Mycobacterial lipid-derived immunomodulatory drug-liposome conjugate eradicates endosome-localized mycobacteria

Abstract

Tuberculosis is a challenging disease due to the intracellular residence of its pathogen, Mycobacterium tuberculosis, and modulation of the host bactericidal responses. Lipids from Mycobacterium tuberculosis regulate macrophage immune responses dependent on the infection stage and intracellular location. We show that liposomes constituted with immunostimulatory lipids from mycobacteria modulate the cellular immune response and synergize with sustained drug delivery for effective pathogen eradication. We evaluate the pH-dependent release of Rifampicin from the mycobacterial-lipid-derived liposomes intracellularly and in vitro, their cell viability, long-term stability, and antimicrobial efficacy. Intracellular drug levels were higher following liposome treatment compared with the free drug in a temporal fashion underlying a sustained release. The drug-encapsulated liposomes were taken up by clathrin-mediated endocytosis and elicited a robust pro-inflammatory immune response while localizing in the recycling and late endosomes. Notably, these were the same cellular compartments that contained the pathogen underlying localized intracellular targeting. Our results also imply a lipid-centric and species-specific selectivity of the liposomal drug formulations. This work provides a proof-of-concept for the dual-action of liposomes derived from the pathogen itself for their effective eradication, in conjunction with the attuned host immunomodulation.

1. Introduction

Tuberculosis (TB) holds the rank of the deadliest malady inflicting millions of people worldwide [1]. The absence of vaccines against TB makes drug therapy the only viable option, however, the emergence of drug resistance calls to attention the development of new therapies for improving patient outcomes [2]. Consequently, considerable efforts are directed towards improved platforms for the delivery of anti-TB drugs, in addition to the discovery of novel chemotypes and/or the improvision of existing drugs. In this context, liposomes are attractive drug delivery vehicles due to their biodegradable, biocompatible, controlled release profile, and generally non-toxic nature [3,4].

Mycobacterium tuberculosis (Mtb), which causes TB, mitigates anti-microbial responses in host macrophages either via suppression of intracellular generation of bactericidal reactive oxygen or nitric oxide species (ROS/NO) and modulates secretion of pro-inflammatory cytokines such as Interleukin-12 (IL-12) and interferon-gamma (IFN-y) [5–8]. Thus, delivery vehicles that enable dual action of organelle-specific delivery of the payload and concomitant modulation of the cellular immune responses to enhance the drug action/pathogen eradication represent intelligent multimodal avenues in liposomal nanomedicine against TB [4]. Lab model of Mtb, M. smegmatis (Msm), has a very similar composition of the cell wall lipids, produces inflammatory cytokines, and induces macrophage apoptosis through the Endoplasmic Reticulum (ER) stress response [9–11]. Interestingly, the immunogenicity of lipid fractions extracted from Msm has been recently documented [12,13]. In general, bacterial-derived lipids possess inherent anti-microbial activity, and nanoparticles coated with lipids extracted from Staphylococcus aureus selectively target macrophage activity [14]. However, likewise, landmarks in tackling TB are rudimentary.

Rifampicin (Rif), a key antibiotic in the frontline treatment of TB, suffers from poor water solubility and stability, and accordingly, its encapsulation in anionic liposomes represents a viable strategy to increase its local concentration and selective delivery to bacteria-loaded phagocytic cells.Msm liposome is negatively charged [15–18]. Further, as isolated lipids of Mtb and Msm possess a spectrum of activity in modulating host cell immune responses shown by us and others [19–23], we hypothesized that the generation of a liposomal drug delivery system composed of extracted lipids of Msm would be capable of eliciting specific immune activation and would complement drug action to enable effective intracellular bacterial eradication. TB patients when administered with exogenous cytokines show improved treatment outcomes, underscoring the synergistic action of immunomodulation and drug action [24]. We report here a detailed physicochemical, immunological, and biological characterization of drug-liposome conjugates derived from Msm membranes. We dissect their mode of intracellular uptake, trafficking, and localization and show them to elicit a pro-inflammatory response. Immunogenic response acts in synergy with enhanced and sustained delivery of Rifampicin within Msm-ingested macrophages to kill the late-endosome localized bacteria in a specifies-specific manner. The inferior killing of intracellular bacteria with drug-loaded lecithin liposomes which are non-immunogenic supported the above findings. These data pave the way for further exploration of bug-centric personalized killing platforms consisting of immunostimulatory lipids derived from the pathogen itself. This should be coupled with the generation of stable liposomal formulations with well-characterized physicochemical properties and drug interactions.

2. Results and discussions

2.1. Physicochemical characterization and biocompatibility of Msm bacterial lipid-derived liposomes (BL-d liposomes)

The main Msm bacterial lipid constituents in the extracted fraction were glycerophospholipids (GPLs), phosphatidylinositol mannosides (PIMs, AC₂PIMs), lipomannans (LMs), and lipoarabinomannans (LAMs) (LMs and LAMs are derivatives of PIMs) and glycolipids and were identified by TLC and mass spectroscopy (Fig. S1A, Table S1) [10,15]. PIMs, LMs, and LAMs are critical for cell wall integrity and immune manipulation via the TLR2 pathway [25]. In highly systemic infections, such immunomodulating agents may improve the host defense mechanisms. The extracted lipids were free from bacterial DNA and protein contaminants (Fig. S2), and were successfully formulated as liposomes with or without Rifampicin (Fig. 1A, Table1); SEM image of THP-1 cell shows a cell-surface docked liposome. TEM micrographs confirmed the formation of unilamellar vesicles by the Msm lipid mixture (Fig. S1B). Further, these had average hydrodynamic diameters of 281 ± 45 (BL-d Lipo) and 398 ± 65 nm (Rifampicin-loaded BL-d; BL-d Rif Lipo), negative zeta potentials of –27.8 (BL-d Lipo) and – 25.6 (BL-d Rif Lipo) and low PDIs (Table 1). These ensure comparable in vitro behaviour (size-dependent), good colloidal stability preventing aggregation, and monodisperse solutions, respectively [26]. Low PDI indicates that the homogeneity of the liposomal size was not affected by drug incorporation. Liposomal stability in both the empty and drug-loaded formulations remained constant when stored for longer durations (Table S2). Biocompatibility of the BL-d liposomes was demonstrated by observing THP-1 macrophage viability (Fig. 1B), where no effect on the metabolic activity of THP-1 macrophages was seen and hence no cytotoxicity (with empty BL-d lipo and BL-d Rif Lipo); at very high concentrations a minor effect was seen.

Fig. 1.

(A) Representative SEM image of (left) Rif-loaded BL-d liposomes representing the uniformity in size, (middle) image of untreated THP-1 macrophages, and (left) represent the SEM image of THP-1 cells treated with BL-d liposomes for 2 h. Scale Bar = 200 nm. (B) In-vitro cytotoxicity of BL-d liposomes (–Rif, +Rif) on THP-1 cells at various indicated concentrations. Physicochemical characterization of bacterial lipid-derived Rif-loaded liposomes (C) Schematic representation of physicochemical characteristics (size, charge, lamellarity, fluidity, packing, and immunogenic) of an anionic liposome (D) (UPPER PANEL) Schematic representation of destabilization of the liposomal membrane before and after fusion with the endosomal membrane, showing a maximum release at endosomal pH. (LOWER PANEL) In vitro cumulative drug release of Rifampicin from BL-d liposomes at various pH. Effect of pH on the fluidity and anisotropy of drug-loaded and empty BL-d liposomes (E) Generalized polarization (F) Laurdan anisotropy (G) DPH anisotropy. BL-d liposome induces pro-inflammatory interleukins in THP-1 macrophages (H) Quantification of various interleukins levels per macrophage. ND: not detected. Concentrations were calculated by fitting the data with four parametric regression methods. Data represent mean ± SD of three independent experiments. Statistical significance was determined using two-way ANOVA with Dunnett’s multiple comparison tests, N = 3.

Table 1. Hydrodynamic size, PDI, and zeta potential of drug-loaded and empty BL-d liposomes at different pH evaluated using DLS.

pH	Liposomes	Size (nm)*	PDI*	Zeta potential (mV)*
7.4	BL-d Liposomes	281 ± 45	0.23	−27.8 ± 2.8
7.4	BL-d Rif Liposomes	398 ± 65	0.24	−25.6 ± 4.3
5.5	BL-d Liposomes	279 ± 73	0.24	−21.2 ± 1.4
5.5	BL-d Rif Liposomes	308 ± 87	0.23	−18.3 ± 6.0
4.5	BL-d Liposomes	296 ± 97	0.23	−19.7 ± 9.3
4.5	BL-d Rif Liposomes	321 ± 65	0.22	−18.9 ± 4.1

^*The results of size and zeta potential are expressed as mean ± SD, N = 2.

2.2. pH-responsive in vitro drug release from BL-d liposomes

Encapsulation efficiency (EE %) of Rifampicin was evaluated by both UHPLC-MS and UV and was found to be around 50% in a 1:10 drug: lipid ratio (Table S3). Since Rifampicin is hydrophobic, it is implicated to reside in the acyl hydrocarbon chain of the liposome, however, other locations within a bilayer are also possible. In general, encapsulation depends on the lipid acyl chain lengths (hence liposome size), lamellarity, packing density or fluidity of the lipid bilayer, head group structure and charges (Fig. 1C), hydrophobic and van Der Waals interactions and hence is lipid-composition specific. The location of the drug impacts its temporal release and is further modified by physiochemical parameters (such as temperature, pH, ionic strength, etc), that would alter lipid bilayer structure, eventually affecting drug partitioning and/or location. As we envisioned achieving targeted intracellular delivery of Rifampicin, a pH-responsive drug release was monitored, wherein the chosen pH conditions relate to the pH gradient found in the endosomal-lysosomal trafficking network. The physiological pH of the lung fluid/cytoplasm, pH = 7.4 (Fig. 1D (upper(U)-lower (L) panel), the phagosome/early endosome, pH = 6.2 (Fig. 1D (U-L panel)), late endosome/phagolysosomal pH = 5.5 (Fig. 1D (U-L panel), and the lysosomal pH = 4.5. The drug release was pH-dependent at later time points. Initially, Rif-loaded BL-d liposomes showed a biphasic drug-release pattern with an initial burst release of 30% during the first 8 h, followed by a 20% sustained release at simulated lung or cytoplasmic fluid (at pH 7.4). An enhanced and sustained drug release was mainly seen at the endosomal pH. The Rif-loaded BL-d liposomes had a maximum of 56.6% drug released under the simulated endosomal (pH 5.5) conditions within 22 h, post which a plateau was reached. The drug release profiles showed 29%, 32%, and 33% release within 10 h in the simulated intestinal, endosomal/phagosomal, and lung fluid pH conditions, respectively. Overall, the drug release at pH 6.5 and pH 4.5 was more sustained and constant, suggesting that low pH might perturb the bilayer, fostering drug release. This was further observed after 18 h at pH 5.5, where an enhanced and sustained release >45% was seen. In vitro drug release confirms that the Rif-loaded BL-d liposomes exhibit a higher release profile at acidic pH, thus hinting towards higher drug release during endosomal acidification inside the TB-infected macrophages.

The difference in the drug release rates between the acidic and neutral pH could be due to the disruption of the lipid bilayer structure, for instance lipid protonation at the head group. Induction of a non-lamellar phase that fuses faster with the endosomal membrane, could also release the drug contents efficiently at low pH (Fig. 1D). The amount of drug released during the first 5 h for Rif-loaded BL-d liposomes was independent of the pH. We attribute this initial burst release to loosely bound Rifampicin fraction to the liposomal membranes. Rifampicin fraction that tightly resides in the lipid bilayer is responsible for the more sustained and controlled release, wherein the pH-induced-destabilization of the bilayer at lower pH would account for higher release. Given the preferential localization of the drug at the interfacial head group region in these liposomes [28], the pH-induced modulation of the lipid head group region is likely to impact the drug residence time and hence alter the release profile. This controlled pH-responsive drug release profile could deliver the maximum drug at the intracellular infection site with a longer therapeutic potential and lower frequency of drug administration.

2.3. Acidic pH alters the packing and fluidity of the Msm lipid-based liposomes

To get further insight into the effects of low pH on BL-d lipid membranes, we calculated Laurdan GP (Generalized polarization, reported for interfacial membrane hydration and packing/order) and anisotropy (r, reporter of membrane fluidity) of two different membrane probes with their known depths and positions (Fig. 1 E-G) [27]. Rif-loaded BL-d liposomes demonstrated a high GP at physiological temperatures (37 °C) compared with free BL-d liposomes (Fig. 1E). This implies a higher packing efficiency at the interfacial head group region in the BL-d liposomes with RIF. Notably, interfacial localization of a structurally similar drug (Rifabutin) in these lipid bilayers is supported by our previous work [28]. The zeta potential of Rif-encapsulated BL-d liposomes decreased with a decrease in pH (Table 1), suggesting that the interaction of RIF with the liposomes is electrostatically driven. RIF is present in zwitterionic form at pH 7.0 with prominent intramolecular hydrogen bonding. At low pH, intermolecular H-bonding with water increases, thereby increasing the number of water molecules near the head group region. Further, the most abundant lipid components of BL-d are acylated PIMs with mannosylated headgroups; however, their acyl chains are primarily saturated [15]. This would afford a tighter intermolecular interaction (as seen in other gel or cholesterol-enriched liquid-ordered (L_o) lipid systems [15]; leading to lower interfacial hydration (and high GP).

Steady-state fluorescence anisotropy (r) of DPH and Laurdan was measured (Fig. 1 F-G) [27]. While DPH localizes at the deep hydrophobic acyl chain region of lipid membranes, Laurdan localizes at the lipid head group interfacial region. (Fig. 1F). We found no significant difference in anisotropy at pH 7.4 and 5.5 (Fig. 1F) for the empty liposomes. The DPH anisotropy at low pH indicated increased fluidity. The head groups with reduced repulsions among each other affect the hydrophobic chains, and thus the molecules have more freedom within a less tightly packed hydrophobic chain environment. The presence of RIF enhanced the anisotropy, suggesting that a fraction of RIF is also localized in the deep acyl chain regions or that the RIF partitioning at the headgroup modulates the acyl chain dynamics. No appreciable change in pH in the Rif-loaded BL-d liposomes suggests that pH does not affect the hydrophobic interactions between RIF and liposomes (Fig. 1G).

2.4. Immunostimulatory effect of Msm lipid-based liposomes

Host defense systems initiate various strategies for eliminating mycobacteria, such as activating pro-inflammatory responses, producing reactive intermediates such as reactive oxygen and nitrogen species, and inducing cell death to inhibit the spread of infection [29–32]. Conversely, the bacteria have evolved several strategies to disturb these defenses, such as interference with phagosomal maturation and acidification, resistance to oxidative stresses, escape to the cytosol, formation of granulomas, and modulation of host cell death [33]. Along these lines, it has recently been demonstrated that cationic liposomes, when combined with additional immunostimulatory agents such as TDB (trehalose 6,6′-dibehenate), MPL (monophosphoryl lipid A adjuvant), and Poly I:C (Polyinosinic:polycytidylic acid), generate a strong immune response against Mtb antigens bolstering the host cell efforts by stimulating both the humoral and cellular immune responses [34,35]. RUTI, a poly-antigenic liposomal vaccine consisting of detoxified and fragmented Mtb cells, is one of the liposomal therapeutic vaccine candidates. It is being developed to prevent active TB in people with latent TB infections by strengthening their immunity [36,37]. Furthermore, RUTI has been demonstrated to lower the bacillary burden and enhance the survival rate of sick animals in both short- and long-term vaccination studies [38]. This strategy of using Mtb fragments in vaccines has inspired the search for novel mycobacterial-derived molecules (instead of cells) and additional mycobacteria species. One of these is Mycobacterium smegmatis, a non-pathogenic strain, that shares numerous glycolipids with Mtb and generates specific humoral immune responses against Mtb infection when enwrapped in liposomes [39].

This prompted us to examine the inflammatory potential of BL-d liposomes. In presence of BL-d liposomes; secretion of IL-8 and TNF-α was 107-fold and 27.9-fold higher than the basal level (Mo macrophages), respectively. Furthermore, in presence of Msm-infected and BL-d liposome-treated macrophages, a 2.9-fold decrease (TNF-α) and a 1.85-fold increase (IL-8) (Fig. 1H, Fig. S3–4) was observed. This supports the inflammatory effect of BL-d liposomes and consequent polarization of Mo to M1 macrophages. Also, as BL-d liposomes elicited a similar immune response as that seen with bacteria (Fig. 1H and S4), this could afford alike host responses to the liposomes, specifically with respect to their intracellular trafficking and fate, and thus, allow for passive targeting of phagocytosed bacteria. Further, absence of any additive effect on the release of TNF-α and IL-6/8 by BL-d liposomes in bacteria-infected macrophages imply a non-detrimental behaviour of BL-d liposomes. At the same time, secretion of IL-10 (Fig. S4) at high liposomal concentrations suggests that BL-d liposomes also elicits some basal anti-inflammatory response as well. The immune response is attributed to the varied lipid constituents of the BL-d liposomes (Table S1), that consist of possible pathogen-associated molecular patterns (PAMPs) (Table S1) that recognize specific macrophages receptors to mediate the initiation of antigen-specific adaptive immune responses and release of inflammatory cytokines critical for intracellular bacterial eradiation (Fig. 1H, Fig. S3–4). The bacterial lipopolysaccharide (LPS) is a known pro-inflammatory mediator and, hence, was used as a positive control and later used to quantify the additive secretion response of inflammatory interleukins in LPS pre-treated and BL-d liposome-treated macrophages. We observed a similar cytokine release pattern i.e., no marked additive effect (Fig. S4–5). In this investigation, media-treated cells served as negative controls.

2.5. Internationalization of Msm lipid-derived liposomes, drug retention within host macrophages, and intracellular bacterial killing

To investigate the internalization of BL-d liposomes within THP-1 macrophages, the cellular uptake of the labelled liposomal formulations was investigated by fluorescence microscopy and FACS (Fig. 2A-B). THP-1 macrophages were treated with N-Rh-DHPE-labelled drug-loaded or unloaded BL-d liposomes in a time-dependent fashion. Fluorescence microscopy confirmed the cell internalization of the liposomes, indicating that BL-d liposomes are efficiently taken up by macrophages (Fig. 2A). We did not observe any extracellular liposomes over time. All fluorescence was limited to the cell periphery indicating fusion and subsequent incorporation of drug-loaded and unloaded N-Rh-DHPE-labelled BL-d liposomes into the THP-1 cell membrane. The fluorescence was homogeneously distributed on the plasma membrane for the initial few hours. Additionally, some fluorescent patches were observed intracellularly. This could be a manifestation of the distinct rearrangements of the cellular membrane in response to Msm lipid [8,22,40] leading to localized enrichment of the fluorescent lipid probe within some areas of the cells. FACS quantified the internalization of N-Rh-DHPE-labelled BL-d liposomes and free N-Rh-DHPE in THP-1 macrophages. The mean fluorescence intensity (MFI) indicated a 6.4-fold increase in cellular uptake of N-Rh-DHPE-labelled BL-d liposomes, compared with free N-Rh-DHPE (Fig. 2B). The data confirms that N-Rh-DHPE-labelled BL-d liposomes reach the intracellular environment efficiently, supporting an active targeting strategy.

Fig. 2. Cellular uptake and drug retention profile of BL-d liposomes in THP-1 macrophages.

(A) Representative confocal images of free N—Rh DHPE dye and BL-d liposomes loaded with N-Rh-DHPE (25 μg/mL) in a time-dependent manner. Scale bar: 10 μm, 63× oil objective. (B) Mean fluorescence intensity of free N—Rh DHPE BL-d liposomes loaded with N—Rh DHPE (25 μg/mL) using FACS. (C) Area abundance of Rifampicin extracted from THP-1 macrophages treated with free rifampicin (25 μg/mL), and BL-d Rifampicin liposomes (25 μg/mL). The area is calculated using the standard curve of Rifampicin, prepared in methanol with a serial dilution and fitted with a linear regression curve. Data represents the mean ± SD of three independent experiments (A-B), while in (C) data is from two independent experiments. Statistical significance was determined using one-way ANOVA with Tukey’s multiple comparison test.

Later, we monitored the drug retention profile of Rif-loaded BL-d liposomes as compared to the free drug in THP-1 macrophages through UHPLC-mass (Fig. 2C, Figs. S5A–C, S6–7). Liposomal formulations showed a 2-fold higher retention of Rifampicin inside the macrophages than the native drug after 12 h, whereas a 3.5-fold higher retention was observed after 24 h. We next quantified the intracellular Rifampicin levels, and found them to be 1.2 μg/10⁶ cells; in contrast, the free Rifampicin was not detectable after the first 24 h. Thus, Rifampicin-loaded BL-d liposomes have superior cellular uptake and drug retention profile as compared to the free Rifampicin and could provide intracellular drug-mediated eradication of the pathogen. To access the same, the antimicrobial efficacy of Rifampicin in free form and within BL-d liposomes was assessed in THP-1 macrophages after infection with intact Msm bacteria. The antimicrobial activities of Rif-loaded BL-d liposomes were 0.6-fold and 2.1-fold higher than that of the free Rifampicin and empty liposomes, respectively (Fig. 3A, Fig. S8). Rifampicin demonstrated lower antimicrobial activity due to its intracellular-uptake properties in vitro, with an intracellular-to-extracellular (I/E) ratios ranging from 0.1 to >20 [30]. Free drug and Rif-loaded BL-d liposomes demonstrated similar MIC values in the in vitro bacterial killing (Fig. S9). We further correlated and evaluated the time-dependent effects of Rifampicin in bacterial replication suppression for both the free drug and Rif-loaded BL-d liposomes using confocal microscopy. A comparison of mycobacterium suppression profiles for the two revealed a similar trend, as observed in CFU counting (Fig. 3A). In control (untreated infected macrophages) cells, the morphology of Msm was more slender, elongated, and intact. After 4–12 h with Rif-loaded BL-d liposomes, bacterial morphology was distorted (Fig. 3B-D). Their membrane was no longer intact and elongated, suggesting that Rifampicin not only suppresses the growth or replication of bacteria but also disrupts the membrane structure and possibly contributes to bacterial killing. The membrane action of RIF has been recently shown by us [41]. A more pronounced effect was observed after 24 h (Fig. 3D). We attribute the enhanced antimicrobial activity within infected macrophages of Rif-loaded BL-d liposomes due to the immunogenic properties of constituent lipids that pre-activate the infected macrophages and synergize the antimicrobial activity of Rifampicin in eradicating the intracellular bacteria. This was supported by observing inferior antimicrobial activity with rifampicin-encapsulated liposomes constituted from non-immunogenic lipid, lecithin (Fig. S10). Further, the ability of Rif-loaded BL-d liposomes to show higher drug release at low pH compared to lecithin liposomes (Fig. S10) might also contribute to a higher bacterial killing at intercellular sites (e.g., endosomes).

Fig. 3. Concentration-dependent and time-dependent intracellular antimicrobial efficacy of free RIF and Rif-loaded BL-d liposomes in THP-1 macrophages.

(A) The concentration-dependent anti-mycobacterial activity of free drug and Rif-loaded BL-d liposomes (5, 25, and 50 μg/mL) was determined using CFU counting by serial dilution and plate counting method for 24 h. (B—D) Time-dependent representative confocal images of Msm-infected THP-1 macrophages treated with Rifampicin (25 μg/mL) and Rif-loaded BL-d liposomes (25 μg/mL). Data represent the means ± SD of three independent experiments. Statistical Significance was determined using one-way ANOVA with Tukey’s multiple comparison test. Scale bar: 10 μm, 63× oil objective.

Furthermore, a species-specific killing is inferred as Msm lipid-derived drug-loaded liposomes did not display a lower MIC against Mtb HRv37 and Rif-resistant strains compared to the free drug (Table S4). This is attributed to the subtle lipidome differences among various bacterial strains which could fine-tune the immunogenic response synergizing with antibiotic potency and lipid bilayer properties including their responses to pH. This also demonstrates that the generation of drug-encapsulated liposomes with lipids extracted from a specific pathogenic organism could serve as personalized bug-killing platforms.

2.6. Mechanism of cellular uptake and subcellular localization of BL-d liposomes

Surface charges play a crucial role in cellular uptake; positively charged nanoparticles are internalized via macropinocytosis, and clathrin/caveolae-mediated endocytosis is the mechanism for the uptake of negatively charged counterparts [34]. Further, the negative charge enhances the binding of liposomes to macrophages and promotes their phagocytosis. Particle size is another important determinant; <1 μm particles are internalized by clathrin-mediated endocytosis, and with the increase in the size of the particles, clathrin-independent mechanisms take over. Hence, for particles with a diameter of 1 μm or above, internalization occurs by phagocytosis after forming actin-coated invaginations in the macrophage membrane. To further investigate the pathway of endocytosis and mechanism of internalization, we employed various endocytosis blockers such as nystatin, chlorpromazine, azide, colchicine, and phenothiazine that block caveolae-dependent, clathrin-dependent endocytosis, ATP synthesis, microtubule-mediated and clathrin-mediated pathways, respectively [42]. The cells pre-treated with the endocytotic blockers and treated with BL-d liposomes exhibited weak signals, except for colchicine (Fig. S11A-E). As BL-d liposomes are negatively charged and < 0.4 nm, we hypothesize their uptake through clathrin-mediated endocytosis. Aptly, in presence of chlorpromazine, 85% uptake inhibition of BL-d liposomes was observed. NaN3 also resulted in a marked decrease in cellular internalization (by 58%) (Fig. 4A-D, Fig. S11A-E). This indicates that internalization is an energy-dependent process, however complete inhibition was not seen because of exogenous ATP and glucose in the serum-free media. The above findings were also correlated with microscopy, where we found similar results. Therefore, it is implied that the internalization of BL-d liposomes follows multiple endocytic pathways such as clathrin, and caveolae-mediated, but is independent of microtubule-mediated endocytosis.

Fig. 4. Uptake mechanism of N-Rh-DHPE-labelled BL-d liposomes (BL-d Rh Lipo).

(A-C) Representative confocal images of THP-1 cells showing inhibition of the uptake of BL-d Rh liposomes after treatment with indicated endocytosis inhibitors for 1 h. (D) Quantitative estimation of the inhibition of liposomal uptake after additional 2 h incubation with BL-d liposomes. (E-J) Subcellular localization of BL-d Rh liposomes. Representative confocal images of THP-1 macrophages treated with BL-d Rh Lipo (25 μg/mL) and later stained with various endosomal antibodies in a time-dependent manner © EEA-1; 30 min (F) EEA-1; 24 h (G) RAB-7; 30 min (H) RAB-7; 24 h (I) LAMP-1; 30 min and (J) LAMP-1; 24 h. The relative intensity of each condition with Pearson correlation coefficients given at the corner of the RI graph and zoomed inset of each image are highlighted with white borderline and the arrow. (K) Manders correlation M2 coefficients depicting the extent of colocalization between the BL-d Rh Lipo and probed the various endosomal markers at various time points. We have used the weighted colocalization coefficients for analysis. (L) Pearson correlation coefficients of each time point and condition treated with BL-d Rh liposomes (25 μg/mL). Confocal images represent BL-d Rh liposomes in red, endosomal markers in green, nuclei in blue, overlay of the compartment, and colocalized particles in yellow. Data represent the means ± SD of three independent experiments. Statistical significance was determined using two-way ANOVA with Dunnett’s multiple comparison test. Scale bar: 10 μm, 63× oil objective. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

To maximize the therapeutic efficacy of any antimicrobial agent, the delivery vehicles should efficiently deliver the drug to intracellular compartments where the pathogens reside and multiply. Thus, to understand the interactions between the BL-d liposomes and pathogen at the subcellular level, first the endocytic distribution or localization of the N—Rh DHPE-labelled BL-d liposomes (BL-d Rh Lipo) was determined using endosomal markers; EEA-1 (early), Rab-7 (late) endosomes, and LAMP-1 (lysosomes). Colocalization with early (EEA-1) endosomes was observed in earlier time points (30 min – 24 h) (Fig. 4E-L, Fig. S12). Later, the highest colocalization in the late (Rab 7) endosomes was observed. A 0.56 fraction of BL-d liposomes colocalized with early endosomes, 0.89 fractions with late endosomes, and 0.63 fractions with lysosomal compartments (Mander’s coefficients and system’s relative fluorescence intensity, (Fig. 4E-L, Fig. S12). These findings suggest that liposomal formulations reside not only in cytoplasmic area but also endosomal compartments. Collectively, THP-1 macrophages could take up and sustain the BL-D Rif liposomes for long-term therapeutic retention.

Next, we studied the correlation, if any, between the trafficked intracellular sites of BL-d liposomes with subcellular sites of mycobacterial residence. FITC-labelled Msm infected macrophages treated with Rif-loaded and N-Rh-DHPE-labelled BL-d liposomes were immune-stained for EEA-1, Rab 7, and LAMP-1 [43] and relative levels of labelled liposomes and bacteria were determined in each endo-lysosomal compartment. Manual counting with a color-coded contrast rendered the colocalization events of endosomes harboring bacteria and liposomes. BL-d liposomes were distributed mainly in late endosomal (RAB-7) and lysosomal (LAMP-1) compartments (Fig. 5A-E). Surprisingly, the subcellular distribution of Msm was identical to that of BL-d liposomes; a 1.7-fold and 1.5-fold higher association with late endosomes (Rab-7) compared to early endosomes and lysosomes, respectively. A 0.892 fraction of BL-d liposomes colocalized with Msm and 0.92 with late endosomes. These data demonstrate that BL-d liposomes localize not only in the cytoplasm but also with early and late endosomes, wherein, an appreciable co-localization of the latter with ingested bacteria is present. This implies that BL-d liposomes traffic to the same subcellular compartments, where the mycobacterium resides (Fig. 5A-E). This could underline higher bacterial eradication due to sustained and higher release of the drug at this pH found in late endosomes, (Fig. 1 D) and intracellular drug retention profile (Fig. 2C). These findings are essential for developing innovative, smart, and dual-action formulations to eradicate tuberculosis and other intracellular infection.

Fig. 5.

Intracellular trafficking and subcellular distribution of M. smegmatis (Msm), BL-d liposomes, and endosomal markers. Representative confocal images of BL-d liposomes showing subcellular distribution in the cytoplasm and other intracellular or endosomal regions within the THP-1 macrophages such as (A) early endosome, EEA-1 (B) late endosomes, RAB-7 (C) lysosomes, LAMP-1. Liposomes are shown in red, endosomal compartments in magenta, bacteria (Msm) in green, and nuclei in blue, and overlay of the compartment and particle in whitish the compartment and particle in whitish-blue containing relative intensities with Pearson correlation coefficients at the corner side of the graph. (D) No. of colocalized bacteria with liposomes containing endosomes/macrophages. (E) The bar graph represents the colocalization profile through Manders correlation coefficients in between the M. smegmatis and BL-d liposomes vs endosomes, BL-d liposomes vs endosomes, and endosomes vs M. smegmatis. Data represent the means ± SD of three independent experiments. Two-way ANOVA (Sidak multiple comparison test) was performed, with a statistical analysis of p > 0.05, p > 0.005, and p > 0.0005. Scale bar: 10 μm, 63× oil objective. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

3. Conclusions

In summary, liposomes generated with lipids derived from Mycobacterium smegmatis encasing rifampicin were developed. The liposomes exhibited good biocompatibility and long-term stability. Improved drug uptake, release, retention, and antimicrobial efficacy against macrophage-ingested Mycobacterium smegmatis compared to the free drug and non-immunogenic liposomes were seen. The drug release was higher and sustained at lower pH, due to the interplay between pH-induced changes in the drug and lipid environments that fine-tuned the drug-lipid interactions. Drug-liposomes trafficked via the clathrin-mediated processes localized in the late endosomes, and exhibited enhanced killing of the intracellular bacteria. M. smegmatis and the drug liposomes interacted at the subcellular level thereby enhancing the antimicrobial effect. The antimicrobial effect was further magnified by the concomitant activation of host pro-inflammatory responses by the constituent bacteria lipids in the delivery vehicle, which is otherwise critical for bacterial clearance. Drug encapsulated within non-immunogenic lecithin liposomes exhibited attenuated killing of the intracellular bacteria. This work demonstrates a novel pathogen-specific nanocarrier design inspired by the lipid fraction derived from the pathogen, while exhibiting multiple functionalities, i.e., immune modulation and intracellular drug delivery. Collectively, this work demonstrates the potential of pathogen-derived multimodal therapeutic approaches for their intracellular eradication.

4. Materials and methods

4.1. Cell culture, antibodies, and reagents

4.1.1. Reagents

Fetal bovine serum (FBS) was procured from Hy Clone (USA), Roswell Park Memorial Institute medium (RPMI-1640), 1× Dulbecco’s phosphate buffer saline (D-PBS), sodium bicarbonate, 20× Antibiotic-antimycotic solution, amphotericin-B, streptomycin, penicillin G, 0.25% trypsin-EDTA, paraformaldehyde, PMA (phorbol 12-myristate-13-acetate), MTT and solubilizing agent were purchased from Hi Media. Middlebrook 7H9 and 7H10 were procured from BD Difco. The drug Rifampicin and the probes DPH (diphenylhexatriene) and TMA-DPH (trimethylamino-diphenylhexatriene) were purchased from Cayman Chemicals, USA. Chloroform of spectroscopic grade was purchased from Spectrochem. AR grade chloroform, methanol, n-heptane, and ethyl acetate were obtained from Merck. The probe laurdan, FITC, and the other chemicals (Tris, Glutaraldehyde, MgCl2, and methanol of spectroscopic grade) were purchased from Sigma-Aldrich. Antibodies (RAB-7 (D95F2) Rabbit mAb, EEA1(C45B10) Rabbit mAb, and LAMP-1 (D2D11) Rabbit mAb) were purchased from Cell signalling technology. Lecithin was a kind gift from Prof. Rinti Banerjee (IIT Bombay). All the products were used without further purification. The water used for aqueous buffer solutions was from Millipore water purification system.

4.1.2. Mammalian culture

THP-1 cells were a kind gift from Prof. Sarika Mehra (Chemical Engineering, IIT Bombay). THP-1 cells were maintained in RPMI-1640 medium supplemented with 10% heat-inactivated FBS and penicillin (100 U/mL), streptomycin (100 mg/mL), and amphotericin B (20 μg/mL) at 37 °C in humidified air containing 5% CO2. THP-1 monocyte was cultured at a lower density (2.5 × 10⁵/mL) with media being refreshed every three days for one week (as per recommended on ATCC site). For the experiments, THP-1 cells were cultured using 20 nM PMA for 48 h; cells were further incubated in complete RPMI media containing no PMA for 24 h before using it for any experiments. All experiments were performed within the cell passages [18–20]. Cells were routinely tested for Mycoplasma and found free of contamination.

4.1.3. Bacterial culture

Mycobacterium smegmatis (MC²155) was a kind gift from Dr. Tyagi (Delhi University). Mycobacterium smegmatis (MC²155) was grown in Middlebrook 7H9 supplemented with 10% (v/v) home-prepared albumin-dextrose-catalase (ADC), 0.1% tween80 and 0.5% glycerol cultured at 37 °C at 180 rpm shaking conditions. Cells were harvested (the early log phase of 0.8 OD600) at 12000 rpm for 5 min and washed twice with PBS (pH 7.4). Removal of tween-80 from the cells was validated by washing the cells twice with either tween80-free Middlebrook 7H9 media or phosphate buffer saline (pH 7.4). The absorption of bacteria was measured at 600 nm using a UV spectrophotometer (Eppendorf, USA). Bacteria were then harvested at 12000 rpm for 5 min and washed twice with PBS (pH 7.4).

Middlebrook 7H10 (Difco) supplemented with 0.5% Glycerol and 10% OADC (oleic acid, albumin, dextrose, and catalase) agar base was used for CFU counting and to select a single colony for infection and bacterial staining related studies. Regularly bacterial broth was streaked on Luria Bertani Agar Base (Hi media) to cross-check the contamination from other bacterial strains.

4.2. Lipid extraction

Lipids were isolated from the harvested cells by the RMS (Reverse micellar solution) extraction technique (Nikaido et al., 1993). For the inner membrane lipid extraction, the RMS-treated cells were washed twice with distilled water and then extracted with 3 mL of chloroform: methanol: water (2:1:0.1) for every 10 mg of dry mass, and the extracts were dried to obtain respective lipids. The mixture was left overnight, and the extractions were carried out in monophasic solutions four times at 20 min. The lipids were recovered from the supernatants, which were filtered, the solvent was removed and further purified, and extracted lipids were dried and reconstituted in chloroform and stored at –20 °C till further use.

4.3. Analysis of lipids

The thin-layer chromatographic (TLC) technique using aluminium-backed silica gel plates (Silica gel 60 F 254; Merck) was used to obtain the pattern of lipid distribution in each extracted fraction. Lipid fractions were run in the mobile phases of chloroform: methanol (95:5 v/v %) and developed using a 1% anthrone solution.

4.4. Preparation of Rifampicin-loaded bacterial lipid-derived (BL-d) liposomes through thin-film hydration method

Liposomal suspensions of extracted lipids were prepared by the thin-film hydration method. The lipid-Rifampicin film was designed in different ratios (1:5, 1:10, 1:15, and 1:20) to optimize drug encapsulation and in-vitro drug release studies. The dried lipid-Rifampicin (0.5 or 0.1 mg/mL) and lipid-lipid probe (N–Rh DHPE, 2μg/mL) films were hydrated, separately, with 0.22 μm-filtered aqueous phosphate buffer saline (PBS, pH 7.4, 5.5, 4.5, and 6.5), at room temperature followed by 5 mins vortexing and 10 mins ultrasonication in a water bath at 65 °C for full 4 cycles. Further, drug-loaded or lipid probe-loaded liposomes were filtered using ultrafiltration tubes to remove the free Rifampicin or lipid probe from the liposomal suspensions. The pellets were collected and resuspended in PBS (as described above). Later, physicochemical properties (size, charge, lamellarity, drug encapsulation, drug loading, and in vitro-drug release) were measured for further experiments.

4.5. Physiochemical characterization of Rifampicin-loaded BL-d liposomes

4.5.1. Drug encapsulation

The encapsulation efficiency of Rifampicin was calculated by following Eq.(1):

$$\text{Encapsulation}\text{Efficiency}\left(EE\%\right)=\frac{W2-W1}{W_2}\times 100$$ (1)

W1 is the amount of free Rifampicin, and W2 is the total added Rifampicin. The concentration of encapsulated Rifampicin was determined through a UV–Visible Elisa plate reader (Thermo scientific) and UHPLC-MS/MS (Thermo scientific). The addition of 100% methanol disrupted the liposome to release all the encapsulated drugs. Methanol-containing liposomal solutions were diluted to 20% in PBS and the EE% was recorded using an Elisa plate reader (Thermo Scientific, USA). Three independent experiments were performed to optimize the final ratio of EE.

4.5.2. Particle Size, shape, and charge: DLS, SEM, LSCM, TEM and stability studies

The particle size, polydispersity index, and zeta potential experiments were performed at RT (Anthon Paar, Instruments, UK), at a scattering angle of 175°. Freeze-dried liposomes (drug-free and drug-loaded) were resuspended in a filtered PBS solution (pH 7.4) before dynamic light scattering (DLS) experiments. The sample was loaded in either an ordinary or capillary (Omega) cuvette to determine particle size or zeta potential. The particle shape analysis was also done by microscopic observation using a Scanning Electron Microscope, SEM (Carl Zeiss, USA), and Laser scanning confocal microscopy (LSCM, Carl Zeiss, USA). SEM analysis enabled visualization of cell surface interactions of liposomes with the THP-1 cells. An aliquot of liposome suspension was dropped-cast on an aluminium foil and dried in a closed vacuum chamber for 10 to 15 min. Later, aluminium foil was affixed to a carbon tape, and platinum-coating was done before conducting the liposomal formulations’ surface analysis. Fluorescent liposomes were prepared using N—Rh DHPE (lipid probe) to examine their shape and cellular uptake. Three independent experiments were performed. For cryo-TEM: a carbon-coated holey grid was used on which 5 mL of the bacterial lipid was deposited. The excess fluid was taken care of with a Whatmann filter paper, after which a Vitrobot (FEI, Hillsboro, OR) with a 1 s blot time was used for plunging in liquid ethane. At a temperature of –173.45 °C, the sample was observed using a Gatan:Orius charge-coupled device camera under a high-resolution Cryo-TEM (JOEL-JEM 2100 electron microscope). ImageJ software was used for further analysis.

4.5.3. In-vitro drug release kinetics

The bacterial lipids-derived liposomes loaded with Rifampicin showed the highest EE at a 1:10 ratio (Table S3) and were used for drug release studies. Freshly, ultrafiltered liposomal pellets were prepared and dispersed in 0.5 mL PBS at different pH (7.4, 6.5, 5.5, and 4.5). The resulting suspensions were transferred to a dialysis bag (DM-50, Himedia with MWCO (molecular weight cut-off) between 12 and 14 kDa that was placed in a falcon tube containing 10 mL of sink medium (PBS) at different pH (7.4, 6.5, 5.5, and 4.5), respectively. The sink medium was maintained at 37 °C with stirring at 120 rpm. 100 μL of the sink solution was withdrawn at specific intervals (0.5, 1, 1.5, 2, 3, 4, 5, 7, 9, 12 h, and so on), and the release was monitored for up to 3 days. The aliquoted sample was replaced by the same volume of the fresh buffer to maintain sink volume. Three independent experiments were performed. Quantitation was done by measuring calibration curves. Percentage of encapsulation efficiency measured through the supernatant was used to calculate the total amount of drug in the dialysis bag. The RIF content at each time point was calculated as % cumulative drug release, after considering the amount of drug removed in the aliquoted volumes.

$$\%Cumulativedrugrelease\frac{Cumulativeamountofdrugreleased}{Totalamountofdruginthedialysisbag}\times 100$$ (2)

4.5.4. Membrane fluidity and anisotropy

4.5.4.1. Laurdan generalized polarization (GP) spectroscopy

The fluorescence intensities of laurdan incorporated BLd liposomes loaded with and without rifampicin was recorded using temperature-controlled Varian Cary Eclipse Fluorescence Spectrophotometer, at physiological temperature of 37 °C. The use of the generic formula of.

$$GP=\frac{I_{440}-I_{490}}{I_{440}+I_{490}}$$ (3)

(Where I indicated the intensity at specified wavelengths).

4.5.4.2. Membrane anisotropy studies

The changes in the local environment of the fluorophores within a membrane system can be marked by monitoring the orientation and rotational correlation time which would be indicated by the fluorescence anisotropy technique. This would indirectly indicate the micro-viscosity of the membrane. Fluorescence anisotropy of DPH/TMA-DPH labelled BL-d liposomes loaded with and without rifampicin were measured using a temperature-controlled Varian Cary Eclipse at physiological temperature of 37 °C. Fluorescence Spectrophotometer attached with a polarizer (Varian Cary Eclipse Manual Polarizer).The samples were excited with vertically and horizontally polarized lights and respective polarized emission intensities were recorded. The degree of fluorescence steady-state anisotropy (r) was calculated from the following equation.

$$r=I_{VV}-G{I}_{VH}{I}_{VV}+2G{I}_{VH}$$ (4)

where I_VV and I_VH are the parallel and perpendicular emission intensities of the vertically polarized excitation beams. G = I_HV I_HH (Eq. 5) is the correction factor to determine the sensitivity of the instrument (G should be ~1). I_HV and I_HH being the parallel and perpendicular emission intensities of the horizontally polarized excitation beams.

4.5.5. Biological evaluation

4.5.5.1. Cellular uptake

dTHP1 macrophages were treated with 2 μg/mL of free membrane probe (N—Rh DHPE) and 25 and 50 μg/mL of BL-d liposomes loaded with (2 μg/mL) N—Rh DHPE in a time-dependent manner (2 h, 4 h, 12 h, 24 h, and 48 h, respectively). The extracellular dye and dye-loaded liposomes were washed off with PBS. Cell lysates were pelleted out and resuspended in filtered PBS for FACS measurements. For microscopy, the cells were washed twice with PBS and fixed with 4% paraformaldehyde for 10 min at RT. The fixed cells were washed with PBS, and then a Hoechst dye 33342 (0.5 μg/mL) (Invitrogen, Paisley, UK) was added, followed by 10 min incubation at RT. Additionally, cells were washed with PBS and mounted on microscopic slides. Using a 60× oil immersion objective lens, the liposomal internalization was observed under a confocal laser scanning microscopy (CLSM, Carl Zeiss, Germany). ZEN Lite 2.6 software was used for image processing.

Furthermore, the quantification of liposomal internalization was analyzed by a flow cytometer FACS (BD Biosciences, USA). Three independent experiments were performed to quantify the maximum amount of internalization inside the cells by measuring the fluorescence intensity (a.u) using Flow Jo software (BD biosciences) and normalizing with appropriate controls.

4.5.5.2. Intracellular content of Rifampicin

dTHP-1 macrophages were plated in a 60 mm dish (Eppendorf) at a density of 1 × 10⁶ cell/mL. The cells were then exposed to BL-d liposomes loaded with Rifampicin (25 μg/mL) or free Rifampicin (25 μg/mL) for various time intervals (2 h, 4 h, 12 h, 24 h, and 48 h) and incubated at 37 °C under an atmosphere of 5% CO₂. The Mϕ (macrophages) cells were resuspended in 0.1% SDS lysis buffer at a density of 1 × 10⁶ cell/mL and homogenized by sonication for 5 min. After that, Rifampicin was released into the homogenized solution from the cells and was dissolved or separated using organic solvent by adding equal volumes of chloroform and methanol/10⁶ Mϕ cells and kept for 4 to 6 h. After 5 h of incubation, drug-containing solutions were separated into three phases (Chloroform, methanol, and aqueous). The chloroform and methanol phase were evaporated using rotavap under reduced pressure, whereas the aqueous phase was lyophilized. Afterward, the dried samples of three separate phases were resuspended in 250 mL of methanol. They were pooled together in a 1.5 ml glass vial, and filtered with a 0.22 μm microporous membrane. Intracellular content of Rifampicin was measured by UHPLC-Q Exactive MS (Thermo Scientific) using a reverse-phase C18-ODS column (STRODS-M 150 mm × 4.6 mm, Shinwa Chemical Industries, Ltd., Kyoto) with a UV detector at 254 nm. 5 μL samples were loaded into the system with the flow rate of 0.3 mL/min. The mobile phase was acetonitrile, and water containing 0.1% (v/v) formic acid (40:60 (v/v)) was used. Two independent experiments were performed to quantify the intracellular drug content by interpolation of calibration curves measured using freshly prepared rifampicin solution (calculating the area under the curve of each specified condition).

4.5.5.3. Inhibition of the cellular uptake of Rifampicin-loaded BL-d liposomes

dTHP-1 cells were seeded in 60 mm and 24-well plates. After maintaining the cells in preconditioned media for 24 h, cells were washed with PBS and treated in separate wells with Sodium azide (1 h, 20 μg/mL), Nystatin (1 h, 20 μg/mL), Chlorpromazine (1 h, 20 μg/mL), Colchicine (1 h, 20 μg/mL), and Phenothiazine (1 h, 20 μg/mL). Following treatment with the inhibitors, N—Rh DHPE (2 μg/mL) loaded BL-d liposomes (25 μg/mL) was added to the cells and incubated for an additional 2 h in the presence of the inhibitors. After incubation with liposomes at 37 °C under an atmosphere of 5% CO₂, cells were harvested and processed as per discussed in earlier section. The incubation time and concentrations of the inhibitors used were carefully titrated, and the concentrations used are consistent with those reported to be non-toxic. Additional confirmation of cell viability was observed by microscopy after treatment with various inhibitors, followed by incubation with dye-loaded liposomes. The fluorescence intensity was recorded using FACS and normalized with control cells (N-RH-DHPE loaded BL-d liposomes without inhibitors), and the percentage was calculated using FlowJo software. Two independent biological replicates were done.

4.5.6. Antimicrobial efficacy of Rifampicin-loaded BL-d liposomes

4.5.6.1. Generation of FITC-labelled Msm bacteria

Bacterial cells were harvested by centrifugation at 10,000 ×g and 4 °C for 5 min and then resuspended in 1 ml of 0.1 M sodium bicarbonate buffer. The optical density was measured at 600 nm from a 1:100 dilution of cell suspension in sodium bicarbonate. After that, cells were diluted to 10¹⁰ cells/mL in 1 mL sodium bicarb. 0.2 μL of FITC stock solution (10 mg/mL) was added to the cell suspension and immediately vortexed. Later, cells were incubated in the dark with end-over-end rotation for 45 min at RT. Cells were washed with HBSS (4×) to remove the unbound dye and resuspended in 1 mL of HBSS. Afterward, the optical density was again measured at 600 nm from a 1:100 dilution, and cells were resuspended and adjusted to 10⁷ cells/mL concentration in RPMI medium for infection-based studies. 1:10 MOI (multiplicity of infection) was selected for bacterial infection.

MIC of Rifampicin and Rifampicin-loaded BL-d liposomes in Msm and virulent Mycobacterial strains.

A Resazurin assay was performed to calculate the MIC (minimum inhibitory concentrations) of Rifampicin on Msm (MC²155 cells). The bacteria were cultured in MB7H9 medium supplemented with ADC/tween80 and harvested in the log phase of growth 0.8 OD. In a 96-well plate, 100 μL cell culture in triplicate were diluted appropriately to a seeding density of 10⁵ cells, were aliquoted in triplicates. Cells were exposed to different concentrations of free Rifampicin, free BL-d liposomes, and Rifampicin-loaded BL-d liposomes, mixed thoroughly and incubated for 24 h at 37 °C in the bacterial incubator.

Resazurin stock solution was prepared aseptically (0.2% w/v) in methanol and added in an amount equal to 10 % of the volume in the well, both treated and untreated wells, and incubated for 4–8 h at 37 °C. Fluorescence was recorded at 560 nm excitation wavelength and 590 nm as emission wavelength using fluorescence spectrophotometer. Blank was used as a medium only. MIC was calculated using the given eq. (3):

$$Percentaged\text{i}fferencebetweentreatedandcontrol=\frac{FI590oftreatedcells}{FI590ofuntreatedcells}x100$$ (3)

For the MIC determination against pathogenic strains, culture grown and assay performed are as described above. However, plates were incubated for 7 days (M. tuberculosis) or 72 h (for M. abscessus, M. chelonae, and M. fortuitum). Afterward, Resazurin was added, as mentioned above, and incubated overnight. The reduction of resazurin dye, as observed by a change in color from blue to pink, is considered bacterial growth. Whereas MIC is the lowest concentration that completely inhibits growth (no change in color).

4.5.6.2. Biocompatibility on macrophages

MTT (3–[44]-2,5 diphenyl tetrazolium bromide) was used to detect the viability of macrophages. The macrophages were grown in 96-well plates and then stimulated with different concentrations of free Rifampicin and Rif-loaded BL-d liposomes for 24 h. After discarding the supernatant, to each well 10 % MTT was added and incubated for 4 h in the dark. After this, the MTT solution was removed, and then the formazan crystals were dissolved using 100% DMSO solution and incubated for 15 min in the dark. Following that, absorbance was recorded at 570 nm, and background subtraction was measured at 630 nm using an Elisa microplate reader. Biocompatibility was measured using the Eq. (4):

$$\%cellviability=\frac{Treated-Blank}{Control-Blank}x100$$ (4)

Small pieces of coverslips were sterilized with 70 % ethanol followed by UV treatment for 30 min. Using sterile forceps, coverslips were transferred into a 12-well plate. After that, cells were seeded at a density of 1 × 10⁴ cells per well. 72 h post-seeding, cells were washed twice in pre-warmed (37 °C) 1 × PBS. Later, cells were treated with BL-d liposomes for 1 h and immediately fixed by gently adding 2.5 % glutaraldehyde diluted in PBS (37 °C) for 30 min. Cells were immediately subjected to alcohol gradient dehydration using 25%, 40%, 60%, 80%, 90%, and 100% ethanol for 15 min at each concentration. To avoid constant handling of the coverslips, they were placed and dried in a vacuum desiccator and incubated for 30 min. A total of 2 coverslips were mounted onto the microscope stage at a time using carbon tape, sputter-coated with platinum, and finally viewed using a (Hitachi S –3400 N) scanning electron microscope for any morphological changes induced by the BL-d liposomes. No such changes were observed, and this experiment was performed once.

4.5.6.3. Intracellular antimycobacterial effect of Rifampicin and Rifampicin-loaded BL-d liposomes

dTHP-1 cells (5 × 10⁶ cells/mL) were infected for 3 h with bacterial cells (5 × 10⁷ cells/mL) at 1:10 multiplicity of infection (MOI) in the serum-free medium without gentamicin at 37 °C with 5% CO₂. Afterward, the extracellular mycobacteria were killed by 1 h of incubation with 250 μg/mL of Gentamicin (Sigma-Aldrich). Cells were washed twice using ice-cold PBS and then treated for 6 h, 12 h, and 24 h with empty liposomes, Rif-loaded BL-d liposomes, and free Rifampicin at different concentrations (5 μg/mL, 25 μg/mL, and 50 μg/mL). Finally, intracellular bacterial growth was assessed by Colony-forming unit (CFU) assay. For this, the cells were lysed with 1% sodium deoxycholate (Sigma-Aldrich), and serially diluted in sterile-filtered ultrapure water (10², 10³, and 10⁴) and immediately vortexed. 100 mL of bacilli containing suspension were plated on 7H10 base supplemented with OADC and incubated for 3 to 4 days to monitor the antimycobacterial growth suppression. The number of CFU were counted and calculated using the given equation:

$$CFU/mL=\frac{no.ofcoloniesxdilutionfactor)}{thevolumeofcultureplate}$$ (5)

4.5.6.4. Confocal microscopy

The intracellular microbicidal effect of Rifampicin was also observed by confocal microscopy, and the infection (1:10) and treatment were performed as described above. In this case, FITC-labelled bacteria were used for infection, which was later adjusted, and diluted to 10⁵ cells/mL. After treatment with free Rifampicin and Rifampicin-loaded BL-D liposomes for 2 h, 4 h, 12 h, and 24 h. The cells were washed with 1× PBS and fixed with 4% paraformaldehyde for 15 min and again washed the cell with 1× PBS at 5 min intervals for 10 min. Further, counterstain was added (Hoesct dye) and incubated for an additional 10 min, followed by washing. Cover slips were mounted on microscopic slides using homemade mounting media and sealed with sealants. Counting was performed manually and normalized with an untreated number of infected macrophages. All experiments were done in independent triplicates. A minimum of 50 cells from randomly selected fields were scored per condition per experiment.

4.5.6.5. Macrophage immune response with bacterial lipids derived (BL-d) liposomes

dTHP-1 cells were plated in a 6-well plate with 1 ×10 cells/mL. After removing PMA, cells were cultured overnight in pre-conditioned media for 24 h and then stimulated with LPS or bacteria (Msm) for 6 h and 3 h, respectively. After stimulation, LPS and bacteria were removed and washed twice with ice-cold PBS, and Gentamicin treatment removed the extracellular bacteria. The cells were further washed with 1× PBS and treated with different concentrations (2 μg/mL, 5 μg/mL, 25 μg/mL, and 50 μg/mL) of BL-d liposomes for another 24 h. For cytokine assay, supernatant was collected from each condition, and cytokines concentration in pg/mL was measured with human IL-8, IL-6, TNF-α, and IL-10 ELISA kits (DUO SET ELISA kit, R&D Systems) according to the manufacturer’s instructions. An ELISA plate reader recorded the absorbance at 450 and 570 nm. Cytokine standard concentrations were selected based on the manufacturer’s instructions and fitted with four parametric-logistic regression models in GraphPad prism software, to calculate the unknown concentrations in each condition by interpolating the calibration curve of each cytokine. Two independent experiments were performed.

4.5.6.6. Sub-cellular distribution and localization

dTHP-1 cells were cultured on coverslips in 24 well plates as described above. The cells were treated with 2 μg/mL of N—Rh DHPE-loaded BL-d liposomes for 2 h, 4 h, 12 h, and 24 h at 37 °C in 5% CO₂. Following the incubation, cells were washed with PBS three times. Further, cells were fixed with 4 % paraformaldehyde solution for 20 min at RT, followed by washing with 1×PBS at 5 min intervals, 3 times. After that, cells were permeabilized using 0.1% triton x-100 for 10 min at RT, washed with 1× PBS 2 times at 5 min intervals. Blocking solutions were prepared using 2 % BSA in PBST, added to the samples in a required volume, and incubated for 1 h at RT. After blocking the unwanted sites, cells were washed and incubated overnight with three primary antibodies (for endosomal markers: RAB-7, EEA-1, LAMP-1), prepared, and diluted in a blocking solution at a 1:300 ratio. The cells were then washed with 1× PBS 5 times at three-minute intervals and incubated with the secondary antibody conjugated to Alexa-flour 488 for 2 h at RT, washed for 5 x using PBS. DAPI was added to stain the nuclei and incubated for 5 min at RT and gently washed with PBS. Subsequently, coverslips were mounted on microscopic slides using a homemade mounting medium. Intracellular colocalization analysis was performed using the Co-loc. JaCop plugin in Image J. Manders correlation coefficients were determined to understand the subcellular localization of N—Rh DHPE loaded BL-d liposomes in THP-1 macrophages.

4.5.6.7. Intracellular trafficking of N—Rh DHPE loaded BL-d liposomes in the acidic endosomal compartment of host macrophages

Briefly, macrophages were infected at 1:10 MOI using FITC labelled bacteria, washed, and treated with freshly prepared 25 μg/mL of N—Rh DHPE loaded BL-d liposomes, and incubated for 12 h at 37 °C with 5% CO₂. Rest, all other experimental conditions were performed similarly as given above. Quantification was done using Image J, and manually endosomes positive cells were counted, containing both mycobacteria and N—Rh DHPE loaded BL-d liposomes (represents bluish-white stain in color) that suggests the perfect colocalization of endosomes, mycobacteria, and dye loaded liposomes. Manders correlation coefficient was calculated using Image J software. Images of both experiments were taken on a laser scanning confocal microscope (Carl Zeiss) using a 63×/1.4NA objective. A minimum of 50 cells were counted and analyzed for each sample. Statistical analysis was performed using the Two-way Anova test. Significant differences: *p < 0.01; ** p < 0.005; *** p < 0.0001.

4.6. Statistical analysis

The results of independent experiments are presented as mean or median values; error bars represent the standard error of the mean (SEM) or standard deviation (SD) as indicated. Statistical significance was tested using an unpaired two-way or one-way ANOVA with Tukey’s/Dunnett’s multiple comparison test. Significant differences: *p < 0.01; ** p < 0.005; *** p < 0.0001.

Supplementary Material

Appendix A. Supplementary data

Supplementary data to this article can be found online at https://doi.org/10.1016/j.jconrel.2023.07.013.

Data availability

Data will be made available on request.