Characteristics of dye tagged biomimetic nano ghost vesicles are evaluated through microscopic and spectroscopic measurements which are further corroborated by in vitro and in vivo studies (Fig. 2A-E). Dye integrated (J-aggregates) ghost nanovesicles (~ 60 nm) show NIR (650–900 nm) absorption and emission which have been tested for solid tumor imaging and localized biodistribution. Engineered optically active biomimetic imaging agents41 are stable for several days (tested up to 30 days, see supporting information for more details) showing promising brightness (779 MESF for 784 ICG/BNV and 2.69 × 1011 BNVs/cm3, Fig. 3 and Table 1) for localized tumor visualization. Moreover, phototransduction response of designed biomimetic optically active ghost nanovesicles (J-aggregates and monomers of ICG dye entrapped nanovesicles) have been evaluated due to their NIR absorption ability. Upon 5 minutes of NIR light exposure, J-aggregates of ICG dye demonstrate better phototransduction (57°C at 2.5×1013 nanoparticles/cm3) compared to ICG monomers (44°C at 2.5×1013 nanoparticles/cm3). Whereas ICG monomers tagged biomimetic nanovesicles exhibit better photosensitizing property as compared to J-aggregates encapsulated nanovesicles (details are discussed below). On the other hand, we have noticed that ICG J-aggregates encapsulated nanovesicles exhibit better brightness (754 MESF units) than their monomer (598 MESF units) may be due to self-quenching of free monomers in nanovesicles. Additionally, it is believed that monomers of small organic dye molecules can diffuse easily from the interior of nanovesicles as compared to their aggregated form. Hence, they may exhibit poor emission property which may not be suitable for bioimaging applications. It should be noted that nanovesicles have lipid assemblies in their membrane. Hence, lipid specific dye remains in the membrane for a long time (at least 12–24 h) due to lipophilic nature of small organic molecules. In the present work, we have used Nile Red, which is specific to the lipid layers with BNVs, but they require high numbers of dye molecules per particle (802) and a number of BNVs per cm3 (7.90 × 1012) to exhibit comparable brightness (748 MESF units) to the ICG J-aggregates in BNVs due to molar extinction coefficient (ε, 38000 cm-1 M-1) barriers. Cancer cell membrane ghost vesicles42,43 (3–5 mg/mL) are prepared from breast cancer cells (5×105-7×105 cancer cells/experiments) in hypotonic conditions treated with various buffers (ice-cold Tris-Magnesium and PBS) followed by freeze-thaw and centrifugal process (Supplementary Fig. 1A). These ghost vesicles are used to prepare nanosized biomimetic cell membrane vesicles under multiple rounds of probe sonication (20 cycles) in ice bath. Uniform size distribution (~ 60 nm, Supplementary Fig. 1B) of prepared biomimetic cell membrane vesicles has been examined by cryo-mode TEM, which is corroborated by dynamic light scattering measurements (0.28 PDI) (Supplementary Fig. 1C). Diameter sharpness in light scattering measurements indicates nanovesicles' spherical morphology, which is clearly ensured by microscopic imaging. On the other hand, the negative surface charge (-12 mV) of nanovesicles supports their colloidal stability in physiological conditions, which may be due to the presence of natural surface biomarkers (Supplementary Fig. 1D). These available inherent surface biomarkers, mostly proteins, help in (i) better particle dispersion in the physiological environment observed from microscopic images shown in Supplementary Fig. 1E, (ii) natural targeting, and (iii) evading macrophage uptake (data are not shown here). Interestingly, controlled particle diameter (~ 68 nm) in the aqueous medium indicated the better dispersion of nanovesicles over 30 days of storage (Supplementary Fig. 1E and Supplementary Fig. 2A).
More importantly, we are trying to address major hurdles such as (i) are these cells derived nanovesicles exhibit homogeneous size distribution and (ii) what is the fate of repeatability and scalability of these biomimetic nanovesicles? Secondly, we also have tried to understand the numbers of encapsulated dye molecules which decides the brightness of particles. To address these critical challenges, we have optimized the preparation recipe and made sure the reproducibility for more than 10 times. However, we have noticed slight changes in the final product concentration (1.20 to 1.32 mg/mL), particles size (60 to 65 nm) distribution and numbers of dye molecules per nanovesicle (201 to 281) during each batch (see Supplementary Table 1 and Table 2 in the supporting information). One the other hand, batch scalability is also a major challenge especially for liposomes and cell derived biomimetic nanovesicles which has been specifically taken care while optimizing the present recipe. We have noticed that concentration of magnesium chloride and sucrose in Tris buffer and Freeze-Thaw time are essential components apart from mechanical forces which decide the cell ghost structures and surface biomarkers. Results of 10 different batches demonstrate the consistency in final product concentration (1.20 to 1.32 mg/mL concentration per 15 mL batch). It should be noted that the biomimetic ghost nanovesicles have been achieved with 5 mg/mL maximum yield but that was for large scale production (100 mL volume of cell suspension) where 65 nm size of nanovesicle can encapsulate 215 ICG dye molecules in its cavity.
$$Weightof 1 BNV \left(mg\right)=\rho \times 4/3\pi r^3 \left(1\right)$$
$$NumberofBNVs/mL=C/(\rho \times 4/3\pi r^3) \left(2\right)$$
$$NumberofDyes/mL=A/ \epsilon \times 6.023\times 10^23 \left(3\right)$$
So far, several studies have been reported on cell membrane-derived biomimetic nanovesicles and their integration with fluorescent dye molecules.14 However, ultra brightness and the importance of integrated dye for the structural characteristic of membrane nanovesicles have yet to be achieved. In the current study, we show the amphiphilic dye molecules (ICG and Nile Red) integrated biomimetic ghost nanovesicles at the nanoscale for targeted tumor imaging and specific biodistribution. In microscopic
images, observed dark patches from the (i) cavity (505 to 784 ICG molecules/nanovesicle), (ii) exterior surface (578 to 828 ICG molecules/nanovesicle), and (iii) bilayer of single nanovesicle represent the assembly of amphiphilic dye molecules (ICG and Nile Red) in the form of J-aggregates (Fig. 2A). Details of nanovesicle numbers, dye per nanovesicles and respective brightness are thoroughly discussed in the supporting information (Supplementary Table 1). These observations are further corroborated with spectroscopic measurements, as shown in Fig. 2B, C. In absorption spectra, aqueous ICG monomers (0.01 mM) exhibit λmax absorbance at 780 nm and a weaker H-aggregate peak at 710 nm. Aqueous dye at high temperature (60°C) incubation result in miniature J-aggregates showing bathochromic shifts in absorption and emission (λabs/emm 890/804 nm) with peak sharpness.
Further, these J-aggregates exhibit significant bathochromic shifts in absorption and emission (λabs/emm 893/806 nm) upon their integration with biomimetic cell ghost nanovesicles which are purified through dialysis (for 2 days using 12 KD dialysis membrane). Photo stability of dye tagged nanovesicles have been evaluated by measuring their emission properties at various time points (1 h to 30 days) with and without treating in serum (Supplementary Fig. 2B, C). It has been calculated that a single membrane ghost nanovesicles accommodate a few hundred dye molecules exhibiting ultra brightness (498 to 778 MESF units viz., Molecules of Equivalent
Soluble Fluorophore). Controlled emission and maintained brightness for a prolonged period (tested up to 30 days) demonstrate the photostability of emissive ghost nanovesicles, whereas free dye J-aggregates demonstrate ~ 40% degradation and reduced emission (Supplementary Fig. 2A-C). Overall, the formed J-aggregates demonstrate better fluorescence intensity and stability under 760 nm excitation when they are within the nanoparticulate formulation.
It should be noted that free ICG in aqueous solution mainly exhibit their monomers with two absorption peaks at 780 and 712 nm, respectively whereas these monomers turn into aggregated form (J aggregates, a sharp extinction peak at 890 nm) due to noncovalent π–π stacking and hydrophobic interactions upon 60°C incubation for 2 h. Further, these ICG J-aggregates integrated ghost nanovesicles exhibit hyperthermia temperature (43°C at 2.5×1013 nanoparticles/cm3) within 2 minutes which is further improved to 57°C after 5 minutes of NIR exposure compared to ICG monomers encapsulated nanovesicles (44.6°C at 2.5×1013 nanoparticles/cm3) as shown in Supplementary Fig. 2D. Whereas ICG monomers tagged biomimetic ghost nanovesicles exhibit better photosensitizing property (80–85%) as compared to J-aggregates encapsulated nanovesicles (20%, see Supplementary Fig. 2E) which indicate their photodynamic therapeutics applicability (data not shown here).
Next, microscopic characteristics reveal that amphiphilic ICG J-aggregates integrated nanovesicles exhibit the exterior dark layer (5–6 nm) around due to the presence of aggregated dye molecules within the available surface biomarkers of nanovesicles which is corroborated by spectroscopic measurements (λmax, absorbance at 892–893 nm and λmax, emission at 806 nm) as shown in Fig. 2B, C. Similarly, the presence of lipid bilayers (2–3 nm) within biomimetic nanovesicles is physiochemically characterized. J-aggregates of lipophilic Nile Red dye in nanovesicles show bathochromic shifts (34 nm red shift) with λmax, absorbance at 604 nm, and emission at 662 nm as compared to their monomers (λmax, absorbance at 570 nm, Supplementary Fig. 3) which is also supported by microscopic characterization. The presence of amphiphilic J-aggregates in nanovesicles is validated through release kinetic patterns in cancer mimicked conditions viz., acidic conditions (Fig. 2D). Dye from the exterior surface of vesicles show rapid release as compared to J-aggregates inside the vesicles (Supplementary Fig. 4 and Supplementary Fig. 5). These nanovesicles demonstrate better cargo release (more than 10%) in late endosomal (pH 2–4) and tumor mimicked environment (pH 6.5) as compared to physiological environment (less than 10% at pH 7.4). Further, the red patches from cancer cells (MDA-MB231 and 4T1) interior indicate the presence of red emissive dye aggregates (ICG and Nile Red) within the BNVs located at different nanovesicle sites. Inherent surface biomarkers of nanovesicles exhibit strong binding with breast cancer cells (MDA-MB231), indicating their natural targeting ability shown in Fig. 2E, Fig. 3, and Supplementary Fig. 6. Empty ghost nanovesicles show distinct cavities as compared to J-aggregates integrated nanovesicles confirmed through microscopic characterization (Fig. 2A, Supplementary Fig. 1B, and Supplementary Fig. 7). These characteristics are discussed here for the first time in the case of cell membrane-derived biomimetic imaging agents.
$$\text{R}\text{e}\text{l}\text{a}\text{t}\text{i}\text{v}\text{e} \text{b}\text{r}\text{i}\text{g}\text{h}\text{t}\text{n}\text{e}\text{s}\text{s}=\frac{\frac{\text{F}\text{L}\text{B}\text{N}\text{V}}{C\text{B}\text{N}\text{V}}}{\frac{\text{F}\text{L}\text{D}\text{y}\text{e}}{C\text{D}\text{y}\text{e}}} \left(4\right)$$
Further, these bright nanovesicles have also been tested for targeted cancer cell imaging (4T1 metastatic breast cancer cells), as shown in Fig. 3A. After 3 h of post-treatment, red emissive dye-tagged nanovesicles are accumulated in cancer cells. It is observed that the brightness of dye-tagged nanovesicles is dependent on the number of dyes per nanovesicle, which define as a multiplication of the quantum yields (ΦF) and the extinction coefficient ε(λ) (M− 1 cm− 1 units) (Fig. 3B and equations 1–4). In the present work, the brightness of nanoparticles is calculated by considering the 2.621×105 M− 1 cm− 1 extinction coefficient at various concentrations of nanoparticles (0.9 to 2.5 mg/mL, Supplementary Table 1). The prepared ultrabright nanovesicles are tested for in vitro and in vivo biocompatibility. In vitro compatibility, BNVs before and after different cargo (J-aggregates of ICG and Nile Red) loading show more than 95% cell viability (200 µg/mL) of healthy cells which is recorded 80–85% at 500 µg/mL (Fig. 3C). It should be noted that engineered BNVs demonstrate multifunctional abilities due to entrapped different cargo molecules (ICG, and Nile Red), which are examined through physicochemical studies and release kinetics (negligible release is seen at neutral pH 7.4), indicating safe entrapment of loaded cargo molecules (Fig. 2, Supplementary Fig. 8 and Supplementary Fig. 9). Brightness and contrast response of J-aggregates (ICG and Nile Red) tagged BNVs have been evaluated in tumor-bearing female mice at different concentrations (0.1-1 mg/mL).
Remarkably, ICG J-aggregates-tagged BNVs demonstrate 20-fold brightness compared to only ICG J-aggregates in the aqueous medium as shown in Supplementary Fig. 10. To evaluate the site-selective tumor imaging and specific biodistribution, a single dose (2.5×1013 nanoparticles/cm3) of J-aggregates (ICG and Nile Red) tagged BNVs are intravenously administrated to 4T1 tumor-bearing female Balb/c mice. After post-injection, whole-body scans are captured through in vivo imaging system at different time points (0.5 h to 24 h), indicating a clear visualization of the tumor site with better brightness signal intensity (Fig. 4A-C). In initial post-injection time (0.5 h to 1 h) the injected ultrabright BNVs contrast agents adjust their distribution (0.98×108 to 0.23×108 p/sec/cm2/sr) among vital organs (heart, liver, spleen) and increase in the tumor
site (0.065×108 to 1.98×108 p/sec/cm2/sr) from 3 h of post-injection time which is further constant (2.2×108 to 2.6×108 p/sec/cm2/sr) up to 24 h of post-injection (Fig. 4A-C and Supplementary Fig. 11 and Supplementary Fig. 12). On the other hand, it has been noted that ICG J-aggregates-BNVs (exterior) exhibit higher signal intensity and brightness (2.4×108 p/sec/cm2/sr) from the tumor site as compared to ICG J-aggregates-BNVs (interior) (2.0×108 p/sec/cm2/sr) indicating the quick release response of emissive dye from exterior surface of BNVs in the cancer mimicked environment as shown in Fig. 4A. Interestingly, we have noticed that injected ultrabright BNVs contrast agents identify the
metastasis in tumor bearing mice. Moreover, Nile Red J-aggregates-BNVs also demonstrate promising signal intensity (1.4×108 p/sec/cm2/sr) from the tumor site which augment (1.6×108 to 2.6×108 p/sec/cm2/sr) with post-injection time (3 h to 24 h). Controlled and uniform fluorescence from the tumor area indicates the better accumulation and distribution of injected red emissive biomimetic nanovesicles than the pre-injected mice (Supplementary Fig. 12 and Supplementary Fig. 13). Biodistribution measurements have been evaluated by considering quantitative ROI values of the fluorescence intensity of the corresponding organs and tumors in vivo (Fig. 4B, C). Further, the promising brightness and fluorescence from the whole body demonstrate the better circulation of nanovesicles, which are entering and stay within solid tumors and exiting via draining tumor lymphatic and then re-entering in the blood circulation as shown in Fig. 4. However, the clear mechanism is not known yet.
Next, these injected nanovesicles are site-specific for solid tumor without affecting healthy tissues (heart, lung, liver, spleen and kidneys) showing their better biocompatibility due to their natural surface biomarkers and biomimetic ability. The tissue toxicity has been evaluated through histopathology examinations as shown in Fig. 4D. Tissue histology exhibits no pathological damage or changes to the vital organs where (i) myofiber and muscle bundle in the heart, (ii) portal triad and central vein in the liver are noticed without any injury. Moreover, the glomerulus and tubules in the kidney are also without any histological change. On account of unique features, we can state that the engineered biomimetic ultrabright nanovesicles are safe to be used as contrast agents for localized tumor imaging and tissue visualization in vivo.