Codelivery of Emodin and Diammonium Glycyrrhizinate by Anti- alpha8 Integrin-Conjugated Immunoliposomes for the Treatment of Renal Fibrosis

The targeted delivery of therapeutics to the kidneys has a profound potential for the management of renal fibrosis. Thus, we developed a drug delivery system that targets mesangial cells by conjugating anti-alpha8 integrin to the surface of liposomes. We coloaded emodin (EMO) and diammonium glycyrrhizinate (DAG) to the immunoliposomes for combined therapy. The coloaded immunoliposomes were small size (92:4 ± 0:4 nm), narrowly distributed, and with nearly neutral zeta potential and good stability. The encapsulation rate of EMO and DAG in immunoliposomes was 45:5 ± 2:0% and 44:3 ± 1:1%, respectively. Using a BCA assay, the actual number of antibody molecules attached to a single liposome was determined as being approximately 41. An in vitro release study showed that EMO and DAG could be ratiometrically released from the immunoliposomes, which means that an optimized synergistic ratio of the two drugs could be achieved. Studies on cellular uptake studies demonstrated an approximately 3-fold increase for immunoliposomes in HBZY-1 cells compared to nonconjugated liposomes. In vitro cell growth inhibition and Western Blot assay revealed that the coloaded immunoliposomes exhibited a stronger and synergistic in vitro antifibrosis effect against NIH3T3 and HBZY-1 cells in vitro. Taken together, it indicated that anti-alpha8 integrin-modified immunoliposomes for codelivery of EMO and DAG have great potential for targeting the kidneys for the treatment of renal fibrosis.


Introduction
Renal fibrosis, which is the final common outcome of progressive kidney diseases, represents an attractive target for therapy in numerous chronic kidney diseases. More than 10% of the global population is affected by renal fibrosis [1,2]. Renal fibrosis is characterized by the activation and proliferation of renal interstitial fibroblasts, excessive accumulation of extracellular matrix (ECM) in the renal stroma, and destruction of renal tissue [3,4]. Numerous investigations have demonstrated that the development of renal fibrosis is closely associated with mesangial cells [5,6]. Glomerular mesangial cells are involved in glomerular inflammatory response, repair of basement membrane, and the initiation of ECM overproduction, which are signs of progress of renal fibrosis. Phenotypic transformation of glomerular mesangial cells is activated after injury, followed by the production of several signaling molecules are produced such as transforming growth factor-β (TGF-β) and platelet-derived growth factor (PDGF), which contribute to excessive proliferation of mesangial cells and increases in matrix synthesis [7]. Mesangial cells (MCs) proliferation and extracellular matrix accumulation are also involved in the occurrence of glomerular inflammation and aggravate its progression to chronic diseases. Therefore, specific inhibition of mesangial cell activation and proliferation is of great significance in the treatment of renal fibrosis.
Alpha8 integrin, an RGD binding integrin, is a subtype of the integrin family and is expressed on mesangial cells. The alpha8 integrin chain combines with β1 integrin on mesangial cells to form the functional heterodimeric proteins.
These proteins play a critical role in cell proliferation, adhesion, differentiation, attachment, apoptosis, and phagocytosis, as well as in migration processes [8,9]. Alpha8 integrin has already been proved to target therapeutic cargoes to mesangial cells. Thus, the treatment of renal fibrosis can be achieved by targeting mesangial cells mediated by anti-alpha8 integrin antibodies.
Clinically, angiotensin-converting enzyme inhibitors (ACEI), angiotensin receptor blockers (ARBs), and rennin inhibitors are recognized as first-choice medications for blocking the progression of renal fibrosis [10]. However, these therapeutic strategies are not effective enough in patients with middle or late stages of renal fibrosis. Moreover, hyperkalemia or even renal dysfunction caused by long-term application of ACEI/ARB is also the weakness of this therapeutic strategy. Furthermore, cytostatic and immunosuppressant drugs administered in clinics commonly produce severe side effects.
Nowadays, novel therapeutics and drug delivery systems have been proposed for slowing, arresting or even reversing, the progression of renal fibrosis including plant-derived active compounds [6]. Emodin (EMO) is an active anthraquinone present in the roots and barks of Chinese herbs, such as Rheum palmatum [11], Polygonum cuspidatum [12], and Polygonum multiflorum [13]. Investigations have shown that EMO ameliorates renal fibrosis in rats by reducing ECM synthesis by suppressing transforming growth factor-β1 (TGF-β1) and fibronectin (FN) overexpression [14], thus decreasing mRNA expression of TGF-β1 [15] and activating the PI3K/Akt/GSK-3β pathway [16]. EMO can also significantly inhibit the proliferation and hypertrophy of mesangial cells and extracellular matrix production. However, clinical trials have proven that EMO does not have insufficient quality and presents significant side effects especially when administered alone, due to its low water solubility and kidney nonspecificity [17], which prevent its use in patients. Diammonium glycyrrhizinate (DAG) is the diammonium salt of glycyrrhizin extracted from the root and rhizome of Glycyrrhiza uralensis Fisch [18], which has been reported as beneficial to renal fibrosis [19]. DAG can inhibit the expression of TGF-β1 and connective tissue growth factor (CTGF) in renal interstitial tissue [20], reduce the accumulation of extracellular matrix, alleviate the pathological damage to renal glomerulonephritis, and produce a positive effect on renal protection [21]. Hence, the combination of EMO and DAG can act on renal fibrosis by utilizing multiple nonoverlapping and synergistic mechanisms, thus improving antifibrosis activity. We explored the cell growth inhibition of EMO and DAG in NIH3T3 cells in a previous study. Results showed that the combined use of EMO and DAG could exert a synergistic effect when the mole ratio between EMO and DAG changed from 1 : 2 to 1 : 10 ( Figures S1 and S2). Moreover, renal fibrosis-associated protein expression including α-SMA, TGF-β1, Col-I, and FN expression was significantly decreased (Figures S3 and S4). Nevertheless, the traditional combination cannot effectively control the dosage regimen and effective drug ratio at the target tissue, which is caused by different physicochemical properties and pharmacokinetics of individual drugs [22]. A promising strategy for addressing these drawbacks is the use of nanosized delivery platforms. Moreover, nanosized delivery platforms, such as liposomes and polymeric nanoparticles, are important targeting strategies to mesangial cells [23]. Thus, liposomes may be an appropriate delivery system for codelivery of EMO and DAG in renal fibrosis, and it can encapsulate both hydrophilic (DAG) and hydrophobic drugs (EMO) based on the amphiphilic properties of phospholipids and guarantee the optimized synergistic effect of DAG and EMO. To date, liposomes have gained momentum in the research of drug delivery with advantages such as improved drug efficacy, low toxicity, and enhanced pharmacokinetics and stability [24]. Significantly, liposomes (LPs) can be conjugated to antibodies on their surface to form immunoliposomes (ILPs). ILPs modified by Thy1.1 antibody or anti-α8 integrin have been used mainly for targeting to mesangial cells. Moreover, ILPs modified with anti-α8 integrin have higher translational potential in humans [23].
The purpose of this investigation is to develop coloaded anti-alpha8 integrin-modified EMO/DAG-ILPs and to evaluate their cell binding and uptake properties by HBZY-1 (alpha8 integrin receptor positively expressed cells) and HK-2 cells (alpha8 integrin receptor negatively expressed cells). Furthermore, the NIH3T3 cell line, which was widely used in the study of renal fibrosis as an in vitro model of fibrosis [25][26][27], was chosen to evaluate the antifibrotic effects and mechanism of EMO/DAG-loaded ILPs, cell growth inhibition, and levels of fibrosis protein expression.

Preparation of EMO/DAG-Loaded
ILPs. The functionalization of liposomes with anti-alpha8 integrin was performed using a modified postinsertion method [28,29]. First, anti-alpha8 integrin was dissolved in HEPES buffer (pH 8.0) containing 2 mM EDTA and then thiolated with Traut's reagent at a molar ratio of 1 : 50, followed by incubation for 1 h at room temperature. The unreacted Traut's reagent was then removed from a HiTrap desalting column eluted with HEPES buffer (pH 7.4). Subsequently, the obtained thiolated monoclonal antibody (mAbs) was directly mixed with DSPE-PEG2000-Mal at a molar ratio of 1 : 100, shaken well and incubated for 12 h at 4°C. Then, EMO/DAG-loaded ILPs (EMO/DAG-ILPs) with anti-alpha8 integrin to lipids at a molar ratio of 1 : 250 were then prepared by mixing the resultant DSPE-PEG2000-Mal-mAbs with preformed liposomes and incubating them for 24 h at 4°C with continuous and gentle shaking. Final formulations were purified on a Sepharose CL-4B column with HEPES buffer (pH 7.4) as the eluent and then stored at 4°C until used. The ILPs containing DiI were prepared according to the above-mentioned method under dark conditions.

Characterization of Liposomal Samples.
Particle size distribution, polydispersity index (PDI), and zeta potential of LPs and ILPs were measured at 25°C using a dynamic light scattering (DLS) and the laser Doppler electrophoresis system (ZEN 3690, Malvern, Worcestershire, UK). Samples were diluted in distilled deionized water and then dropped on the surface of a copper grid and air dried in order to observe the morphology of liposomes. The grid was observed using transmission electron microscopy after staining with 2.0% phosphotungstic acid hydrate (H-7650, Hitachi, Japan). Entrapment efficiency (EE%) in liposomal formulations was determined using the ultrafiltration centrifugation method. To achieve this, 1 mL of liposomal formulations was added into an ultrafiltration centrifuge tube (MWCO, 30 kDa, Millipore, USA) and centrifuged for 10 min at 10000 r/min. The amounts of EMO and DAG in the filtrate were analyzed using a valid HPLC method. The total liposome fraction was also disrupted and solubilized in menthol. The samples were then diluted, sonicated, and analyzed in triplicate using the same HPLC method mentioned above. The C18 column (Ultimate XB-C18, 5 μm, 4:6 × 250 mm, Welch Inc., USA) was used for HPLC analysis. The mobile phase consisted of acetonitrile and 0.2% phosphoric acid, and the UV detection wavelength was 250 nm.
EE% was calculated according to the following equation: where W total means the amount of drug in the total fraction and W unencapsulated means the weight of unencapsulated drug.

Antibody Conjugation Efficiency in ILPs.
The conjugation efficiency of anti-alpha8 integrin coupled to the surface of liposomes was determined though BCA protein assay, described as follows [30]. Briefly, the BCA working reagent was configured by mixing reagent A with the reagent B at a volume of 50 : 1. Then, 20 μL of each protein standard sample was added into 96-well plates, followed by adding 200 μL of working reagent and incubated at 37°C for 30 min. Subsequently, absorbance was measured at 562 nm using a microplate reader (SpectraMax M2, Molecular Devices, USA). The number of conjugated antibodies on the surface of the ILPs was calculated considering the following assumptions: (i) the liposome was unilamellar; (ii) the number of phospholipid molecules per liposome was approximately 80000 [31]; and (iii) the molecular mass of anti-alpha8 integrin was 117KDa [32]. Antibody conjugation efficiency and number of antibodies on the ILPs were calculated using equations as follows: Conjugation efficiency % ð Þ= the amount of mAbs in immnolipsomes the total mAbs employed for conjugation × 100%,

Number of antibodies on the immunoliposomes =
where W mAbs means the weight of anti-alpha8 integrin in ILPs, MW mAbs means the molecular mass of anti-alpha8 integrin, and M DSPC means the amount of substance of DSPC.
3 Journal of Nanomaterials 2.8. In Vitro Drug Release. The in vitro release behavior of EMO and DAG from the liposomes (EMO/DAG-LPs and EMO/DAG-ILPs) was evaluated according to the dialysis method in phosphate buffer solution (PBS, pH 7.4) containing 20% ethanol for 48 h. Briefly, 5 mL of formulation equivalent to 0.42 mg EMO and 5.46 mg DAG was sealed in a dialysis bag (MWCO 3500) and immersed into 200 mL of release medium with continuous stirring at 37°C. At prearranged time intervals, 1 mL of the solution was withdrawn and replaced with the same volume of fresh medium. The collected samples were filtered with 0.22 mm filters, and the amounts of EMO and DAG were determined using the HPLC method.
Furthermore, to evaluate the influence of plasma on release behavior, the liposomal formulation was mixed with plasma, which was diluted twice with phosphate buffer solution (PBS, pH 7.4), at a final concentration of 0.11 mg/mL for EMO and 1.39 mg/mL for DAG, sealed in a dialysis bag, and then suspended in 200 mL of phosphate buffer solution (PBS, pH 7.4) containing 20% ethanol. All experiments were carried out in triplicate.
2.9. In Vitro Cell Delivery and Uptake of ILPs. Studies on specific binding and uptake of liposomes and ILPs were performed by incubating HBZY-1 and HK-2 cells with DiIlabeled liposomal samples, which were prepared using fluorescent DiI at 16 μmol as of lipid bilayers. Cellular delivery of the liposomal samples was monitored using confocal laser scanning microscopy (CLSM). To achieve this, HBZY-1 cells and HK-2 cells were seeded onto 4-well plates at a density of 5 × 10 5 cells/well and incubated at 37°C overnight, respectively. After washing twice with phosphate buffer solution (PBS, pH 7.4), the cells were incubated with DiI-labeled liposomes (DiI-LPs) or DiI-labeled immunoliposomes (DiI-ILPs) at 37°C for 2, 4, or 6 h, respectively. The cells were then rinsed and fixed with 4% (v/v) formaldehyde for 10 min in a darkroom. Finally, the cells were further rinsed, stained with Hoechst 33342 (12 μg/mL) for 10 min, and then visualized using a confocal laser scanning microscope (Zeiss LSM510, Carl Zeiss, Germany).
To further quantify the cellular uptake, HBZY-1 and HK-2 cells were seeded onto 6-well plates at a density of 5 × 10 5 cells/well and incubated at 37°C overnight, respectively. After incubation with DiI-LPs or DiI-ILPs for 4 h, the cells were harvested, trypsinized, and washed with 4°C phosphate buffer solution (PBS, pH 7.4), and finally the fluorescence intensity was analyzed by flow cytometry (Guava easyCyte 12HT, Merck Millipore, Germany).
2.10. In Vitro Cell Growth Inhibition Assay. The inhibition effect of different formulations on cell growth was investigated in both NIH3T3 and HBZY-1 cells through MTT assays. In this study, the cells were seeded in 96-well plates at a density of 2 × 10 4 cells/well and incubated for 24 h at 37°C. Then, the culture medium was discarded, and the cells were washed with phosphate buffer solution (PBS, pH 7.4) before the different drug-loaded formulations or blank ILPs were added. After incubation for 48 h, 20 μL of MTT (5 mg/mL) solution was added to each well and incubated for another 4 h at 37°C. Subsequently, the supernatant solution containing MTT was carefully removed, and the formazan salt that formed was dissolved in 150 μL of DMSO. Absorbance was measured in each well at a wavelength of 570 nm using a microplate reader (SpectraMax M2, Molecular Devices, USA). Percentage growth (%) was calculated according to the following equation: where OD 570ðsampleÞ was the absorbance of cells incubated with various formulations, OD 570ðcontrolÞ was the absorbance of cells treated with fresh cellular culture medium, and OD 570ðblankÞ was the absorbance of culture medium without cells. The mean drug concentration required for 50% growth inhibition (IC 50 ) was calculated by the nonlinear regression equation by using the SPSS Statistics 22.0 software. Each concentration was tested in five wells, and data was presented as mean ± standard deviation (SD).
2.11. Western Blot Analysis. The antifibrotic effects of ILPs were investigated using Western Blotting by observing the expression of fibrosis protein (α-SMA, Col-I, FN, and Smad3). NIH3T3 cells and HBZY-1 cells were plated into 6-well plates at a density of 1:5 × 10 4 cells/well. After incubation overnight, the cells were treated with 5 ng/mL of TGF-β1 and incubated for 48 h [33]. The cell culture medium was then substituted with different formulations (EMO solution, DAG solution, EMO/DAG solution, EMO/DAG-LPs, and EMO/DAG-ILPs) and incubated at 37°C for 24 h. As a control for the profibrotic effect of TGF-β1, cells were cultured in medium alone (without TGF-β1 or any formulations). The cells were then harvested and lysed using RIPA buffer, and the supernatants were collected after centrifugation (12000 rpm at 4°C for 10 min). The protein concentration of each sample was determined using a BCA protein quantitative kit. Aliquots corresponding to 30 μg of the protein were separated on 8% SDS-PAGE gel for 1 h, followed by transfer onto the PVDF membranes for 1.5 h. The membranes were blocked with 5% skimmed milk for 1 h followed by overnight incubation at 4°C with diluted primary antibodies against α-SMA (1 : 500, Abcam, USA), Smad3 (1 : 500, Abcam, USA), Col-I (1 : 1000, Abcam, USA), FN (1 : 1000, Abcam, USA), and GAPDH (1 : 1000, Abcam, USA) antibodies and were used as the internal loading control. The membranes were incubated with a secondary antibody (goat anti-rabbit IgG) at room temperature for 1 h after three washings. Finally, protein bands were detected using the LI-COR/Odyssey infrared image system (LI-COR Bioscience, Lincoln, NE) for capturing images.
2.12. Statistical Analysis. Data were presented as mean ± SD. The statistical significant analysis was performed using Student's t-test through SPSS 21.0, and a P value less than 0.05 was considered statistically significant.

Preparation and Characterization of ILPs.
The morphology of liposomal formulations appeared as spherical unilamellar structures with good dispersity as shown in Figure 1.
The physicochemical properties of LPs and ILPs are shown in Table 1. Particle sizes of LPs and ILPs were 86:2 ± 1:1 nm and 92:4 ± 0:4 nm in diameter, respectively, indicating that surface modification of liposomes with anti-alpha8 integrin did not show any significant difference (P > 0:05) in particle sizes. All liposomal formulations show homogeneous dispersibility with a PDI of <0.2. The zeta potentials of LPs and ILPs were comparable and were 5:6 ± 0:4 mV and 4:3 ± 0:2 mV, respectively. Additionally, in this investigation, thiolated anti-alpha8 integrin was successfully conjugated to the DSPE-PEG2000-MAL present on the surface of liposomes with a conjugation efficiency of 60:8 ± 2:0%, corresponding to approximately 41 anti-alpha8 integrin molecules per liposome. It turned out that prepared ILPs are sufficient to achieve an efficient targeting effect, which is consistent with previous reports that at least 10-20 ligands/liposomes are sufficient to perform selective targeting delivery [34,35]. EMO was entrapped into liposomes according to the ethanol injection method with EE% of 46:7 ± 1:1% and 45:5 ± 2:0% for EMO/DAG-LPs and EMO/DAG-ILPs, respectively, as shown in Table 1. The relatively low entrapment efficiency of EMO is due to the application of stearylamine cationic agent in the preparation. The existence of a charged interface leads to a decrease in the volume of the lipid bimolecular layer [36]. Considering its strong hydrophilicity (log P = −2:89), DAG was actively loaded into the liposomal formulations according to the calcium acetate gradient method to increase its EE%. The method resulted in an EE% of DAG of 58:2 ± 5:0% in liposomes. Nevertheless, the EE% of DAG in ILPs decreased by approximately 14% compared to the LPs, which was only about 44:3 ± 1:1% and may be due to the modification process. Similar results reported by Manjappa et al. [37], Devi et al. [38], and Peres-Filho et al. [39] showed that the encapsulation efficiency of cargoes in liposomal formulations was reduced by 5%, 10%-12%, and 16%, respectively, after modification with anti-neuropilin-1, lactic acid, and folic acid, respectively. Furthermore, the determined loading molar ratio between EMO and DAG was 1 : 3.7 and 1: 2.7 in the prepared LPs and ILPs, respectively, which was consistent with the designed radiometric of EMO and DAG for synergistic effects (Figures S1 and S2). The final EMO and DAG concentrations in immunoliposomes were 0.08 mg/mL and 0.49 mg/mL, respectively.
As shown in Figure 2(a), no significant changes in particle size and PDI of ILPs were observed after incubation with 10% FBS for 12 h. Moreover, the zeta potential of ILPs hardly changed over 12 h (Figure 2(b)). Thus, both the LPs and ILPs were stable in serum despite of the low zeta potential value, which might be due to the PEG surface modification. Many studies have proved that nanoparticles stability not only depends on the repulsive electrostatic double layer (zeta potential value) but also is closely related to the van der Waals and structural forces (surface functionalization) [40][41][42]. Surface functionalization can reduce the probability of nanostructure aggrega-tion, prevent the nonspecific adsorption, and increase the steric stability of liposomes [42][43][44]. For instance, Smith et al. found that PEG modification greatly improved the stability of DOTAP liposomes and DOP liposomes [45].

In Vitro Drug
Release. The in vitro release profiles of EMO and DAG from liposomal formulations in release medium over the period of 72 h are shown in Figure 3. It is obvious that the release behavior of EMO and DAG from LPs and ILPs are similar, demonstrating that antibody modification does not affect drug release, which is consistent with previous research [46]. The limited ability of the phospholipid bilayer membrane to retain hydrophilic drugs led to higher release of DAG compared to EMO, which was 93.7% from the LPs and 91.3% from the ILPs, while the cumulative release of EMO from the LPs and ILPs was 75.8% and 71.6%, respectively. The release pattern of DAG could be divided into two stages, including initial burst release followed by sustained release. More than 20% of the drug was released within 2 h followed by an additional 70% release within 72 h. Moreover, although the molar ratio of EMO/DAG released from both LPs and ILPS exceeded 1 : 10 within the initial 2 h due to the burst release of DAG, both the LPs and ILPs can better control the molar ratio of EMO/DAG around 1 : 5 after 4 h, which is in accordance with the optimized synergistic effect.
Addition of plasma to the LPs and ILPs did not cause obvious aggregation within the dialysis bag for the duration of 72 h, neither was any noticeable change observed in the DAG release behavior. However, the presence of plasma significantly delayed the release of EMO from the LPs and ILPs, which was only approximately 41.8% and 45.5%, respectively, as illustrated in Figure 3(b). This may due to the high protein-binding efficiency of EMO, which is up to 99.6% [47].

Cellular Uptake and Intracellular Disposition.
To evaluate the intracellular disposition and uptake behavior of ILPs, both alpha8 integrin receptor positively expressed cell line, HBZY-1, and negatively expressed HK-2 cell line were used in this study. As shown in Figure 4(a), fluorescence intensity is boosted up with the increase of incubation time, demonstrating that liposomes have ability to deliver different formulations to these two cell lines. Furthermore, the fluorescence intensity of ILPs was significantly stronger in the HBZY-1 cell line than in that of the LPs, indicating enhanced uptake, possibly mediated by endocytosis through anti-alpha8 integrin and alpha8 integrin receptor interaction., No remarkable difference was seen between the LPs and ILPs in HK-2 cells. This result is consistent with previous studies, which reported that high expression of alpha8 integrin facilitated phagocytosis by renal mesangial cells [8].
Flow cytometry analysis was conducted to further quantify intracellular uptake by HK-2 and HBZY-1 cells. As shown in Figure 4  Journal of Nanomaterials HBZY-1 cells (P < 0:01), suggesting that anti-alpha8 integrin fragments conjugated to the surface of liposomes were reactive, thus facilitating the binding of ILPs to alpha8 integrin-overexpressed cells (i.e., HBZY-1 cells) specifically, as shown in Figures 4(b) and 4(d). The cellular uptake rate of ILPs by HBZY-1 cells was approximate 3-folds as high as that of the LPs (P < 0:01). These results further confirmed the effectiveness of anti-alpha8 integrin-modified ILPs in the cellular uptake of HBZY-1 cells, which is in consistent with the findings of Scindia et al. [48], who constructed anti-alpha8 integrin-modified ILPs and demonstrated that mesangial cell-targeted delivery by anti-alpha8 integrin ILs opens up a novel method for treating glomerular diseases.
3.4. In Vitro Cell Growth Inhibition Activity of ILPs. In vitro biocompatibility of blank liposomal formulations was first evaluated using the MTT assay in NIH3T3 and HBZY-1 cells. As illustrated in Figures 5(a) and 5(c), cell viabilities were more than 85% after exposure to blank liposomal formulations at a phospholipid concentration range of 0.037 to 1.184 mM. This result showed that blank liposomes, whether they were modified with anti-alpha8 integrin or not, exhibited low cytotoxicity and high biocompatibility on HBZY-1 cells and NIH3T3 cells. The MTT assay was conducted using different formulations to compare the inhibitory effect of EMO and DAG in solutions and in liposomal formulations. As shown by a higher proportion of inhibited cells (Figures 5(b) and 5(d)) and lower IC 50 values (Table 2), EMO and DAG combined treatment, whether in solutions or in liposomal formulations, showed an explicit enhanced inhibition effect in both NIH3T3 and HBZY-1 cells compared to individual treatments (P < 0:05). Furthermore, EMO/DAG-LPs and EMO/DAG-ILPs inhibited the growth of NIH3T3 cells more significantly than EMO/DAG solution (P < 0:01). This could be attributed to the enhancement of cellular uptake and effective delivering capability of liposomal formulations. However, compared to EMO/DAG-LPs, EMO/DAG-ILPs conjugated with anti-alpha8 integrin antibody did not show a significant decrease in cell viability in NIH3T3 cell lines, which does not express alpha8 integrin receptor. On the other hand, EMO/DAG-ILPs had reduced IC 50 values in HBZY-1 cells compared to the EMO/DAG-LPs, which was expected and was consistent with the cellular uptake finding, indicating that the EMO/DAG-ILPs were endocytosed via the alpha8 integrin receptor-mediated pathway. We also found that, compared to NIH3T3 cells, HBZY-1 cells are more sensitive to EMO/DAG-ILPs (P < 0:05), which might be due to the expression of alpha8 integrin receptors. Thus, we hypothesized that ILPs could weaken the transdifferentiation of MCs into fibroblasts. These findings imply that ILPs treat renal fibrosis by inhibiting the proliferation of MCs and attenuating the transformation of MCs into fibroblasts.

Regulating Effects on Renal Fibrosis-Related Proteins.
It is well known that TGF-β1 is a crucial pathological factor in promoting renal fibrosis which can directly induce the production of ECM, including collagen I (Col-I), and fibronectin (FN). TGF-β1 can also act directly on various kidney cells, such as promoting mesangial cell proliferation [49]. In addition, numbers of studies have identified Smad3 as major signaling mediators downstream of the TGF-β1 signal and have substantiated that the development of renal fibrosis is positively correlated with Smad3 [50]. α-SMA is the hallmark protein of myofibroblasts, which plays an important role in renal fibrosis, such as accelerating fibrous tissue contraction   In this investigation, NIH3T3 cells were simulated by TGF-β1 (5 ng/mL), and the renal fibrosis-related protein expression was explored after treatment with EMO/DAG-ILPs. As shown in Figure 6, the protein levels of Smad3, α-SMA, Col-I, and FN in the model group were significantly elevated after stimulation with TGF-β1 compared to the control group. After treatment with different formulations, the protein levels of Smad3, α-SMA, Col-I, and FN were significantly reduced except for the DAG group. This result suggested that both free EMO and the combination of EMO and DAG are feasible for delaying the process of renal fibrosis by inhibiting the expression of α-SMA expression and reducing ECM synthesis. It was noteworthy that the combination of EMO and DAG, whether in solution or in liposomal formulations, could significantly reduce the protein levels of Smad3, α-SMA, Col-I, and FN compared to the EMO group (P < 0:01), indicating that the combination of EMO and DAG has a synergistic effect on the treatment of renal fibrosis. Interestingly, compared to liposomal formulations, the EMO/DAG solution has shown stronger modulatory effects on fibrosis-related protein levels. This result can be explained by the differences in cellular uptake mechanisms between liposomal formulations and free drugs [51]. Endocytosis is the main method by which liposomes enter the intracellular environment, which is slower than the transmembrane process of free drugs. Furthermore, the sustained release of EMO and DAG from liposomal formulations also contributed to this phenomenon. Hence, we speculate that the expression of these renal fibrosis-related proteins will 9 Journal of Nanomaterials be further tempered by the extension of incubation with liposomal formulations, which are at least as effective as the EMO/DAG solution with regard to its in vitro antifibrotic properties [52]. Moreover, the ILPs were superior to LPs in modulating protein levels of Smad3, α-SMA, Col-I, and FN, especially for Col-I. The downregulation effect of LPs and ILPs on α-SMA and FN protein in NIH3T3 cells is superior to that in HBZY-1 cells (P < 0:05), except for Smad3 and Col-I (P > 0:05). This result may be related to the difference of cell lines.
Changes in renal fibrosis-related proteins in HBZY-1 cells were also evaluated after the cells were treated with different formulations considering the important role that MCs play in kidney fibrosis. As shown in Figure 7, formulations including EMO/DAG solution, EMO solution, EMO/DAG-LPs, and EMO/DAG-ILPs all showed strong suppression of renal fibrosis-related proteins. Note that the combination of EMO and DAG, whether in solutions or in liposomal formulations, had a superior ability to downregulate the Smad3 protein. It is well recognized that Smad3 is the key mediator of TGF-β1induced ECM production and tissue fibrosis [53,54].

Conclusion
We have successfully designed and characterized EMO/DAG coloaded anti-alpha8 integrin-conjugated immunoliposomes for targeted delivery and combined therapy. The coloaded immunoliposomes presented nanometric particle size, high drug entrapment, and sufficient conjugation efficiency. In vitro drug release demonstrated that the molar ratio of the released agents was within the synergy range after a small burst release of DAG. Cells with a high level of alpha8 integrin expression showed better cellular uptake of immunoliposomes and higher cell growth inhibition. Moreover, the combination of EMO and DAG can inhibit the expression of renal fibrosis-related proteins (α-SMA, Smad3, Col-I, and FN), thereby delaying the progress of renal fibrosis. Based on the promising results of in vitro studies, we anticipate that the EMO/DAG-loaded immunoliposomes will prove to be an efficient delivery system for the treatment of renal fibrosis in vivo.

Data Availability
On data availability statement, the raw/processed data required to reproduce these findings cannot be shared at this time as the data also forms part of an ongoing study.

Conflicts of Interest
No potential conflict of interest was reported by authors.