Intracellular synthesis of gold nanoparticles by Gluconacetobacter liquefaciens for delivery of peptide CopA3 and ginsenoside and anti-inflammatory effect on lipopolysaccharide-activated macrophages

Abstract Probiotic Gluconacetobacter strains are intestinal microbes with beneficial effects on human health. Recently, researchers have used these strains to biosynthesize metal and non-metal nanoparticles for treating various chronic diseases. Despite their importance in nanotechnology, gold nanoparticles (AuNPs) biosynthesized by Gluconacetobacter species have not been clearly identified for treating inflammation and inflammation-associated diseases. While ginsenoside CK has strong pharmaceutical activity, it also has strong cytotoxicity and hydrophobicity which is hurdle to make formulation. Peptide–nanoparticle hybrids are gaining increasing attention for their potential biomedical applications, including human inflammatory diseases. Herein, we developed peptide CopA3 surface conjugated and ginsenoside compound K (CK) loaded gold nanoparticles (GNP-CK-CopA3), which intracellularly synthesised by the probiotic Gluconacetobacter liquefaciens kh-1, to target lipopolysaccharide (LPS)-activated RAW264.7 macrophages. The synthetic GNP-CK-CopA3 was characterised by various instrumental techniques. The results of our cellular uptake and MTT assays exhibited obvious drug intracellular delivery without significant cytotoxicity. In addition, pre-treatment with GNP-CK-CopA3 significantly ameliorated LPS-induced nitric oxide (NO) and reactive oxygen species (ROS) production and suppressed the mRNA and protein expression of pro-inflammatory cytokines in macrophages. Furthermore, GNP-CK-CopA3 efficiently inhibited the activation of the nuclear factor-κB (NF-κB) and mitogen-activating protein kinase (MAPK) signalling pathways. Taken together, our findings highlight the potential of using peptide–nanoparticle hybrids in the development of anti-inflammatory approaches and providing the experimental foundation for further application. Graphical Abstract


Introduction
Panax ginseng Meyer, a perennial plant of the Araliaceae family, is one of the most well-known herbal medicines and has been used in Eastern countries for more than two thousand years. Ginseng exhibits various pharmacological effects including, anti-cancer, anit-diabetes, anti-inflamaroty, anti-oxidanat, stimulation of immune system, and nervous system, hyperglycaemia, and cardiovascular diseases [1][2][3][4]. Ginsenoside compound K (CK) is a major intestinal bacterial metabolite of the protopanaxadiol-type ginsenosides. CK has attracted wide attention because of its multiple pharmacological actions, such as anti-cancer, anti-inflammatory, antiaging, and anti-diabetic effects [5,6]. However, its clinical application is significantly restricted by its low water solubility and poor permeability [7]. To overcome those issues, various nanotechnology-based systems have been developed to improve drug delivery and thereby overcome low therapeutic effectiveness [8]. Inflammation is the major response raised by the body to address injuries. Despite inflammation establishing a vital constituent of tissue repair, it is well recognised that unrestrained or chronic inflammation becomes deleterious, leading to progressive tissue damage. This complex, tightly regulated process serves as a rapid defence mechanism to contain potential pathogens, limit further tissue damage, and stimulate repair mechanisms; consequently, inflammation is crucial for human health [9]. Ginsenoside CK is well-known for its anti-inflammatory effects in lipopolysaccharide (LPS)-activated RAW 264.7 cells [4], LPS-induced septic mouse brains [10], and 2,4,6-trinitrobenzene sulphuric acid-induced colitic mice [11]. Its underlying mechanisms include inhibition of activation of nuclear factor NF-jB and mitogen-activated protein kinases [10,11].
The potential applications of gold nanoparticles (GNPs) in diagnosis, therapy, and drug delivery have been extensively studied [8]. Various chemical, physical, and electrochemical technologies have been developed to generate GNPs. However, the toxic reagents limited the biomedical applications of nanoparticles [12]. So recently, many researchers, including our group, have focussed on the biological synthesis of GNPs to allow easy synthesis, biocompatibility, distinctive physicochemical properties, and other benefits [13]. Microorganisms can be used for the synthesis of GNPs because of their considerable ability to reduce metal ions without producing any toxic by-products [14,15]. Previous studies showed that Lactobacillus strains can be used to extracellularly or intracellularly biosynthesize GNPs [15,16]. Herein, we synthesised GNPs using probiotic Gluconacetobacter liquefaciens kh-1 (family Acetobacteraceae) for CK loading.
Herein, CK and CopA3 have been loaded and bioconjugated on GNPs and then evaluated its anti-inflammation in LPS-induced RAW264.7 macrophages.

Analysis of nanoparticle-producing strain and growth condition evaluation
Strain kh-1 was kindly provided by Seung-Cheol Koh (Professor, Korea Maritime University) and cultured on a solid nutrient medium containing 5 g/L yeast extract, 3 g/L peptone, 25 g/L mannitol, and 15 g/L agar at 30 C for 48 h. The colonies were picked from agar plates and suspended in a modified YPM medium containing 5 g/L yeast extract, 3 g/L peptone, 25 g/L mannitol, and 15 g/L agar. All culture media were sterilised at 121 C for 15 min. The phylogenetic tree showed in Supplementary Figure 1, was constructed following our previous method [11], and the identified 16S ribosomal RNA sequences were deposited in NCBI GenBank under accession No. MN209894.

Intracellular synthesis of GNP-CK
The ginsenoside CK-loaded GNPs were synthesised according to our previously published with few modifications [16]. G. liquefaciens kh-1 was cultivated in MRS broth at 27 C for 24 h, and the activated culture was inoculated into MRS broth at 27 C on a shaker bed at 200 rpm for 24 h. After incubation, the fresh biomass was collected by centrifuging at 3,000 rpm for 10 min at 4 C. The collected bacterial pellets were washed and resuspended in the sterile water (10 ml). The gold chloride trihydrate solution (HAuCl 4 Á3H 2 O, final concentration of 1 mM) and CK (final concentration of 1.6 mM) were inoculated with G. liquefaciens and incubated at 27 C in a shaker bed for 24 h. Afterward, the synthetic of GNPs was visually observed according to the colour change. The mixtures were conducted ultra-sonication in ice for 30 min to release GNPs and then centrifuged at 3000 rpm for 10 min to remove the bacterial pellets. Finally, the CK-loaded nanoparticles (GNP-CK) were collected by high-speed centrifugation (12,000 rpm for 10 min at 4 C), washed exhaustively and resuspended in sterile water.

Conjugation of GNP-CK with peptide CopA3
To conjugate the peptide CopA3, a carboxylic acidterminated alkanethiol monolayer was first formed on the surface of the nanoparticles, as previously reported [28]. Afterward, 100 lM CopA3 (dissolved in phosphate-buffered saline, PBS) was added to the 1-ethyl-3-(3-dimethylamino-propyl)carbodiimide hydrochloride (EDC) and N-hydroxysulfosuccinimide (Sulfo-NHS) activated nanoparticles solution, allowed to react using a rotary shaker at room temperature for 4 h, and stored at 4 C overnight. 1 M glycine and 10 mM Tris (pH 7.5) were added to quench the excess hydroxylamine. The solution was centrifuged at 12,000 rpm for 20 min to remove any unbound peptide and collect the precipitate. Finally, the synthetic peptide-nanoparticles hybrids, GNP-CK-CopA3, were resuspended in PBS and stored at 4 C for further study.

Cell viability assay
NHDFs, HaCaTs, and RAW264.7 cells were separately seeded in 96-well plates at a density of 1 Â 104 cells/well and incubated for 24 h to allow attachment. In vitro cytotoxicity was analysed after treatment with or without GNP-CK-CopA3 at the indicated concentrations. After 24 h, 100 lL MTT (0.5 mg/ mL) was added, and cells were incubated for another 3 h. Then formazan crystals were dissolved in 100 lL DMSO and the optical density was measured at 570 nm by a microplate reader (Molecular Devices Filter Max F5; San Francisco, CA, USA).
Enhanced dark-field (EDF) microscopy RAW264.7 cells were seeded and grew on 22-mm coverslips in 6-well plates for 24 h and then were incubated with GNP-CK-CopA3 (40 mg/mL) for 3 h. After rinsing with PBS twice, cells were fixed with 4% paraformaldehyde, prepared as slides, and sealed by clear fingernail polish, and finally observed (100Â) in a high-resolution illumination microscopy system (Cytoviva Inc., Auburn, AL, USA).

Transmission electron microscopy (TEM)
RAW264.7 cells were seeded in 6-well plates for 24 h and treated with 20 or 40 mg/mL of GNP-CK-CopA3 for 3 h. According to the previously reported [31], collected cell pellets were fixed with 2.5% glutaraldehyde at 4 C for 8 h, post-fixed with 1% osmium tetroxide for 2 h, and gradually dehydrated with 50, 70, 90, and 100% ethanol for 15 min each. Samples were then embedded in Epon (Sigma-Aldrich). Ultrathin sections (70 nm) were cut in a Leica EM ultramicrotome (Wetzlar, Germany) and put on Cu grids. The sections were finally captured on JEM-1010 TEM (Joel, Tokyo, Japan) operated at 80 kV.

Staining of Lyso-Tracker
RAW264.7 cells grown on the 22-mm coverslips in 6-well plates were treated with GNP-CK-CopA3 for 6 h. After incubation, cells were gently washed with PBS and then 1 ml prewarmed medium containing 50 nM LysoTracker TM Green DND-26 (Invitrogen, Ltd., Paisley, UK, Ex/Em 504/511 nm) were added. Following a 30 min incubation, the cells were observed by a Leica DM IRB inverted fluorescence microscope with green excitation as the fluorescence filter.
Nitric oxide (NO) assay RAW264.7 cells grown in 96-well plates were pre-treated with GNP-CK-CopA3 (10-100 lg/mL) for 1 h and followed by adding LPS (1 lg/mL) for 24 h. 100 mL culture supernatant was reacted with 100 mL Griess reagent, and the absorbance was measured at 570 nm by a microplate reader.

Intracellular reactive oxygen species (ROS) production
Cells were treated with GNP-CK-CopA3 (20 and 40 mg/mL) for 1 h prior to exposure with LPS (1 lg/mL) for 24 h. LPSinduced intracellular ROS release was detected according to a Cellular ROS/Superoxide Detection Assay Kit (Cambridge, MA, USA, Ex/Em 490/525 nm). Fluorescence was measured using an LSM 510 and 510 META laser scanning microscope.

Reverse transcription-polymerase chain reaction (RT-PCR)
Cells were pre-treated with GNP-CK-CopA3 for 1 h and then treated with LPS (1 lg/mL) for another 6 h. Total RNA was prepared by a Trizol reagent kit (Invitrogen, Carlsbad, CA, USA). The AccuPower RT PreMix Kit and AccuPower HotStar PCR PreMix Kit (Bioneer, Daejeon, Korea) were performed for RT-PCR following the previously published [29]. Sequences for all primers are provided in Table 1.

Western blot analyses
Cells were pre-treated with GNP-CK-CopA3 for 1 h and then treated with LPS (1 lg/mL) for another 2 or 4 h. The protein was extracted using Pierce TM RIPA Buffer (Thermo Scientific, Rockford, USA) and quantified using Protein Assay Dye Reagent Concentrate (Bio-Rad, CA, USA). The total proteins (50 lg) were separated by 10% SDS-PAGE and transferred to PVDF membranes using Protein Gel Electrophoresis Chamber System and iBlot 2 Dry Blotting System (Thermo Fisher Scientific). The membranes were blocked by 5% skim milk at room temperature for 1 h followed by incubation with primary antibodies overnight at 4 C. After that, the membranes were conjugated with the appropriate horseradish peroxidase-linked secondary antibodies (Santa Cruz, CA, USA) at room temperature for 1 h. The bands were detected with the West-Q Pico ECL Solution (GenDEPOT, Barker, USA) and quantified using ImageJ software.

Immunofluorescence staining
The effect of GNP-CK-CopA3 nuclear translocation of NF-kB p65 was examined by immunofluorescence assay. Cells were fixed in 4% paraformaldehyde and permeabilized with 0.1% Triton X-100 in PBS, and blocked with 2% BSA for 1 h at 37˚C. The anti-NF-jB p65 (1:1000) antibody was applied for 1 h followed by a 30 min incubation with fluorescein isothiocyanate (FITC)-conjugated donkey anti-rabbit IgG. After washing in PBS, nuclei were stained with Hoechst 33258, and fluorescence was captured (400Â magnification) using a Leica DM IRB inverted fluorescence microscope and quantified using ImageJ software. The nuclei-positively stained cells were counted from five random fields in each well.

Characterisation of GNP-CK-CopA3
Gluconacetobacter liquefaciens kh-1 was identified as Gluconacetobacter liquefaciens IFO 12388 with high similarity (98%) and G þ C mole% content (39.70 mol%) of the genomic DNA have been well characterised previously (Supplementary Figure S1). The biosynthetic GNP-CK-CopA3 was confirmed using a UV-Vis spectral analysis showing significant absorbance at 550 nm (Figure 1(A)). TEM images indicated that the nanoparticles were spherical-shaped with a diameter of 10-30 nm (Figure 1(B)). Meanwhile, these nanoparticles displayed polycrystalline orientation, and the lattice fringes with d-spacing of 0.23 nm corresponded to the (111) lattice plane of gold ( Figure  1(C)). The distributions of gold elements were clearly visible in the elemental mapping images (Figure 1(D,E)), and the characteristic peak of metallic gold at 2.3 keV was determined by EDX spectroscopy (Figure 1(H)). XRD and SAED analyses showed four major diffraction peaks at 2h values of 38.18 , 44.40 , 64.56 , and 77.55 , which correspond respectively to the (111), (200), (220), and (311) lattice planes of Bragg's reflection (Figure  1(F,G)). The photoluminescence (PL) spectrum showed the intensity of GNP-CK-CopA3 at 500 nm (Figure 1(I)). We suspect that those peaks result from, on the one hand, the physical interactions (electrostatic, hydrophobic, and affinity interactions) between the ginsenoside CK and GNPs. On the other hand, the peptide CopA3 was covalently linked to GNPs via the coupling of the peptide with amine groups to the carboxylic-group-functionalised of GNPs.

Uptake and intracellular localisation of GNP-CK-CopA3
Cellular uptake of GNP-CK-CopA3 in RAW264.7 macrophages was detected using bright-field and enhanced dark-field (EDF) microscopy ( Figure 2(A)). After 3 h incubation, GNP-CK-CopA3 appeared to be uptake by macrophages, as indicated Table 1. List of primers sequences used in this study.

Gene name
Forward Reverse in the aggregated bright white spots. Subsequently, intracellular localisation of GNP-CK-CopA3 was further observed using TEM images (Figure 2(B)). GNP-CK-CopA3 were detected on the membrane and membrane-bound organelles such as endosomes, lysosomes, and large endosomalvacuoles in RAW264.7 cells. Furthermore, we examined the fluorescence responses of LysoTracker green after treating with GNP-CK-CopA3 for 6 h in RAW264.7 cells. Hoechst 33258 nucleic acid stain was used for staining the nucleus. In GNP-CK-CopA3-treated cells, lysosomes were visible as specific foci (Figure 2(C)), indicating that GNP-CK-CopA3 disrupts lysosomes in macrophages.

Cytotoxicity of GNP-CK-CopA3
The cytotoxicity of GNP-CK-CopA3 was investigated using the MTT assay in the NHDFs, HaCaTs, and RAW264.7 cells. In NHDFs (Figure 3(A)), no toxicity was noticed from 2.5 to 80 lg/mL GNP-CK-CopA3, in HaCaTs, a small inhibitory effect was noticed at 80 lg/mL, indicating that HaCaTs was more sensitive (Figure 3(B)) than NHDFs. However, it is clear that 80 lg/mL GNP-CK-CopA3 caused toxicity (31.9% death) to the RAW264.7 cells (Figure 3(C)). As a control, the cytotoxicity of CopA3 was investigated using the MTT assay in RAW264.7 cells (Supplementary Figure S2), and also the cell cannot survive under same concentration of CK.

Effects of GNP-CK-CopA3 on NO production
The NO secretion in LPS-activated macrophages was indirectly estimated by nitrite determination. As shown in Figure 3(D), pre-treatment with GNP-CK-CopA3 (20 and 40 mg/mL) significantly inhibited the LPS-induced (1 lg/mL) nitrate levels in macrophages without cytotoxicity (Figure 3(C)).

ROS production
It has been reported that ROS are essential for LPS-induced inflammation. [32] ROS secretion was elevated by 76.6% after LPS treatment (Figure 3(E,F)). However, pre-treatment with 20 and 40 mg/mL GNP-CK-CopA3 effectively inhibited ROS production by 40.4 and 65.05%, respectively.

Effects of GNP-CK-CopA3 on the NF-jB signalling pathway
In macrophages, various intracellular signalling pathways, such as the nuclear factor kappa B (NF-jB) pathway, have been found to be highly associated with the anti-inflammatory action of GNPs [34]. Herein, RAW264.7 cells were pretreated with GNP-CK-CopA3 (20 or 40 lg/mL) for 1 h and then irritated by LPS (1 mg/mL) for 2 h [35]. In LPS-activated inflammatory condition, inhibitor kappa B-alpha (IjBa) was phosphorylated and ubiquitinated, leading to the nuclear translocation of NF-jB from the cytoplasm ( Figure 5(A-E)), whereas pre-treatment with the nanoparticles efficiently inhibited the protein IjBa phosphorylation and degradation. Additionally, nuclear NF-jB p65 subunit protein expression was also inhibited by pre-treatment GNP-CK-CopA3, indicating the nuclear transcription of NF-jB was suppressed in LPSirritated RAW264.7 cells. By immunofluorescence staining, we further examined whether GNP-CK-CopA3 inhibits the activation of NF-jB in RAW264.7 cells. As expected, LPS stimulation induced translocation and accumulation of NF-jB p65 from the cytoplasm to the nucleus, however, it was dramatically inhibited in GNP-CK-CopA3 ( Figure 5(F))

Effects of GNP-CK-CopA3 on the MAPK signalling pathway
The mitogen-activating protein kinases (MAPKs) are one of the key transcription factors in regulating inflammation-specific genes. The phosphorylation of extracellular signal-regulated kinase (ERK) and p38 MAPK significantly increased after LPS treatment, but pre-treatment with GNP-CK-CopA3 (20 or 40 lg/mL) for 1 h dramatically attenuated the LPS-induced phosphorylation ( Figure 6). Meanwhile, the total expression levels of ERK and p38 did not differ significantly among the groups.

Discussion
Recently, GNPs have been used to deliver drugs for cancer therapies and immune-regulation, such as 5-fluorouracil [36], doxorubicin [37], and curcumin [7]. In our previous report, we used Lactobacillus kimchicus DCY51 T to synthesise GNPs for the delivery of ginsenoside CK. The CK-loaded GNP was confirmed as an effective photothermal therapy agent with synergistic chemotherapeutic effects for the treatment of cancer [38]. Additionally, to improve the stability and function of the CK-loaded GNPs, we conjugated with the peptide CopA3. The CopA3 is a 9-mer peptide (LLCIALRKK-NH 2 ) derived from coprisin [39]. Its amino acid residues were used to covalent coupling to the carboxylic-group-functionalised GNPs [17]. Yang et al. [40] suggested that peptidecapped GNPs achieved more cellular internalisation and prolonged intracellular retention compared with citratecapped GNPs because of their positive surface charges. GNP-CK-CopA3 nanoparticles was smaller in size and were accumulated intracellularly by endocytosis. And it endured by EDF, TEM, and Lyso-Tracker images indicated that GNP-CK-CopA3 were trapped as clusters in either endosomes or lysosomes in RAW264.7 cells. CK has excellent physiological activity but it also showes cytotoxicity. Synthesised GNP-CK-CopA3 contain a small amount of CK (lower than 10 ppm), which may lower cytotoxicity with good permeability effect in to Raw264.7 cells. Peptide conugated GNP-CK-CopA3 can be suggested for reducing the cytotoxicity of drug, inducing more biological activity, and increasing durg absorption into the target cells.
Macrophages are the primary contributors to the initiation of potentially pathological inflammation and the progression of autoimmune and auto-inflammatory diseases including cancer, rheumatoid arthritis, cardiovascular disease, atherosclerosis, and Alzheimer's disease [41]. In the early phase of a disease, the innate immune system and Toll-like receptor (TLR) signalling can recognise and respond to diverse microbial epitope pathogen-associated molecular patterns such as LPS [42]. However, excessive TLR responses cause many acute and chronic human inflammatory diseases and cancers. Activated TLR4 elicits the production of pro-inflammatory mediators and cytokines, including NO, iNOS, COX-2, TNF-a, IL-1b, and IL-6. During cellular metabolism, the overproduction of ROS can also contribute to the development of cancer, aging, and inflammation [32]. Zhu et al. reported that GNPs significantly attenuated the rise in ROS/RNS-mediated inflammatory responses [43]. Similar to the previous findings, our preliminary study also indicated that CopA3 inhibited the NO production in LPS-stimulated RAW264.7 cells (Supplementary Figure 2A) [27]. Therefore, in our study, the effective inhibition of GNP-CK-CopA3 in LPS-induced production of NO and overexpression of mRNA and protein of pro-inflammatory cytokines was probably due to the combined role of both CopA3 and nanoparticles.
Studies have suggested that both NF-jB and MAPK signalling activation are responsible for the initiation and progression of LPS-induced inflammation [44]. Ginsenoside CK has been widely studied the immunopharmacological activities both in vivo and in vitro. It was confirmed to regulate inflammation through inhibition of systemic inflammatory cytokines production, MAPKs activation [45], and NF-jB activation [46]. Dysregulated NF-jB is accompanied by the ubiquitination and degradation of IjBa, and the nucleus translocation of activated NF-jB subunits, which ultimately leads to the induction of pro-inflammatory cytokines [47]. In our study, GNP-CK-CopA3 showed anti-inflammatory action by suppressing the protein IjBa phosphorylation and NF-jB nuclear translocation in LPS-activated RAW264.7 cells.
Oxidative stress and ROS are also the causative factors of various inflammatory diseases. Overproduction of ROS induced the phosphorylation of three MAPKs, including JNK, ERK, and p38 [48]. Manna et al. [49] reported that pomegranate peel extract-stabilized GNPs reduced the pro-inflammatory burden by modulating the MAPK and NF-jB pathways. In the current study, the phosphorylated form of ERK and p38 was decreased by GNP-CK-CopA3 in LPS-stimulated RAW264.7 macrophages. It seems unclear that GNP-CK-CopA3 has an anti-inflammatory effect through directly inhibiting NF-kB-signalling pathway or through neutralising LPS, since CopA3 is an AMP. Compared to CopA3, GNP-CK-CopA3 showed more significant anti-inflammatory effect towards RAW264.7 cells, can be suggested for reducing cytotoxicity of drug, inducing more biological activity, and increasing drug absorption into the target cells. Whether or not direct interaction of GCk-CopA3 with LPS may thus have profound impact on antiimflamatory signalling events.