Antifibrotic therapy by sustained release of low molecular weight heparin from poly(lactic-co-glycolic acid) microparticles on bleomycin-induced pulmonary fibrosis in mice

Heparin and low molecular weight heparin (LMWH) have recently been considered useful treatment tools for inflammation. Heparin has antifibrotic activity, mediated by cellular secretion of hepatocyte growth factor (HGF). HGF has antifibrotic properties demonstrated in experimental models of lung, kidney, heart, skin, and liver fibrosis. The ability of LMWH for HGF secretion is similar to that of normal heparin. Poly (lactic-co-glycolic acid) (PLGA) is widely used for sustained drug release, because of its biocompatibility and low toxicity. LMWH-loaded PLGA microparticles are prepared by a conventional water-in-oil-in-water emulsion method. Interstitial pneumonia is a life-threatening pathological condition that causes respiratory failure when it progresses. In the present study, we investigated the therapeutic effect of LMWH-loaded PLGA microparticles in a mouse model of bleomycin-induced lung fibrosis. The ratios of fibrotic area to total area were significantly lower in mice administered LMWH-loaded microparticles than in mice administered bleomycin alone. The microparticle administration did not further enhance the gene expression for inflammatory cytokines. In a cell culture study, HGF secretion by mouse and human lung fibroblasts was significantly increased by LMWH addition. We conclude that LMWH showed anti-inflammatory activity, through the effects of LMWH-loaded PLGA microparticles on cells at sites of inflammation.

Cell culture. HPF and Mlg2908 cells were cultured in FGM2 and DMEM/F-12 containing 10% FBS and 1.0% Pen-Strep, respectively. A549 cells were cultured in F12K medium containing 10% FBS. MLE-12 cells were cultured in DMEM/Ham's F-12 containing 2% FBS, L-glutamine (2 mM), insulin (5 ug/mL), transferrin (10 µg/ mL), sodium selenite (30 nM), Hydrocortisone (10 nM), β-estradiol (10 nM), HEPES (10 mM). The wells of 96-well multiwell culture plates (Sumitomo Bakelite Co. Ltd., Tokyo, Japan) were coated with collagen type I aqueous solution. Briefly, 1% collagen solution (pH 3.0) was placed in each well and incubated for 10 min. After removal of the solution, the wells were dried for 30 min and washed twice with PBS (pH 7.4; 100 µL). Cells were seeded into the coated wells at 5000 cells /well in medium 100 (µL) supplemented with or without LMWH, and cultured for 24 h at 37 °C in an atmosphere of 5% CO 2 /95% air. The cell numbers were counted using Cell Count Reagent SF according to the manufacturer's protocol. Briefly, the medium was exchanged for 100 µL of fresh medium containing 2-(2-methoxy-4-nitrophenyl)-3-(4-nitrophenyl)-5-(2,4-disulfophenyl)-2H-tetrazolium and further incubated for 1 h. The absorbance of the medium was measured at 450 nm. The percentage of cell viability was expressed relative to 100% for cells cultured without fragmin. HGF immunoassay. Supernatants were evaluated for their HGF contents using an HGF ELISA Quantikine Kit (R&D Systems Inc., Minneapolis, MN, USA) according to the manufacturer's protocol. The experiments were carried out six times independently for each sample, and in duplicate.

Microparticle preparation.
Fragmin solution (1000 IU/mL; 200 µL) was mixed with DCM containing PLGA (50 mg/2 mL) and 2% PVA (2 mL) for 60 s at 3000 rpm using a vortex to create the first emulsion (W 1 /O). The first emulsion was poured into a 2% PVA aqueous solution (16 mL) and mixed with a vortex mixer for 60 s. This procedure permitted the formation of a double emulsion (water-in-oil-in-water: W 1 /O/W 2 ), in which the W 1 phase was homogeneously dispersed in the O phase. PVA aqueous solution (20 mL) was then added, and the W 1 /O/W 2 emulsion was continuously stirred for 6 h at room temperature until the DCM had completely evaporated. The microparticles were filtered through a 40-µm cell strainer (Corning, New York, NY, USA) and washed with double-distilled water by centrifugation three times. The mixture was freeze-dried in a freeze drier (VD-550R; Taitec, Saitama, Japan) to produce powdered microparticles.
Scanning electron microscopy observation. LMWH/PLGA-MPs were fixed on an aluminum support with carbon-adhesive glue and coated with a 10-nm thick layer of palladium using an ion sputter (30 mA; E-1030; Hitachi Ltd., Tokyo, Japan). The samples were observed using a scanning electron microscope (S-5000; Hitachi Ltd.). LMWH/PLGA-MPs size was measured by ImageJ (n = 1,617). Histological analysis and assessment of fibrotic area. The right middle lung of each mouse was fixed for 6 h in 4% PFA/PBS followed by overnight incubation in 20% sucrose/PBS. The tissues were embedded in OCT compound (Sakura FineTek, Tokyo, Japan), cut into 5-µm sections, and processed for hematoxylin and eosin (H&E) or Masson's trichrome staining for analysis of morphological or fibrotic changes. For each section, the five most severely injured non-overlapping fields (magnification: 200×) of the lung parenchyma were evaluated. Quantification of lung fibrosis in the histological specimens was performed using a numerical scale (modified Ashcroft score) 20 . The severity of the fibrotic changes in each microscopic field of a given lung section was assessed and assigned a modified Ashcroft score from 0 to 8. The overall severity in each lung section was expressed as the mean score for the examined microscopic fields. All histological examinations were assessed by three independent observers. The blue area after Masson's trichrome staining was determined as the fibrotic area. The fibrotic area/total area ratio was measured by a WinROOF Ver 6.1 computerised morphometry system (Mitani, Fukui, Japan) for each section.

Scientific Reports
Quantitative real-time RT-PCR (qPCR). Total RNA was extracted from lungs of mice at 21 days after receiving BLM (n = 5 per group). After RNA extraction with an RNeasy Mini Kit (Qiagen Ltd., Manchester, UK), cDNA was synthesised using an ExScript RT Kit (Takara, Shiga, Japan) and amplification was performed in a Sequence Detection System 7000 (Thermo Fisher Scientific Inc., MA, USA) according to the manufacturer's instructions. The primer sequences for matrix metalloproteinase-2 (MMP-2), matrix metalloproteinase-9 (MMP-9), tissue inhibitor of metalloproteinase-1 (TIMP-1), tissue inhibitor of metalloproteinase-2 (TIMP-2), collagen type 1 alpha 1 (COL1A1), hepatocyte growth factor (HGF), interleukin-1 beta (IL-1β), interleukin-6 (IL-6), interleukin-12 (IL-12), tumor necrosis factor-α (TNF-α), and glyceraldehyde 6-phosphate dehydrogenase (GAPDH) as a housekeeping gene are summarised in Table 1. The amount of each target gene expression was normalised to the housekeeping gene expression. The following PCR conditions were used: 95 °C for 3 min, followed by 40 cycles of 95 °C for 10 s, 56 °C for 30 s, and 95 °C for 10 s, and a final extension at 95 °C for 50 s. The relative mRNA expression level of each target gene was calculated by the comparative CT method 21 . The experiments were carried out five times independently for each sample, and in duplicate.

Statistical analysis.
All results are expressed as mean ± standard deviation (SD). Data were analysed by non-parametric one-way ANOVA followed by a multiple-comparisons test. Differences were considered significant for values of p < 0.05. All statistical analyses were performed using GraphPad Prism 7.04 software (Graph-Pad Software, San Diego, CA, USA).
HGF secretion. Figure 2 shows the amounts of HGF secretion. The amounts of HGF secreted by Mlg2908 ( Fig. 2A) and HPF (Fig. 2B) cells were significantly increased after LMWH addition.
Microparticles observation. Figure 3A-C shows images of LMWH/PLGA-MPs at different magnifications. The microparticle size distribution were shown in Fig. 3D. The microparticles size were 6.4 ± 4.2 µm. The minimam and maximan size of the microparticles were 0.97 and 31.2 µm, respectively. The observations confirmed that microparticles of less than 40 µm were prepared.

LMWH/PLGA-MPs reduce lung fibrosis.
Representative lung sections were subjected to Masson's trichrome staining at 28 days after starting BLM administration (Fig. 4A-F). The continuous subcutaneous infusion of BLM using an osmotic minipump forms fibrotic lesions with collagen deposition mainly on pleural side (small write arrows). The lung sections in the BLM-alone group showed a diffuse increase in collagen deposition (Fig. 4B,E) compared with the normal lung sections (Fig. 4A,D). The increase in collagen deposition in the lung was reduced by administration of LMWH/PLGA-MPs (Fig. 4C,F). The fibrotic area/total area ratio (Fig. 4G) and fibrosis score (Fig. 4H) were significantly higher in the BLM-alone group than in the normal group, but significantly lower in the BLM-LMWH group compared with the BLM-alone group.

LMWH/PLGA-MPs reduce fibrosis in the BLM-IP lung.
To determine whether the decrease in collagen was caused by reduced synthesis, whole lung mRNA expression was analysed by quantitative real-time PCR for COL1A1, TIMP-1, TIMP-2, MMP-2, MMP-9, and HGF ( Fig. 5A-F). Significant increases in COL1A1 and TIMP-1 relative mRNA expression levels were observed at 28 days after BLM administration. The COL1A1 and TIMP-1 relative mRNA expression levels at 28 days after starting BLM administration were significantly lower in the BLM-LMWH group compared with the BLM-alone group. There were no significant changes in the TIMP-2, MMP-2, MMP-9, and HGF relative mRNA expression levels among the groups.

Evaluation of anti-inflammatory effects.
To evaluate the anti-inflammatory effects of LMWH, whole lung mRNA expression was analysed by quantitative real-time RT-PCR for the following inflammatory cytokines: IL-1β, IL-6, IL-12, and TNF-α (Fig. 5G-J). There were significant increases in the IL-1β, IL-6, and IL- www.nature.com/scientificreports/ 12 relative mRNA expression levels and a tendency toward an increase in the TNF-α relative mRNA expression levels at 28 days after BLM administration. However, there were no significant changes in the IL-1β, IL-6, IL-12, and TNF-α relative mRNA expression levels in the BLM-LMWH group compared with the BLM-alone group.

Discussion
This study demonstrates that LMWH showed anti-inflammatory activity, through the effects of LMWH/PLGA-MPs preventive against fibrogenesis in the BLM-IP mice. The Mlg2908 and HPF cells had served as models of pulmonary fibroblasts. The MLE-12 and A549 cells had served as models of alveolar Type II pulmonary epithelium. The LMWH showed significant cytotoxicity toward Mlg2908 cells at higher concentrations (10 and 100 µg/ mL). However, the amounts of HGF secretion were significantly increased by LMWH addition (Fig. 2A). It may be considered that the LMWH simply suppressed proliferation of Mlg2908 cells and increased HGF secretion at higher concentrations (10 and 100 µg/mL). There were no significant differences in the HGF relative mRNA expression levels between the BLM-alone and BLM-LMWH groups. However, in the in vitro cell culture experiments, the amounts of HGF secretion were significantly increased by LMWH addition. Similar to our results, there are several reports that heparin or LMWH increased HGF secretion without affecting HGF relative mRNA expression levels 9,22 . These results suggest that LMWH may affect the promoter region of the HGF gene without increasing the gene expression 10 . Activation of HGF resulted in phosphorylation of the HGF receptor (c-Met) 23 . Furthermore, the anti-fibrotic effect of the HGF/Met pathway is not restricted to the lung and has been consistently demonstrated in experimental models of liver, kidney, heart, and skin fibrosis [24][25][26][27][28][29] . The roles of HGF/ met signaling pathway have been reported. Administration of HGF to bleomycin-induced pulmonary fibrosis in mice increased lung MMP activities and enhanced myofibroblast apoptosis 30 . HGF also induces the expression of cyclooxygenase-2 (COX-2), which is a prominent source of the potent antifibrotic prostaglandins 31 . The In the present study, LMWH/PLGA-MPs capable of sustained LMWH release were produced. The LMWHtreated group significantly suppressed lung fibrosis compared with the untreated group in BLM-IP mice. IP induced by inhalation of BLM through the airway causes focal inflammatory/fibrotic lesions around bronchioles, which is not clinically compatible with IP in histopathology. In contrast, the continuous subcutaneous infusion of BLM using an osmotic minipump forms inflammatory/fibrotic lesions mainly on pleural side, which is suitable for an IP model. The mechanisms by which LMWH suppressed lung fibrosis were suggested to be suppression of fibrosis formation by HGF and extracellular matrix (ECM)-degrading enzymes. The antifibrotic effect of LMWH has been experimentally confirmed in mouse models of CCl 4 -induced hepatitis 22 and unilateral ureteral obstruction kidney fibrosis 33,34 .
We conducted animal experiments by using ICR mice, which are easy to administration of PLGA-MPs via the tail vein. A pulmonary fibrosis model can be prepared using this outbred mice by administering bleomycin similarly to inbred C57BL/6 mice 35 . In qPCR analyses using lung tissue, LMWH/PLGA-MP administration significantly suppressed the gene expression of COL1α1 and TIMP-1 in the BLM-IP mouse lung. These findings indicate that LMWH suppressed the gene expression of fibrosis-related factors involved in epithelial and endothelial cell injury, fibroblast proliferation, and abnormal repair in lung tissue. In lungs with IP, MMPs are activated during the process of tissue repair and there is a mechanism for ECM degradation 30,36,37 . In the present study, BLM tended to promote MMP gene expression in lung tissue, but there was no significant difference compared with normal mice. Possibly because of this effect, LMWH/PLGA-MP administration tended to suppress MMP-2 gene expression, but there was no significant difference, and MMP-9 gene expression remained unchanged. A reason why the expression of MMP genes in lung tissue was not significantly promoted by BLM may be that the expression levels were only evaluated in one phase during the time course of IP lung fibrosis. In the future, it may be useful to evaluate the expression at various time phases during the time course of IP lung fibrosis. The expression level of inflammatory cytokine genes tended to decrease in the LMWH/PLGA-MP administration, but there was no significant difference against BLM-alone (Fig. 5G,H,J). A more detailed anti-inflammatory effect of LMWH/PLGA-MP may be evaluated by early phase of inflammation. www.nature.com/scientificreports/ The dose of a drug can be reduced and its side effects can be alleviated using PLGA MPs 38 . In BLM-IP mice, gene expression of inflammatory cytokines in lung tissue was enhanced. However, LMWH/PLGA-MP administration did not further enhance the gene expression of inflammatory cytokines. These findings suggest that LMWH/PLGA-MPs did not cause inflammation in the lungs of the BLM-IP mice evaluated in the present study.

Conclusions
When administered alone, almost all LMWH exhibits anticoagulant activity. In the present study, LMWH showed anti-inflammatory activity, through the effects of LMWH/PLGA-MPs on cells at sites of inflammation. The LMWH/PLGA-MP administration did not further enhance the gene expression of inflammatory cytokines. Combinations of LMWH/PLGA-MPs and anti-inflammatory drugs or mesenchymal stem cells with anti-inflammatory activity may be more effective as anti-inflammatory treatment systems.

Data availability
The data that support the findings of this study are available from the corresponding author upon reasonable request.