Treatment of Idiopathic Pulmonary Fibrosis by Inhaled Silybin Dry Powder Prepared via the Nanosuspension Spray Drying Technology

Abstract

Idiopathic pulmonary fibrosis (IPF) is a kind of life-threatening interstitial lung disease characterized by progressive dyspnea with accurate pathogenesis unknown. At present, heat shock protein inhibitors are gradually used to treat IPF. Silybin, a heat shock protein C-terminal inhibitor, has high safety and good application prospects. In this work, we have developed a silybin powder able to be used for inhalation administration for the treatment of IPF. Silybin powder was prepared by the spray drying method and identified using cascade impactometry, particle size, scanning electron microscopy (SEM), differential scanning calorimetry (DSC), X-ray diffraction (XRD), and Fourier transform infrared (FT-IR) spectroscopy. A rat model of bleomycin-induced IPF was used to assess the effect of inhaled silybin spray-dried powder. Lung hydroxyproline content, wet weight, histology, inflammatory factor expression, and gene expression were examined. The results showed that inhaled silybin spray-dried powder alleviated inflammation and fibrosis, limited hydroxyproline accumulation in the lungs, modulated gene expression in the development of IPF, and improved postoperative survival. The results of this study suggest that silybin spray-dried powder is an attractive candidate for the treatment of IPF.

Idiopathic pulmonary fibrosis (IPF) is an end-stage lung disease characterized by fibroblast proliferation and accumulation of a large amount of extracellular matrix (ECM), accompanied by inflammatory damage and tissue structure destruction. Although the accurate cause of IPF is still unknown, the common pathological process is characterized by failure of alveolar reepithelialization, activation of fibroblasts, and destruction of normal lung architecture by excessive deposition of ECM including collagen.¹⁻⁴

Among the complex factors that contribute to IPF, transforming growth factor-β (TGF-β) is the key cytokine in the fibrotic process and is involved in the differentiation and activation of fibroblasts resulting in collagen accumulation and other extracellular matrix component production.⁵ Thus, inhibiting TGF-β or blocking its signal transformation can inhibit IPF. Recently, the effect of heat shock protein-90 (HSP90) on the TGF-βR signaling pathway has been proved in renal fibrosis,⁶ skin fibrosis,⁴ cancer cell lines,⁷ or pulmonary fibrosis.⁸ Tanespimycin (17-AAG), a classical N-terminal inhibitor of HSP90, was found to block the formation of the HSP90/TGF-βR complex, terminate the TGF-β signaling pathway, and block myofibroblast transdifferentiation of lung fibroblasts or epithelial cells. However, 17-AAG can destroy glucocorticoid receptors, resulting in significant side effects and difficult clinical application.

Silybin, a flavonoid lignin component extracted from the seeds of Silybum marianum, is a commonly used liver protection drug that has a certain therapeutic effect on liver fibrosis.⁹ Studies have found that silybin has a role in the treatment of cancer,¹⁰ pneumonia,¹¹ and pulmonary fibrosis.¹² Silybin has recently been discovered as a novel C-terminal inhibitor, which can inhibit the efficiency of HSP90 α/β CTD binding to its co-chaperone PPID/cyclophilin D in the low millimolar range. Moreover, the hepatotoxicant behavior of silybin solely occurred at concentrations several thousand times higher than those of the HSP90 N-terminal inhibitor 17-AAG. Therefore, it is highly desirable to investigate whether silybin could have the same antipulmonary fibrotic effect as 17-AAG.

In this study, we developed an inhalable silybin powder and evaluated its efficacy in a bleomycin-induced IPF rat model. We performed gene transcriptomic analysis on rat lungs and further explored their effects on IPF. At the same time, the physicochemical properties, release curve, lung deposition characteristics, and particle size of silybin powder were also optimized.

Results

Preparation of Silybin Dry Powders

Pretreatment of Silybin Spray Drying Suspension

The particle sizes of the silybin suspensions were measured for 0, 4, 8, and 12 cycles of homogenization by an APV-2000 High-Pressure Homogenizer (SPX FLOW, German) and 0, 1, 2, and 3 h of ball milling by an NT-0.3L Nanometer sand mill (Langling Machinery Co., Ltd., China), respectively. As shown in Table 1, the particle size of the silybin suspension was reduced to 552.2 nm after 12 cycles of high-pressure homogenization. The change of particle size in the last four cycles was less than 50 nm, indicating that the high-pressure homogenization had limited influence on it. In contrast, the particle size of silybin could reach about 300 nm after ball milling. Therefore, silybin suspension after ball milling for 3 h was selected as the research object in the subsequent experiments.

Table 1. Particle Size of Silybin Suspensions after High-Pressure Homogenization and Ball Milling.

pretreatment method	particle size (nm)	PDI
homogen 0 cycles	1397.7 ± 214.0	0.994
homogen 4 cycles	705.5 ± 66.8	0.385
homogen 8 cycles	600.9 ± 53.6	0.143
homogen 12 cycles	552.5 ± 40.9	0.209
ball-milling 0 h	1147.3 ± 175.1	0.932
ball-milling 1 h	362.4 ± 10.1	0.403
ball-milling 2 h	355.6 ± 4.1	0.343
ball-milling 3 h	318.9 ± 4.0	0.293

Optimization of Spray Drying Conditions

As shown in Table 2, solid content (A), inlet temperature (B), feed volume flow rate (C), and gas volume flow rate (D), four factors were selected, and performed a four-factor three-level orthogonal test according to L9 (34) orthogonal table. The combined score of fine particle fraction (FPF) and yield was used as an indicator, combined score = FPF × 0.6 + yield × 0.4.

Table 2. Optimization Factors for the Spray Drying Process of Silybin Spray-Dried Powder.

	factors
level	A (solid content/%)	B (inlet temperature/°C)	C [flow rate/(mL/min)]	D [gas volume flow rate/(L/h)]
1	1.5	120	1.5	357
2	1	130	3	473
3	0.5	140	6	610

As can be seen from Table 3, among the investigated four factors, gas volume flow rate had the greatest influence on the experimentally investigated index, followed by solid content and inlet volume flow rate, and the least influence on the overall score of yield and deposition rate was temperature, i.e., D > A > C > B. The analysis of variance (ANOVA) results in Table 4 indicated that solid content, inlet temperature, gas volume flow rate, and inlet volume flow rate all had significant effects on the experiment. Combined with the results of the orthogonal tests, we determined that the optimum spray drying conditions for silybin spray dry powder were A₂B₃C₂D₃. From the validation results (Table 5), it can be seen that the relative standard deviation (RSD) of the comprehensive score of silybin powder aerosol obtained under the optimum spray drying conditions is 1.02 < 2%, indicating that this condition was stable and feasible. In summary, a final silybin suspension with 1% solids was prepared at an inlet temperature of 140 °C and a gas volume flow rate of 610 L/h at a feed volume flow rate of 3 mL/min to obtain the silybin spray dry powder with the highest overall score.

Table 3. Arrangement and Results of an Orthogonal Test for Optimizing Spray Drying Conditions for Silybin Spray-Dried Powder.

number	A	B	C	D	yield (%)	FPF (%)	score (%)
1	1	1	1	1	25.72	13.92	18.64 ± 0.96
2	1	2	2	2	41.91	19.07	28.20 ± 0.63
3	1	3	3	3	44.97	25.37	33.21 ± 0.50
4	2	1	2	3	56.45	24.73	37.42 ± 0.70
5	2	2	3	1	27.21	19.46	22.56 ± 0.41
6	2	3	1	2	56.22	27.31	38.87 ± 1.32
7	3	1	3	2	41.30	26.65	32.51 ± 1.58
8	3	2	1	3	54.76	26.56	37.84 ± 0.86
9	3	3	2	1	31.00	24.26	26.96 ± 0.72
K1	80.05	88.57	95.35	68.16
K2	98.85	88.61	104.92	99.58
K3	97.31	99.04	88.28	108.47
k1	26.68	29.52	31.78	22.72
k2	32.95	29.54	34.97	33.19
k3	32.44	33.01	29.43	36.16
extreme difference	6.26	3.49	5.55	13.43
primary and secondary order	D > A > C > B
excellent level	A2	B3	C2	D3
excellent combination	A2B3C2D3

Table 4. Analysis of Variance for Spray Drying of Silybin Spray-Dried Powder.

factor	bias sum of squares	degree of freedom	F-value	significance
A	217.6	2	126.9	0.000
B	72.8	2	42.5	0.000
C	25.4	2	14.8	0.000
D	897.0	2	523.3	0.000
error	15.4	18

Table 5. Validation of the Best Process for Silybin Spray-Dried Powder.

number	yield (%)	FPF (%)	score (%)	average	standard deviation	RSD
1	60.9	27.8	41.02	40.72	0.42	1.02%
2	59.0	28.8	40.89
3	57.1	29.0	40.25

Preferential Selection of Silybin Spray-Dried Powder

From the scanning electron microscopy (SEM) pictures shown in Figure 1A–F, we found that most of the spray-dried silybin powders showed a spherical shape with a porous structure formed by many small particles. Such a porous structure had a greater surface area than that of a smooth crystalline form, which is expected to increase contact with the lung surface and facilitate lung absorption.

Figure 1.

Scanning electron microscope images (magnification 5000× and 20,000×) of silybin API (A, B) and (C, D) silybin spray-dried powder and (E, F) silybin spray-dried powder (2.5% Poloxamer 188). Pulmonary deposition rates of silybin spray-dried powder at different concentrations of Poloxamer 188 and Tween 20 (G). Stability test (H), variable flow rate test (I), aeration test (J), and permeability test (K) for silybin spray-dried powder (2.5% Poloxamer 188). Differential scanning calorimetry (DSC) (L), infrared (M), X-ray diffraction (N), and in vitro dissolution profiles (O) of silybin API, silybin spray-dried powder, and silybin spray-dried powder (2.5% Poloxamer 188). *P < 0.05 compared to P-2.5% group; **P < 0.01 compared to P-2.5% group (n = 3).

After determining the optimum spray drying process, we examined the concentrations for the previously added Tween 20 and Poloxam 188 in an aqueous solution. To keep the amount of surfactant as low as possible, the maximum amount we set is 2.5% of the weight of the silybin. Although there was no visual difference between the spray-dried powder with and without the addition of Poloxamer 188, the spray-dried powder of silybin supplement with 2.5% of Poloxamer 188 had the highest lung deposition rate (Figure 1G).¹³

As shown in Figure 1H, the SI of the spray-dried powder of silybin (2.5% Poloxamer 188) was 1.31 and the curve was essentially smooth, indicating that the sample was stable (the optimum SI ≈ 1). The variable flow rate determination showed an FRI of 2.22 (1.5 < FRI < 3), indicating that the spray-dried powder of silybin (2.5% Poloxamer 188) was the nature of the vast majority of powders (Figure 1I). As shown in Figure 1J, the spray-dried powder of silybin (2.5% Poloxamer 188) had AR (aeration energy ratio) = 0.81 < 1, and AE10 (aeration energy for the 10th test) = 10.04 MJ > 10 MJ, indicating that the particles are viscous and have high cohesion with each other, which needs to be improved by the addition of lactose. As shown in Figure 1K, as the pressure applied to the sample increased, the number of airflow channels decreased and the resistance to the passage of air through the interior of the sample increased. It can be seen that the silybin spray-dried powder (2.5% Poloxamer 188) has a moderate permeability, similar to most powders.

Differential scanning calorimetry, infrared spectroscopy, and X-ray diffraction were used to detect changes in the properties of silybin spray-dried powder supplemented with 2.5% Poloxamer 188 [SLB-SPR (2.5% P)] compared to the API of silybin. As seen in Figure 1L, the melting endothermic peak of silybin API was evident and became smaller after spray drying. The peak was further reduced after the addition of 2.5% of Poloxamer 188. The smaller the melting endothermic peak, the closer the powder state was to the amorphous state. SLB-SPR and SLB-SPR (2.5% P) have glass transitions, where the T_g of SLB-SPR is 48.4 °C and that of SLB-SPR (2.5% P) is 36.0 °C. IR results showed that spray drying did not affect the chemical structure of the silybin (Figure 1M). XRD diffraction results showed that the corresponding peak intensity decreased after spray drying, indicating that the crystallinity of the sample was reduced (Figure 1N), which is consistent with the results of DSC. The dissolution of silybin in the first 2 h was greatly enhanced after spraying dry, which facilitated the release of silybin and drug absorption (Figure 1O). Together, we determined that the silybin powder prepared by spray drying of nanosuspension with the addition of Poloxamer 188 was superior for further formulation.

Pharmacodynamic Study of Inhaled Silybin Dry Powder for the Treatment of IPF

The present experimental modeling design was to observe the improvement of IPF in rats modeled with bleomycin in the presence of silybin. The total dose of bleomycin was (3U/kg) through two atomization inhalations, achieving 100% death-free survival (Figure 2A). Groups III and IV represented the administration of silybin during the inflammatory (0–14 days) and fibrotic (15–28 days) stages, respectively.

Figure 2.

(A) Photographs of bleomycin modeling and inhalation drug delivery for treatment of IPF. (B) Body weight changes in rats in the control, BLM, and BLM + SLB (0–14 days) groups were dissected on day 14, n = 3. *P < 0.05 compared to the control group. (C) Body weight changes in rats in the control, BLM, and BLM + SLB (15–28 days) groups were dissected on day 28, n = 3. *P < 0.05 compared to the control group.

Effect of Inhaled Silybin on the Body Weight of Bleomycin Rats

As shown in Figure 2B, the body weight of rats in the model group decreased significantly from day 0 after bleomycin administration, and inhaled silybin did not ameliorate the trend toward weight loss during the first 3 days. However, the body weight of rats in the BLM + SLB (0–14 days) group began to gradually return to the level of normal rats from day 4. After bleomycin was administrated for a second time on day 7, the body weight of the BLM + SLB (0–14 days) group decreased but has been steadily increasing since then without excessive fluctuations. In contrast, the body weight of rats in the BLM group remained at the lowest level after both moldings.

As shown in Figure 2C, the body weight of the rats in the BLM group remained unchanged after 14 days, while the rats in the BLM + SLB (15–28 days) group slowly increased in weight, approaching the control group. Finally, only the body weight of the BLM group was significantly different from that of the control group.

SLB Inhalation Attenuates Bleomycin-Induced Fibrosis

According to the hematoxylin–eosin (H&E) staining images on day 14 (Figure 3A), the control group showed a typical normal alveolar structure with no inflammation and a normal lung structure. The lungs of bleomycin-treated rats exhibited a marked disruption of alveolar structure with diffuse severe alveolar damage and interstitial inflammatory cell infiltration. However, the alveolar structure was significantly improved in the BLM + SLB (0–14 days) group compared to the BLM group. The BLM group showed joint fibrous masses, while the silybin inhalation resulted in mostly single fibrous masses. Inflammatory infiltrates and pulmonary edema were also significantly reduced, which was consistent with the Ashcroft score (Figure 3B). Masson staining results showed larger blue areas and fibrous tissue hyperplasia in the alveolar wall of the BLM group, suggesting significant lung fibrosis. In contrast, the BLM + SLB (0–14 days) group was lower than the BLM group, and the collagen volume fraction was also significantly lower than that of the BLM group (Figure 3C). The results of the Sirius scarlet staining showed that the collagen fibers were more pronounced in the remaining two groups compared to the control group, with the BLM group showing aggregation of myofibers. In contrast, the BLM + SLB (0–14 days) group did not show significant fiber aggregation. At the same time, changes in lung mass were observed between the control group and the BLM group on day 14. By inhaling silybin powder into the lungs, it was observed that the lung weight of the BLM + SLB (0–14 days) group was lower than that of the BLM group, although there was no statistical difference (Figure 3D). According to our findings, silybin inhalation may effectively reduce IPF by modulating the inflammation caused by bleomycin in the early stages of inflammation.

Figure 3.

(A) Lung photographs obtained from rats and photographs of H&E, Masson, and Sirius scarlet staining of rat lung tissue sections at 14 days (scale = 100 μm). (B) Quantification of fibrosis by Ashcroft score; n = 9. (C) Results of collagen volume fraction, n = 9. (D) Lung weight of rats in the 14-day group, n = 3. (E) 14 days group rats lung hydroxyproline content (HYP), n = 3. (F) Lung photographs were obtained from rats and photographs of H&E, Masson, and Sirius scarlet staining of mouse lung tissue sections at 28 days (scale = 100 μm). (G) Quantification of fibrosis by Ashcroft score, n = 9. (H) Results of collagen volume fraction, n = 9. (I) Lung weight of rats in the 28-day group, n = 3. (J) Hydroxyproline content in the lungs of rats in the 28-day group, n = 3. Mean ± standard deviation (SD), *P < 0.05, **P < 0.01 compared with the control group; ^#P < 0.05, ^##P < 0.01 compared with the BLM group.

According to the pathological sections of the lungs, we found that silybin inhalation had significant effects on improving IPF during the fibrotic stages (Figure 3F). The lung fibrosis score (Figure 3G) and collagen content (Figure 3H) in the BLM + SLB (15–28 days) group were significantly lower than those in the BLM group. Masson staining showed that the blue area in the BLM + SLB (15–28 days) group was evenly distributed around the ruptured alveolar wall. The results of Sirius red staining showed that the red collagen was more concentrated in the rats of BLM treatment at day 28, indicating that the fibrosis continued to develop after 14 days. During this period, inhalation of silybin could also improve the alveolar morphology and reduce collagen content. Compared with the control group (1.24 ± 0.02 g), perfusion of BLM in the model animals increased the lung mass of the animals (1.62 ± 0.19 g) (Figure 3I), which may be due to the combined effect of increased lung collagen content and lung edema.

The degree of IPF and the changing trend of collagen in the lung are often estimated by HYP assay. As shown in Figure 3J, HYP was produced mainly in the fibrosis stage (15–28 days) rather than the inflammatory stage (0–14 days). The HYP content of the control group and the BLM group showed a significant difference on day 28 (P < 0.01). Compared with the BLM group, the content of HYP in lung tissue of the BLM + SLB (15–28 days) group decreased significantly (P < 0.01) by 24.44%. More importantly, there was no significant difference in HYP content between the SLB group and the control group (P > 0.05), which could well reduce the collagen content to normal levels.

Mechanism Analysis of Inhaled Silybin Powder for the Treatment of IPF

After modeling or treatment for 14 days, there were no statistically significant differences in interleukin 1β (IL-1β), interleukin 6 (IL-6), tumor necrosis factor-α (TNF-α), and transforming growth factor-β1 (TGF-β1) levels in the lungs between rats in the control and model group. Inhaling of SLB slightly raised the lung inflammatory factors such as IL-1β, IL-6, and TNF-α, while slightly reducing TGF-β1 level (Figure 4A–D). Compared with the control group, serum levels of IL-1β and IL-6 were increased in the BLM group and BLM + SLB group at day 14, while levels of serum TNF-α and TGF-β1 were not significantly changed (Figure 4E–H).

Figure 4.

Enzyme-linked immunosorbent assay (ELISA) measures the levels of IL-1β, IL-6, TNF-α, and TGF-β1. Levels of IL-1β (A), IL-6 (B), TNF-α (C), and TGF-β1 (D) in the lungs of rats in the 14-day group (n = 3). Levels of IL-1β (E), IL-6 (F), TNF-α (G), and TGF-β1 (H) in the serum of rats in the 14-day group (n = 3). ELISA measures the levels of IL-1β, IL-6, TNF-α, and TGF-β1. Levels of IL-1β (I), IL-6 (J), TNF-α (K), and TGF-β1 (L) in the lungs of rats in the 28-day group (n = 3). Levels of IL-1β (M), IL-6 (N), TNF-α (O), and TGF-β1 (P) in the serum of rats in the 28-day group (n = 3). Mean ± SD, *P < 0.05, **P < 0.01 compared with the control group; ^#P < 0.05, ^##P < 0.01 compared with the BLM group.

In the rats dissected on day 28, the levels of IL-1β and IL-6 in the lung changed more significantly as the lung condition developed after molding. The BLM group had the highest levels of IL-1β and IL-6, which decreased after silybin inhalation. The serum levels of IL-1β in the BLM group were significantly higher than those in the control group (P > 0.01) and were significantly downregulated after silybin inhalation compared to the BLM group (P > 0.05). There were no significant differences between the groups in lung and serum levels of TNF-α on day 28. All rats had less than 6 pg/mL of TNF-α in serum. Rats in BLM + SLB (15–28 day) group showed a down-regulation of lung TNF-α levels after administration of silybin, with levels closer to those of the control group (P < 0.05).

TGF-β1 is a powerful fibrogenesis cytokine that promotes the growth and differentiation of fibroblasts and plays an important role in the fibrosis process. The lung and serum TGF-β1 levels were significantly higher in the rats of the BLM group than in the control group (P < 0.01) at day 28. After inhaling silybin, both the lung and serum TGF-β1 levels decreased significantly (P < 0.05), and the difference with the control group was not statistically significant.

Genome Analysis and Validation

As shown in Figure 5A, when the difference multiplier was 1.5 (P < 0.05), there were 703 differential genes in the BLM group and the control group, of which 346 genes were upregulated and 357 genes were downregulated. There were 493 differential genes in the BLM + SLB (0–14 days) group and the BLM group, of which 274 genes were upregulated and 219 genes were downregulated. In total, there were 380 differential genes between the BLM + SLB group and the control group, of which 229 genes were upregulated and 151 genes were downregulated. In terms of the number of genetic differences, the rats were more similar to the control group after the inhalation of silybin, with the number of differential genes reduced from 703 to 380. The Venn diagram shows the overlap of 178 genes differing between the BLM group and the control group (Figure 5B). Of these, 100 genes that were reduced in expression because of bleomycin were upregulated by the administration of SLB. At the same time, 72 genes that had increased expression because of bleomycin modeling were reduced by the administration of SLB (Figure 5C). Therefore, the inhalation of silybin had a back-regulatory effect on these genes.

Figure 5.

Gene expression profiles of rat lungs from the control, BLM, and BLM + SLB groups. Sequencing was performed on three samples from each group. (A) Volcano plots of altered differentially expressed genes (DEGs) in BLM versus control, BLM + SLB versus BLM, and BLM + SLB versus control groups of rats. Upregulated and downregulated genes were distinguished according to log2-fold changes and adjusted FDR p-values (P < 0.05). (B) Venn diagram showing the overlap of differential genes between the BLM and the control groups and differential genes between the BLM + SLB and the BLM groups. (C) Heat map of overlapping genes for differential genes between the BLM and the control group and differential genes between the BLM + SLB and the BLM group. (D) Bubble plots of gene ontology (GO) enrichment analysis of overlapping genes for differential genes between the BLM and the control group and differential genes between the BLM + SLB and the BLM group. ELISA measures the levels of matrix metalloproteinase 13 (MMP-13), matrix metalloproteinase 14 (MMP-14), and matrix metalloproteinase inhibitor 1 (TIMP-1) (E–J). Levels of MMP-13 (E), MMP-14 (F), and TIMP-1 (G) in the lungs of rats in the 14-day group (n = 3). Levels of MMP-13 (H), MMP-14 (I), and TIMP-1 (J) in the lungs of rats in the 28-day group (n = 3). Mean ± standard deviation (SD), *P < 0.05, **P < 0.01 compared with the control group; ^#P < 0.05, ^##P < 0.01 compared with the BLM group.

We used gene ontology (GO) to functionally enrich these specific genes and found that they were mainly categorized in terms of response to stress, response to external stimulus, and immune system process. Among the differential genes, we found that αSMA, COL7A1, MMP-13, and MMP-14, which were involved in dysregulated tissue remodeling, were back-regulated after silybin administration. The genes COL17A1, COL6A5, IL2RB, and CCL19, which were involved in fibroblast motility, were reduced after modeling, and the expression of these genes was close to that of normal rat lungs after administration. To further confirm the effect of inhaled silybin, we examined the levels of MMP-13, MMP-14, and TIMP-1 in the lungs of rats. It was found that the MMP-13 content in the lungs of the BLM group was significantly higher than that of the control group at day 28 and was not significantly different from that of the control group after the administration of silybin (Figure 5H). Likewise, silybin did have a back-regulatory effect on MMP-14 (Figure 5I).

Security Analysis

The safety of silybin inhalation to the lung was evaluated by a comparison between the lungs of the control group of rats and the lungs of rats that inhaled silybin 7 times. Results showed that the lungs of both groups of rats were red and had a healthy luster (Figure 6A). No significant differences were seen in the H&E, Masson, and Sirius Red sections of the two groups of rats. The lung fibrosis scores and collagen content of the lung tissue sections of the silybin group were not significantly different from those of the control group as shown in Table 6. The lung weight and lung HYP levels were also not significantly different from the control group. The levels of IL-1β, IL-6, TNF-α, and TGF-β1 in the lungs and serum were also not significantly different from those in the control group.

Figure 6.

(A) Lung photographs and photographs of H&E, Masson, and Sirius scarlet staining of rat lung tissue sections of the control group and the silybin group at 14 days (scale = 100 μm). (B) Cell viability of RLE-6TN cells at silybin concentrations of 2, 20, 50, and 100 μM (n = 4). (C) Cell viability of RLE-6TN cells at 17-AAG concentrations of 50 nM, 100 nM, 1 μM, and 10 μM (n = 4). Mean ± standard deviation (SD), *P < 0.05, **P < 0.01 compared with the control group.

Table 6. Results of Ashcroft Score, Collagen Volume Fraction, Lung Weight, Hydroxyproline Content, IL-1β, IL-6, TNF-α, and TGF-β1 in Rats of the Silybin Group.

test part	test items	SLB group	significant difference with the control group	repeat number
lung	Ashcroft score	0.33 ± 0.50	no	n = 3
lung	collagen volume fraction (%)	5.02 ± 1.17	no	n = 3
lung	lung weight (g)	1.41 ± 0.06	no	n = 3
lung	hydroxyproline content (μg/mg)	0.91 ± 0.04	no	n = 3
lung	IL-1β (pg/mL)	191.43 ± 71.32	no	n = 3
lung	IL-6 (pg/mL)	35.44 ± 6.06	no	n = 3
lung	TNF-α (pg/mL)	164.83 ± 29.53	no	n = 3
lung	TGF-β1 (pg/mL)	190.05 ± 42.87	no	n = 3
serum	IL-1β (pg/mL)	4.86 ± 1.72	no	n = 3
serum	IL-6 (pg/mL)	24.46 ± 13.39	no	n = 3
serum	TNF-α (pg/mL)	1.61 ± 1.84	no	n = 3
serum	TGF-β1 (pg/mL)	6631.94 ± 1470.64	no	n = 3

We further investigated the effect of silybin and 17-AAG on the viability of RLE-6TN. It was found that silybin at doses up to 50 μM had a good safety profile and had no significant effect on the cell viability of RLE-6TN cells. In contrast, 17-AAG had a significant effect on the cell viability of RLE-6TN cells even at a dose of 50 nM (P < 0.05). At doses of 100 nM and above, 17-AAG inhibited cell viability of RLE-6TN cells in several significant manners (P < 0.01).

Discussion

Since IPF requires long-term drug treatment, oral drugs are easy to accumulate in the liver and cause side effects, and it is difficult to meet the requirements of safe and effective drug use after intravenous injection. Based on the above, local pulmonary administration of SLB is expected to improve the therapeutic efficacy of IPF with the advantages of low systemic exposure, low side effects, and administration frequency.¹⁴

In this study, an easily available and economically safe aqueous solution was used as the solvent for silybin spray dry powder. Due to the insoluble nature of silybin, it was also considered whether the addition of ethanol to improve solubility could improve its final lung deposition rate. However, preexperiments revealed no significant advantage in its final lung deposition rate.¹⁵ We prepared a smaller particle size of silybin nanosuspension by ball milling and improved the lung deposition rate of the powder by adding surfactants. We also carried out to optimize the best spray drying process by orthogonal experiments and obtained the best-combined level of pulmonary deposition rate and yield of the spray-dried silybin powder.¹⁶ Afterward, the surfactant amount and type with the best lung deposition rate were selected by single-factor experiments: the addition of 2.5% poloxamer 188 resulted in a powder lung deposition rate of 32.12 ± 2.38% and a yield of 66.80%.¹⁷ It was found that the use of spray drying for the preparation of silybin powder resulted in a more rapid release of the powder with a wrinkled surface.^18,19

We succeeded in establishing a rat model of IPF after bleomycin was nebulized in two separate doses. This modeling approach allowed all rats to survive and more scientific statistics. No rats died in the middle of the experiment and overall the lung damage was less severe, with the lungs appearing partially white in the photographs. According to the characteristics of bleomycin-induced IPF, we conducted a two-stage trial on the therapeutic effect of silybin: the first two weeks and the second two weeks correspond to inflammation and fibrosis stages, respectively. The administration of SLB during the first two weeks of molding was also used to test whether SLB could have a preventive effect during the external damage to the lungs. In addition, collagen and fibrosis reduction were observed in lung sections on day 14 and day 28, respectively. Results showed that there was no significant difference in the levels of hydroxyproline, IL-1β, IL-6, TNF-α, and TGF-β1 between the BLM + SLB (0–14 days) group and the BLM group at day 14. However, there were significant differences in the levels of hydroxyproline, IL-1β, and TGF-β1 between the BLM + SLB (15–28 days) group and the BLM group at day 28. This may be due to the low dose of bleomycin, which leads to less significant changes in related factors during the development of inflammation stage in the first 2 weeks. However, mild but persistent inflammation still led to the development of fibrosis over the next two weeks. It also showed that the therapeutic effect of SLB on IPF was mainly in the fibrotic stage rather than the inflammatory stage.

Interleukins play an important role in transmitting information, activating and regulating immune cells, and mediating T and B cell activation, proliferation, and the inflammatory response.^20,21 We selected IL-1β and IL-6 for testing to further investigate the effects of silybin inhalation on the regulation of immune and inflammatory responses. When IL-1β is present at low local concentrations, it is involved in immune regulation. In addition, when IL-1β is produced in large quantities, it is involved in endocrine effects. IL-6 was originally found to be a cytokine that inhibits viral replication when fibroblasts are stimulated with Poly I-C. IL-6 effectively promotes TNF and IL-1-induced cachexia; promotes glucocorticoid synthesis, among other effects. Silybin controlled the overproduction of factors such as IL-6 and IL-1β, which in turn reduced inflammation.

For the analysis of differential genes, we used a P < 0.01 difference multiplier of 8 to select specific genes, and 16 genes were screened out. Among them, MMP-13, Golga7, and Rrm2 showed a dramatic increase in expression after drug administration. NGP, Defb105A, DEFA5, and Epx decreased significantly after drug administration. Of these, MMPs were found to correlate significantly with IPF, and a further search revealed that silybin also had a significant back-regulatory effect on MMP-14. The remaining genes with large variations were mainly associated with defensins, neutrophils, and eosinophils.

Matrix metalloproteinases (MMPs) and tissue inhibitors of matrix metalloproteinases (TIMPs) are important enzymes that regulate the dynamic balance between degradation and reorganization of ECM components, to eliminate specific signals, reveal implicit signals, or even release or activate bioactive substances already present in the matrix, and thus participating in the IPF process. The search for specific MMPs and TIMPs to maintain a stable equilibrium between MMPs/TIMPs and prevent or delay the onset of IPF is likely to be a new approach to the clinical management of IPF. It was found that silybin affected the content of MMPs and TIMPs. One of the effects of silybin on IPF is achieved by maintaining the MMPs/TIMPs balance.

Silybin is currently taken orally or intravenously, generally three times a day, once a normal 70 mg. The dose of silybin administered to rats in our study was 3 mg/kg, which after conversion is lower than the single dose of 35 mg of silybin administered to humans. The frequency of administration of silybin powder was once every two days by inhalation. Changes in lung weight and collagen scoring of lung sections enable a preliminary judgment to be made about the safety of the mode of administration and the dose administered. There was no previous premise for the direct inhalation administration of silybin through the lungs. Clinical doses should be limited by further toxicological studies.

Silybin, a nonhepatotoxic, de novo biotin-like HSP90 inhibitor, binds CTD to induce changes in HSP90 conformation and alter HSP90 co-chaperone-client interactions. In terms of safety, silybin was far superior to 17-AAG, with no cytostatic effect on normal rat alveolar type II epithelial cells as long as the dose was 50 μM and below. In contrast, 17-AAG had an inhibitory effect on cell viability even at 50 nM. Silybin was assayed to have a regressive effect on TGF-β₁. The effect of silybin on TGF-β may be through a reduction in the formation of the HSP90/TGF-βR complex. The effect of silybin on MMPs may also be produced via the TGF-β pathway.²² These need to be studied in depth in subsequent experiments.

Conclusions

In conclusion, the current study shows that silybin, as a heat shock protein c-terminal inhibitor, possesses a pharmacological effect in reducing IPF and has a good safety profile through pulmonary inhalation. Silybin controls the severity of IPF primarily by inhibiting the overproduction of IL-1β, TGF-β, MMP-13, and MMP-14. These findings could give ideas for the development of drugs for IPF and the selection of drugs with a high safety profile, good adaptability, and some preventive effect. This study also provides additional support for the role of heat shock protein inhibitors in IPF.

Materials and Methods

Materials

Silybin was extracted and isolated from silymarin by our laboratory, consisting of silibinin A and silibinin B in a molar ratio of 1:1, with a purity of more than 98%. The bleomycin hydrochloride injection was provided by Hisun Pfizer Pharmaceuticals Co., Ltd. (Fuyang, China). Methanol for the high-performance liquid phase was the chromatographic grade. All other chemicals (ethanol, Poloxamer 188, Tween80, sodium stearate) were of analytical grade and have been purchased from common suppliers.

Content Analysis of Silybin

According to the detection wavelength of 288 nm, the concentration of silybin was determined by HPLC using Agilent 1260. The HPLC was performed on an RP-C18 column (4.6 mm × 250 mm, 5 μm, Agilent Technologies) using a mobile phase consisting of methanol–water–glacial acetic acid (48: 52: 1, v/v) at a flow rate of 0.8 mL/min.

Preparation of Silybin Dry Powders

To prepare a silybin powder with the highest possible lung deposition rate, we chose the method of spray drying silybin aqueous suspension. Silybin was first configured into an aqueous suspension of 0.5 mg/mL, sonicated for 30 min, and pretreated with a homogenizer at 18,000 rpm for 1 min. The ball-milling method (1500 rpm) or homogenization method (1000 bar) was then performed to check further suspension handling methods for mutual comparison. The suspension was finally spray-dried with a Buqi B-290 spray drier.

Characterization of Silybin Dry Powders

Particle Size Determination

The particle size of silybin nanosuspensions was measured by Malvern Zetasizer Nano ZS90 (Malvern Equipment Co., Ltd., U.K.) at 25 °C. Each sample was analyzed in triplicate.²³

Differential Scanning Calorimetry (DSC) Analysis

For the determination using a differential scanning calorimeter Model 200F3 (NETZSCH, Germany), about 5 mg of powder is added to an aluminum crucible and the change in thermal effect was measured using a blank crucible as a reference. The procedure was set as follows: the sample was heated up from −20 °C under nitrogen purge, raised to 250 °C, stayed for 4 min, and then cooled down. The heating rate was 10 °C/min. The obtained results were analyzed by the software Proteus Analysis.²⁴

Determination of In Vitro Fine Particle Delivery

A simplified impinger is used to detect its percentage of respirable particles. The interior of the artificial throat is also coated with a thin layer of silicone oil to simulate a mucous membrane and the collection disc of the cascade impactor is also covered with silicone oil to reduce particle bounce and secondary entrainment. The simplified impinger is based on the Anderson Cascade Impactor (ACI), which has only two layers corresponding to the ACI’s layer 2 (near 5 μm) and layer 5 (near 1 μm) compared to the Anderson Impactor’s 7 or 8 collection layers.

The powder is weighed precisely to approximately 60 mg and the weighed powder aerosol is evenly filled into 10 gelatin capsules No. 3. The capsules were broken in turn and placed on the nozzle adaptor, which was then fitted to the artificial throat of the aerosol particle size distribution sampler. The gas volume flow rate was set to 28.3 L/min and removed after 30 s of inhalation. The before and after weight of the collection tray and the weight of the broken capsules were measured separately, and the percentage of fine particle fraction (FPF) was calculated.^25,26

Scanning Electron Microscopy (SEM)

Samples of silybin API and silybin spray-dried powder were taken and placed fixed on a sample table and the samples were subjected to vacuum ion spray gold, respectively. The particle size and shape of the samples were observed using a Regulus 8100 scanning electron microscope (Hitachi, Japan).²⁷

X-ray Powder Diffraction (XRD)

The sample was placed in the effective measurement area and scanned at 10–50°, 0.02°/s with a D8 Advance diffractometer (Bruker, Germany), the test results were saved, and the XRD pattern was plotted.²⁸

Determination of Comprehensive Properties of the Powder

The purpose of the stability procedure is to assess whether the powder will change due to flow. During the measurement, the blade passes through the powder at a fixed rate, and the stability of the sample is evaluated by measuring the resistance of the blade. After 7 tests under the same conditions, if the stability of the powder is good, the results of each measurement should be similar. The variable flow rate program is designed to evaluate the fluidity of powders. The powder was initially subjected to the standard flow rate of 100 mm/s, the flow energy was measured, and then the flow rate was decreased and the energy change was measured four times.

The aeration procedure is designed to assess whether the powder flow properties are altered by the powder becoming aerated. No air is supplied for the first measurement, the air is introduced from the second and the test speed is increased one by one to a maximum air speed of 10 mm/s.

The permeability procedure is designed to measure the ease with which a fluid, in this case, air, can pass through a material. The test detects the resistance to the passage of air through the powder by applying pressure.^29,30

Dissolution Studies

The release characteristics of the spray-dried powder of silybin were examined for 24 h and compared with the API of silybin. Samples (10mg) were placed in separate dissolution cups containing 500 mL of phosphate-buffered saline (PBS) containing 1% Tween80 (v/v). The test was carried out using the paddle rotation method at 100 ± 1 rpm and the temperature was maintained at 37.5 ± 0.1 °C. The concentration of silybin in the extracted samples filtered through a 0.45 μm membrane was determined using the HPLC analytical method mentioned in Content Analysis of Silybin Section.³¹

Infrared Spectrum

The iS5 FTIR spectrometer (Thermo Fisher Technology Co., Ltd.) was used to scan the infrared spectra of the sample powder in the range of 4000–600 cm^–1.

Animals and Ethics Approval

The Experimental Animal Centre provided male Sprague–Dawley (SD) rats (age 6 weeks, weight 190–210 g) purchased from Viton Lever (Beijing, China). All animals were placed in a controlled environment with 12 h of light/darkness and were provided with free food and drink. Animal facilities are designated pathogen-free species. The NJUCM Animal Care and Use Committee approved the protocol for all animal experiments (202205A004). All animals were kept for 7 days before the experiments to acclimatize to their new environment. All animal care and experiments comply with national and institutional ethical regulations, and all efforts were made to minimize animal suffering.

Animal Model of IPF and Treatment

To obtain the rat model of IPF, bleomycin was administered via direct intratracheal nebulization at 1.5 U/kg (about 50 μL/rat) by an HRH-MAG4 lung quantification nebulizer (Beijing Huironghe Technology Co., China) on day 0 and day 7, respectively.^32,33 The atomization administration method is referred to our previous article.³⁴ Briefly, the rats were anesthetized and then suspended through their front incisors on an inclined bench. The tongue was then gently squeezed with forceps, and the atomizer was used to spray the drug into the rat’s airway at the moment of opening the vocal valve.

The SD rats were randomly divided into five groups: Group I: the control group inhaled air only (Control); Group II: the model group given bleomycin nebulization on days 0 and 7 (BLM); Group III: the treatment group received bleomycin nebulization and silybin spray-dried powder via pulmonary inhalation from day 1 every other day for a total of 7 doses [BLM + SLB (0–14 Day)]; Group IV: the treatment group received BLM nebulization and silybin spray-dried powder via pulmonary inhalation from day 15 every other day for a total of 7 doses [BLM + SLB (15–28 Day)]; Group V: The silybin group was only given silybin spray-dried powder via pulmonary inhalation from day 1 every other day for a total of 7 doses (SLB). During the treatment period, rats in each group were checked daily and the body weight changes or deaths were recorded.

Silybin dry powder was given at a dose of 3 mg/kg by using a DP-4 Dry Powder Insufflflator (Penn-Century, Inc., Wyndmoor, PA 19038), which is a drug delivery device designed to produce a puff of fine powder from the end of a small-diameter delivery tube. At day 14, all of the rats of the BLM + SLB (0–14 Day) group and SLB group, and half of the rats in the control group and BLM group were euthanized and the serum lung tissue was taken out. At day 28, the other half of the rats in the control group, the BLM group, and all of the rats of the BLM + SLB (15–28 Day) group were euthanized and the lung tissue was removed for further analysis. The collected lung tissue was washed with sterile saline, and whole lung weights were recorded. The middle lobe of the right lung was placed in 4% paraformaldehyde fixative for 24 h, while the isolated lungs were rapidly placed in liquid nitrogen and transferred to a −80 °C refrigerator for storage.

Assay of Hydroxyproline, IL1β, IL-6, TNF-α, TGF-β1, MMP-13, MMP-14, and TIMP-1

At the end of the experiment, rats were euthanized, lungs and serum were extracted, and 100 mg of the right lung was homogenized in 900 μL of PBS (pH 7.4). Hydroxyproline content was determined using the Hydroxyproline Assay Kit (Nanjing Jiancheng Biological Engineering Research Institute, Nanjing, China). Lung hydroxyproline content was quantified per the manufacturers’ instructions, according to the following equation:

1

The levels of interleukin 1β (IL-1β), interleukin 6 (IL-6), tumor necrosis factor-α (TNF-α), and transforming growth factor-β1 (TGF-β1) were measured in lung tissue and serum using commercially available ELISA kits. (Multi Sciences, China).

Matrix metalloproteinase 13 (MMP-13), matrix metalloproteinase 14 (MMP-14), and matrix metalloproteinase inhibitor 1 (TIMP-1) in rat lung tissue were measured using a commercially available ELISA kit. (Nanjing MALLBIO Biological Technology Co., Ltd.).

Lung Histomorphological Analysis and Ashcroft Scoring

The middle lobe of the right lung was immersed in 4% paraformaldehyde fixative for 24 h, and then the samples were paraffin-embedded to obtain 4 μm sections and dehydrated. The sections obtained were stained with hematoxylin–eosin (H&E), Masson’s trichrome, and Sirius scarlet stain to detect fibrosis in the lung samples. Full scans were performed and sections were photographed using the Vectra 3.0 Quantitative Pathology Imaging System. The degree of IPF was determined using the semiquantitative Ashcroft method and collagen volume fraction score. Changes in Masson-stained fibrotic lung samples were assessed semiquantitatively according to the modified Ashcroft method on a scale of 0–8. This was done on all rat lungs, using three slices per lung.³⁵

Cell Viability Assay Experiment

Rat alveolar epithelial cells type 2 (RLE-6TN) cells were inoculated into 96-well plates at 100 μL per well (10,000 cells) and precultured in an incubator at 37 °C and 5% CO₂ for 24 h until the cells were well attached and morphologically normal.

100 μL of different concentrations of the sample to be tested were added to the plate in the form of a fluid change and incubated in the incubator for 24 h. Blank wells are wells with the appropriate amount of cell culture medium and cell counting kit-8 (CCK-8) solution, but no cells were added. Control wells are wells to which cells and CCK-8 solution were added but no drug concentration was present.

Each well was spiked with 10 μL of CCK-8 solution and incubated in the cell culture incubator for 2.5 h before the absorbance value was measured at 450 nm using a Multifunctional microplate reader (EnVision). Cell viability was calculated using eq 2.

2

Transcriptome Sequencing

At day 14, the anesthetized rats were sacrificed and the lungs were removed. According to the manufacturer’s instructions (Invitrogen), total RNA was extracted from tissues using TRIzol reagent (plant RNA purification reagent for plant tissues). Genomic DNA was removed using DNase I (Takara). After monitoring RNA degradation and contamination on 1% agarose gel, RNA quality was determined by a 2100 bioanalyzer (Agilent Technologies) and quantified using ND-2000 (Nanodrop Technologies). Only high-quality RNA samples (OD260/280 = 1.8–2.2, OD260/230 ≥ 2.0, RIN ≥ 8.0,28S: 18S ≥ 1.0, >1μg) were used to construct sequencing libraries. According to the manufacturer’s instructions (Illumina, San Diego, California), library preparation and sequencing RNA purification, reverse transcription, library construction, and sequencing were performed at Majoro Biohazard Biotechnology Co., Ltd. (Shanghai, China).

Differential expression analysis and functional enrichment. To identify differentially expressed genes (DEGs) between two different samples/groups, the expression level of each gene was calculated according to the transcript per million reads (TPM) method.³⁶ In essence, DEGs with |log2 (foldchange)| ≥ 1 and P-adjust ≥ 0.05 are considered to be significantly DEGs. In addition, functional enrichment analysis was performed, including gene ontology (GO, http://www.geneontology.org).³⁷⁻³⁹

Statistical Analysis

All data were characterized as the mean ± S.D. P-values were calculated by one-way ANOVA followed by Bonferroni post hoc test. Mean differences were considered significant when P < 0.05 (*P < 0.05 and ** P < 0.01); (^#P < 0.05 and ^##P < 0.01).

Glossary

Abbreviations

IPF

idiopathic pulmonary fibrosis

SEM

scanning electron microscopy

DSC

differential scanning calorimetry

XRD

X-ray diffraction

FT-IR

Fourier transform infrared spectroscopy

ECM

extracellular matrix

HSP90

heat shock protein-90

17-AAG

tanespimycin

TGF-β1

transforming growth factor-β1

ACI

Anderson cascade impactor

SI

stability index

FRI

flow rate index

PBS

phosphate-buffered saline

IL-1β

interleukin 1β

IL-6

interleukin 6

TNF-α

tumor necrosis factor-α

MMP-13

matrix metalloproteinase 13

MMP-14

matrix metalloproteinase 14

TIMP-1

matrix metalloproteinase inhibitor 1

H&E

hematoxylin–eosin

CCK-8

cell counting kit-8

DEGs

differentially expressed genes

Supporting Information Available

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acsptsci.3c00033.

• Conditions of the LC–MS (Table S1); liquid-phase conditions of LC–MS (Table S2); ¹H NMR spectra of silybin (Figure S1); infrared spectra of silybin (A) and standard infrared spectrum of silybin in WILEY database (B) (Figure S2); nuclear magnetic spectrum of silybin (Figure S3); and high-performance liquid chromatogram atlas of silybin (Figure S4) (PDF)

Author Contributions

^† X.M. and K.X. contributed equally to this work.

This work was funded by a project of the National Natural Science Foundation of China (81873013) and a Project supported by the Science and Technology Innovation Fund of the Dantu District (GY2021001).

The authors declare no competing financial interest.