol−1) of the interaction between L-K6 and LPS were large, as evident from the exothermic binding process.
Binding of L-K6 and L-K5 to LPS.
|N||1.39 ± 0.08||0.97 ± 0.02|
|Ka (µM−1)||0.11 ± 0.041||2.90 ± 1.02|
|ΔH (kcal mol−1)||−5.52 ± 0.54||−6.58 ± 0.20|
|ΔS (calmol−1 deg−1)||5.36||8.44|
|N||1.39 ± 0.08||0.97 ± 0.02|
|Ka (µM−1)||0.11 ± 0.041||2.90 ± 1.02|
|ΔH (kcal mol−1)||−5.52 ± 0.54||−6.58 ± 0.20|
|ΔS (calmol−1 deg−1)||5.36||8.44|
The neutralization ability of L-K6 and L-K5 to LPS was further examined by measuring the peptides’ inhibition of LPS-induced activation of the TAL enzyme using an endotoxin detection kit. As seen in Figure 5B, L-K6 can significantly inhibit LPS-mediated activation of the TAL enzyme and neutralize more than 30% of endotoxin at a concentration of 20 µM, in which cell viability was greater than 80%, but L-K5 showed significantly lower binding ability to LPS than L-K6, with only 18% inhibition of the TAL enzyme at a concentration of 20 µM.
The significant efficacy of L-K6 and L-K5 at inhibiting the binding of FITC-LPS to the surface of U937 cells was further evidenced by flow cytometry. U937 cells were incubated with FITC-LPS in the absence or presence of L-K6 and L-K5 for 60 min, and FITC fluorescence was detected by flow cytometer. As shown in Figure 5C, the addition of L-K6 or L-K5 significantly decreased FITC fluorescence, suggesting a crucial role of L-K6 and L-K5 in inhibiting FITC-LPS binding on U937 cell surfaces. Approximately 20 µML-K6 was able to inhibit 80.2% of LPS binding, and 40 µML-K6 was able to inhibit 82.1% of LPS binding. Compared with L-K6, L-K5 exhibited a low inhibition rate (48.4 versus 52.2% for L-K5 and L-K6, respectively), further indicating that the binding ability of L-K5 to LPS was weaker than that of L-K6.
As the binding of the peptide to LPS is concomitant to folding, we analyzed the CD spectra of the peptide dissolved under LPS conditions. As shown in Figure 5D, L-K6 and its analogs showed an unordered conformation in SP buffer solution, but they exhibited conformational changes in LPS micelles with a positive maximum at 192 nm and a negative maximum at 222 nm, suggesting that they adopted a significant degree of α-helical and β-sheet structures. Based on the molar ellipticity, the calculated helical contents in LPS solution for L-K6 and L-K5 were 65 and 72%, respectively. In addition to α-helical conformation, L-K6 had ∼1% β-sheet and 6% turn structures, and L-K5 had ∼2% β-sheet and 7% turn structures.
Uptake of L-K6 and L-K5 by U937 cells
It has been demonstrated that L-K6 is able to enter into breast cancer cells (MCF-7 cells) by endocytosis . Here, L-K6 and L-K5 inhibited the proinflammatory response by down-regulating the NF-κB and MAPK signaling pathways; thus, we wanted to investigate whether L-K6 and L-K5 were internalized in human macrophage cell line U937 using confocal microscopy. Thus, we used FITC-labeled peptides to monitor cellular uptake in U937 cells after 30 min of stimulation. The results showed that L-K6 and L-K5 were taken up into the cells and were already detected throughout the cytoplasmic region at 30 min after stimulation (Figure 6Aa, b and c). Interestingly, we found that L-K6 uptake into U937 cells significantly decreased under the condition of 10 µg/ml LPS-pretreatment for 60 min (Figure 6Ah, i and j) compared with that under the control treatment of LPS and peptide simultaneously (Figure 6Ad, e and f); however, L-K5 uptake into U937 cells slightly decreased under the conditions of LPS-pretreatment, suggesting that LPS-pretreatment inhibited L-K6, not L-K5, uptake into U937 cells. We further quantitatively evaluated the effect of LPS-pretreatment on L-K6 uptake by flow cytometry. As shown with histogram analysis (Figure 6B), the uptake of FITC-labeled L-K6 decreased 27% with LPS, whereas the uptake of FITC-labeled L-K5 slightly decreased with LPS.
Effect of LPS on the uptake of L-K6 and L-K5 by U937 cells.
Peptide uptake into U937 cells is an active process
L-K6 is a membrane-activity peptide [20,21,30]. Membrane depolarization was detected using DiBAC4(3), an anionic and membrane-potential-sensitive dye, to determine whether L-K6 and L-K5 were internalized by damaging membrane permeability. The depolarization of cell membranes leads to an uptake of DiBAC4(3) inside the cells, resulting in an increased fluorescent signal. As seen in Figure 7A, after a 30-min exposure of 20 µM L-K6 or L-K5, the U937 cells had no significant increase in the fluorescence intensity of DiBAC4(3); however, a significant increase in the fluorescence intensity of DiBAC4(3) was observed in the positive control (Triton-100-treated cells) after 1 min. This result suggested that L-K6 and L-K5 cell uptake was not dependent on the disruption of cell membranes.
Uptake of peptide into U937 cells is an active process.
U937 cells were incubated with L-K6 or L-K5 at 4°C to determine whether L-K6 was taken up into cells through an active or a passive mechanism. As shown in Figure 7B, when the U937 cells were treated with peptides at 4°C for 1 h, L-K6 and L-K5 uptake into cells was inhibited by 63.6 and 81.3% compared with their uptake rates at 37°C, respectively, revealing that low temperature diminished peptide uptake. This result suggested that the uptake of L-K6 and especially L-K5 into U937 cells was temperature dependent, as would be expected if uptake were dependent on receptors.
Next, we investigated the role of membrane receptor CD14 and TLR4 on L-K6 uptake by U937 cells. CD14 is a necessary membrane receptor that binds LPS to TLR4. U937 cells were treated with L-K6 and L-K5 for 60 min prior to FITC-labeled anti-CD14 antibody for 30 min to detect the binding ability of L-K6 and L-K5 with CD14. The fluorescence intensity of FITC was detected by flow cytometry. As shown in Figure 7C, L-K6 almost completely inhibited the binding of FITC-labeled anti-CD14 antibody onto CD14 membrane protein on the surface of U937 cell membranes. Less than 10% of FITC-labeled anti-CD14 antibody was binding onto CD14 membrane protein. However, L-K5 had little effect on the binding of FITC-labeled anti-CD14 antibody on CD14 protein on the surface of U937 cell membranes, as the binding of CD14 antibody with cells was more than 90% compared with the control. The result indicated that L-K6, not L-K5, may bind to CD14 protein, consequently restraining the binding of anti-CD14 antibody onto CD14 protein.
TLR4 may be internalized into the endosome network, where the second signaling pathway is triggered through the adaptors TRAM and TRIF . L-K6 can be incorporated into TLR4 by either CD14 or LPS. Based on these aspects, we hypothesized that L-K6 itself is incorporated into U937 cells by TLR4 endocytosis. Thus, TAK242, an inhibitor of TLR4, was used to elucidate whether L-K6 can enter the cell by TLR4 endocytosis. After pretreatment without or with 10 µg/ml and 100 µg/ml of TAK242 for 40 min, cells were incubated with FITC-labeled peptides for 1 h, and the intracellular peptides were quantitated by flow cytometry. As shown in Figure 7D, the addition of TAK242 significantly inhibited L-K5 uptake into U937 cells, and uptake decreased as TAK242 concentration increased. Only 12.7 and 3.7% of L-K5 was incorporated into U937 cells in the presence of 10 and 100 µg/ml TAK242, respectively. Compared with L-K5, L-K6 uptake into cells was less affected by TAK242, and the incorporation of L-K6 into U937 cells was the same in the treatments with 10 and 100 µg/ml TAK242. These observations suggest that TLR4 endocytosis is important for the uptake of L-K5 by U937 cells.
Anti-inflammatory activity of L-K5 and L-K6 by the intracellular target
L-K6 and L-K5 can interact with LPS, resulting in the inhibition of the proinflammatory response; however, our results also demonstrated that the two peptides can be incorporated into U937 cells. Human macrophage U937 cells were pretreated with L-K6 and L-K5 for 60 min to force the peptides to be incorporated into cells before LPS (10 µg/ml) stimulation for 24 h to determine whether the main target of L-K6 and L-K5 was LPS or intracellular proteins. TNF-α and IL-6 production was monitored. As shown in Figure 8, LPS-induced TNF-α production was suppressed 72.6% with a simultaneous treatment with L-K6 and LPS and was suppressed 60.5% in L-K6 pretreatment for 60 min. LPS-induced IL-6 production was suppressed 81.0 and 72.6% under L-K6 and LPS treatment and L-K6 pretreatment, respectively. Compared with the control, pretreatment with L-K6 decreased the inhibition of LPS-induced TNF-α and IL-6 production, suggesting that L-K6 can inhibit the inflammatory response with an intracellular target; however, 80% of the inhibition resulted from interaction with LPS.
Inhibition of the intracellular peptides on the production of cytokines in LPS-treated U937 cells.
Wound closure is enhanced by L-K6
The wound area was smaller in the groups treated with 1 or 2×MIC L-K6 (Supplementary Figure S2A) than in the control group 3 and 6 days after wounding. No sign of inflammation could be observed. The wound closure was ∼90% on day 6 in the groups treated with 1 or 2×MIC L-K6. The number of bacterial CFU in wound tissue was counted after overnight culture at 37°C. As seen in Supplementary Figure S2B, the mean CFU in skin treated with1 or 2×MIC L-K6-treated groups was found to reduce 79–89.5% on day 3 and 97.1–98.8% on day 6 compared with that of the bacterial recovered from the physiological saline-treated wounds on day 3 and day 6, respectively. Collagen deposition was quantified by hydroxyproline assay in the lesions at 3 and 6 days after wounding. As shown in Supplementary Figure S2C, hydroxyproline levels in the L-K6-treated group were significantly higher than those in the physiological saline-treated group, and they increased as the L-K6 concentration increased with values ranging 52.0–56.6% at day 3 and 52.4–58.8% at day 6. The mupirocin ointment-treated group had lower hydroxyproline levels than that of L-K6-treated group, but it levels were higher than those of the physiological saline-treated group. No differences were found in hydroxyproline levels at 3 and 6 days.
L-K6 inhibits LPS-induced proinflammatory response by MAPK and IκB signaling pathways
Our previous studies have demonstrated that L-K6 and its analogs can effectively kill gram-positive and gram-negative bacteria with an MIC of 3.125 µM against E. coli and S. aureus; they can even completely kill E. coli and S. aureus at 12 µM in 60 min by rapidly inducing membrane depolarization, resulting in the disruption and disability of plasma membranes [20,21]. Meanwhile, we also found that L-K6 preferentially kills cancer cells with much lower toxicity to noncancerous cells via a nucleus-targeting mechanism and DNA damage and without significant cell membrane, cytoskeleton or mitochondrial disruptions . In this study, we showed that L-K6 and its analogs at a concentration of 20 µM (at which cell viability was more than 90%) effectively suppressed the production of proinflammatory cytokines, including TNF-α, IL-6, MIP-1 and MCP-1, in LPS-stimulated human macrophage U937 cells. These results agreed with observed changes in proinflammatory cytokine/chemokine gene expression. When U937 cells were treated with LPS, the expression levels of five genes (TNF-α, IL-6, IL-8, MIP-1 and MCP-1), which are closely related inflammatory cytokines, were up-regulated. L-K6 and its analogs effectively suppressed mRNA expression levels in LPS-simulated U937 cells. Notably, 20 µM L-K6 can completely kill E. coli and S. aureus but has no effect on human macrophage U937 cells, suggesting that L-K6, which has broad-spectrum antimicrobial activities, is also a potent anti-inflammatory drug.
L-K6V1 and L-K6V2 were prepared by replacing Leu5 residues with Val residues and replacing Leu5 and Leu11 residues with Val residues in L-K6, respectively (Table 1). The introduction of Val residues led to decreases in the anti-inflammatory activity of L-K6V1 and L-K6V2compared with L-K6. L-K6V1 and L-K6V2 have the same net positive charge (+7) with L-K6, but their low hydrophobicity values (10.9 and 10.1, respectively) may affect anti-inflammatory activity compared with L-K6 (H value is 11.6), suggesting that hydrophobicity of peptides is important for anti-inflammatory activity. L-K5 has higher hydrophobicity values than L-K6, its inhibition for the production of inflammatory factors TNF-α, IL-6, MCP-1 and MIP-1 was similar to that of L-K6.
In the LPS-induced inflammatory response, LPS first binds to TLR4 through complex formation with LBP and membrane receptor CD14, subsequently causing the release of proinflammatory cytokines, such as tumor necrosis factor TNF-α, IL-6, IL-8, IL-10 and MCP-1 by LPS through the nuclear factor (NF)-κB and MAPK signal transduction pathways [32–34]. MyD88 is the first intracellular signal protein trigged by TLR 4 in the LPS-induced inflammatory response . We found that L-K6 and its analog L-K5 significantly suppressed the expression of MyD88 in LPS-treated U937 cells, indicating that L-K6 and L-K5 inhibited the LPS-induced inflammatory action by a MyD88-dependent signaling pathway. MyD88 may trigger LPS-induced production of proinflammatory cytokines/chemokines by activating the MAPK or NF-κB signaling pathway. Western blot results showed that L-K6 significantly suppressed phosphorylated MAPKs, including ERK, JNK and p38, in LPS-simulated U937 cells but had no effect on the intensities of total ERK, JNK and p38. However, L-K5 failed to suppress the expression of three types of phosphorylated MAPK proteins, namely, ERK, JNK and p38. In addition, L-K5, like L-K6, exhibited effects on the other signaling pathways, significantly suppressing the expression levels of IκB and NF-κB in cells treated with LPS. Thus, the data suggest that L-K6 prevented LPS from inducing the MAPK pathways and subsequent NF-κB translocation; however, L-K5 inhibited LPS-induced NF-κB activation by inhibiting the phosphorylation of IκB in U937 cells.
L-K6 neutralization and dissociation of LPS micelles as major responsible for its anti-inflammatory mechanism
LPS, a negatively charged LPS, is anchored to the outer membrane in gram-negative bacteria through its hydrophobic lipid A moiety . As a pathogen-associated molecular pattern (PAMP), LPS can bind to pathogen recognition receptor TLRs in macrophages, rapidly activating innate immune responses . Previous studies have explored blocking the LPS-induced cascade by directly binding to and neutralizing LPS to preclude or inhibit LBP binding. It was reported that cathelicidin peptide (CAP)11, beta-defensin (DEFB)123 and anti-LPS factors (ALFs) can abrogate inflammatory responses, reduce organ damage and increase survival rates of mice infected with bacteria by down-regulating TNF-α expression and modulating the inflammatory response in RAW264.7 cells [36–38]. Our ITC experiments also demonstrated that L-K6 has a strong ability to neutralize LPS by an exothermic interaction between net positively charged peptides and highly negatively charged LPS. A similar result was also confirmed by the TAL assay. L-K6 can significantly inhibit the LPS-mediated activation of TAL enzyme and neutralize 30% of LPS at a concentration of 20 µM, in which cell viability was greater than 80%. The neutralization of L-K6 to LPS may inhibit the binding of LPS onto cell membrane receptor TLR 4 and gives L-K6 its anti-inflammatory response ability. Compared with L-K6, L-K5 showed significantly low-binding ability to LPS, with only 18% inhibition of the TAL enzyme at a concentration of 20 µM, and the high dissociation constant Kd. Notably, L-K5 only lacks one Lys residue at the COOH-terminal compared to L-K6. Why an apparently small difference between amino acid components results in a notable impact on the binding ability to LPS? Maybe the additional Lys residue at the COOH-terminal of L-K6 brings the change in the secondary structure. Previous study indicated that COOH-terminal of peptides played an important role in helical content [39,40]. Our previous work also showed that the observed hydrophobicity of L-K6 (the retention time of 4.8 min) dramatically decreased compared with that of L-K5 (the retention time of 11.0 min), but the H values varied only from 12.3 to 11.6 . The large change in the hydrophobicity suggests that the additional Lys residue may affect the conformation of the peptide.
CD spectra data of L-K6 and L-K5 in LPS solution can further confirmed the structural changes of peptides. L-K6 and L-K5 showed no defined secondary structure in SP buffer, but they exhibited significant degree of conformational changes in LPS micelle environments. L-K5 adopted ordered secondary structure of ∼81%, but L-K6 showed only 65% α-helical structures and 1% β-sheet structures.
The conformational transitions in peptide/LPS complexes have previously been correlated to anti-inflammatory effects . Previously reported LPS interactions with peptide NK-2 demonstrated that electrostatic interactions are necessary for efficient LPS neutralization . Singh et al. also found that cationic peptides have stronger binding affinities for LPS than non-cationic peptides .
It has also been found that LPS aggregates have a particularly potent endotoxic effect; oligomers can also inhibit the LPS-induced proinflammatory response and reach anti-endotoxic effects to disaggregate LPS . We have found here that L-K6 significantly reduced LPS aggregates, thus preventing them from reacting with LBP or cellular receptors. In line with this finding, Rosenfeld et al. reported that human LL-37 and magainin led to the dissociation of LPS aggregates .
Mookherjee et al. summarized two central emerging themes regarding the mechanism of the anti-inflammatory activity of cationic host defense peptides: the ability to directly bind to LPS and the ability to modulate signaling through LPS-induced TLR in the NF-kB pathway . However, the mechanisms of the anti-inflammatory activity of these peptides appear to be more complex and unknown, but there is emerging evidence that these peptides act at multiple points of intervention [16,45].
L-K5 is taken up into U937 cells by TLR 4 endocytosis and partly inhibits the inflammatory response through intracellular molecules
Our previous study has demonstrated that L-K6 can be rapidly internalized into breast cancer cells (MCF-7 cells) through mainly clathrin-independent macropinocytosis without significant cell surface disruption, cytoskeleton disruption or mitochondrial impairment . In this study, we also demonstrated the cellular uptake and cytosolic localization of FITC-labeled L-K6 and L-K5 in human U937 cells using confocal microscopy and flow cytometry. L-K6 can disrupt the stability of bacterial cell membranes, which leads to bacterial death [19,20]; however, in contrast with the energy-independent manner of pore-forming peptides, the membrane permeability of human macrophage U937 cells was not impacted by L-K6 and L-K5, as L-K6 and L-K5 did not change the membrane potential of U937 cells. This result is not similar to those of earlier studies, which suggested that most cationic AMPs are internalized into cells mainly through pore formation, causing cell lysis, membrane permeabilization or other forms of bilayer disruption [46,47]. Here, L-K6 and L-K5 uptake into cells was not dependent on the disruption of cell membranes. However, low temperature diminished peptide uptake, indicating that the uptake of L-K6 and especially L-K5 into U937 cells occurred through a thermo-sensitive active mechanism, as would be expected if uptake were dependent on receptors.
CD14 is a necessary membrane receptor that binds LPS to TLR4. L-K6 almost completely inhibited the binding of FITC-labeled anti-CD14 antibody onto CD14 proteins on the surface of U937 cell membranes, and the inhibition rate was 91.1%. However, L-K5 had little effect on the binding of FITC-labeled anti-CD14 antibody on CD14 proteins on the surface of U937 cell membranes, and the inhibition was only 8.6%.
TLR4 may be internalized into the endosome network, where the second signaling pathway is triggered through adaptors TRAM and TRIF . L-K6 can be incorporated into TLR4 by either CD14 or LPS. Based on these aspects, we hypothesized that L-K6 itself is incorporated into U937 cells by TLR4 endocytosis. However, the addition of TAK242, an inhibitor of TLR4, did not affect L-K6 uptake into cells, whereas it significantly inhibited ∼97% of L-K5 uptake into U937 cells. These observations suggest that TLR4 endocytosis is important for the uptake of L-K5 by U937 cells. Interestingly, LPS-pretreatment inhibited L-K6 uptake, but not L-K5 uptake, into U937 cells.
Based on our study, we can speculate that the action of L-K6 and L-K5 on human U937 cells is as follows. L-K6 may be internalized by U937 cells by direct insertion into the cell membrane, as proposed by previous studies on cationic peptides [48,49]. However, L-K5 may be internalized by U937 cells, mainly by endocytosis of the cell surface receptor TLR4 or partly by direct insertion into the cell membrane.
Previous work has suggested that cellular uptake may be essential for the immunomodulatory activity of cationic AMPs with antimicrobial and anticancer activities, and some intracellular interacting protein partners and putative cell surface receptors have been identified [45,50–52]. Human macrophage U937 cells were pretreated with L-K6 and L-K5 for 60 min to force the peptides to be incorporated into cells before LPS stimulation for 24 h to determine whether the main target of L-K6 and L-K5 is LPS or intracellular proteins. Our results showed that LPS-induced TNF-α and IL-6 production was lower in simultaneous treatment with L-K6 and LPS than with L-K6 pretreatment for 60 min, suggesting that L-K6 can inhibit the inflammatory response by the intracellular target; however, 80% of the inhibition resulted from the interaction of LPS. The identification of intracellular receptors for L-K5 and L-K6 in U937 cells warrants further investigation and is beyond the scope of this article. Interactions of the peptides with putative intracellular protein partners may facilitate alterations of innate immune signaling pathways, resulting in the overall modulation of inflammatory responses. The results from this study provide evidence that supports further research into the development of peptides as potential therapeutics in immune-mediated chronic inflammatory diseases.
This study shed light on several steps regarding the mechanism of L-K6 and its analogs on inhibiting inflammatory response induced by LPS and enabled better understanding of the specific roles of the cationicity and hydrophobicity of the peptide in LPS binding and permeation. L-K6 inhibits proinflammatory responses by interaction with LPS, including neutralizing LPS and diassociating LPS micelles to inhibit LPS from triggering the proinflammatory signaling pathway, and by partially inhibiting inflammatory responses by the intracellular target. However, L-K5 may mainly inhibit proinflammatory responses by intracellular reporters to modulate the NF-κB signaling pathway. These results indicate that L-K6 and L-K5, which possess both antimicrobial and anti-inflammatory activities, may be promising molecules for the developing therapies for infectious inflammation.
bis-[1,3-dibutylbarbituric acid] trimethine oxonol
dynamic light scattering
isothermal titration calorimetry
laser scanning microscope
mitogen-activated protein kinases
pathogen-associated molecular pattern
Tachypleus Amebocyte Lysate
Toll-like receptor 4
tumor necrosis factor alpha
D.S. designed the project. W.D. and X.Z. performed ELISA, qPCR, WB and FACS experiments. X.Z., Y.Y. and X.Y. performed cytotoxicity, TAL, dissociation of LPS and anti-endotoxin experiments. X.Y. performed CD, DLS and ITC experiments. L.S. performed animal experiments. W.D. and D.S. analyzed the data, wrote and edited the manuscript.
This work was supported by grants from National Natural Science Foundation of China .
The Authors declare that there are no competing interests associated with the manuscript.