Antiangiogenesis Roles of Exosomes with Fei-Liu-Ping Ointment Treatment are Involved in the Lung Carcinoma with the Lewis Xenograft Mouse Model

Exosomes display efficient biocompatibility and represent valuable vehicles for drug or effective material delivery in a tumour-therapeutic approach. Following treatment with Fei-Liu-Ping (FLP) ointment, a traditional Chinese herbal formula, which is used for treating lung cancer patients, could inhibit lung carcinoma growth in clinical and animal studies. In the present study, the values of VEGF and PDGF, which were closely related to angiogenesis, were estimated in serum and carcinoma tissue exosomes to unveil the FLP effects on angiogenesis. The common inflammatory factors of IL-6, IL-1β, TNF-α, and TGF-β in serum exosomes were also detected with the Lewis xenograft model. Methods. Male C57BL/6 mice were randomly divided into four groups, namely, normal, model, cyclophosphamide (CTX), and FLP treatment groups. Histological structures were observed and imaged by H&E. CD31 expressions in tumour tissues were detected by immunofluorescence (IF) and western blot (WB). VEGF, PDGF, and PDGFR levels in exosomes, serum, tumour, and lung tissues were detected by enzyme-linked immunosorbent assay (ELISA), immunohistochemistry (IHC), and WB, respectively. IL-6, IL-1β, TNF-α, and TGF-β levels in exosomes were measured by multiplex immunoassay panels. Results. The results showed that FLP had tumour growth inhibition rate (39.31%). CD31 protein expression was obviously decreased in tumour tissues of CTX- and FLP-treated MO mice, compared to that of MO mice (P < 0.05 or P < 0.001). VEGF, PDGF, and PDGFR expression levels with FLP treatment were downregulated in exosomes, serum, tumour, and lung tissues compared to model group (P < 0.05 or P < 0.01). The expressions of IL-6, IL-1β, and TNF-α were downregulated in exosomes compared to the model group (P < 0.05 or P < 0.01). Conclusions. This study suggested that FLP had the ability of inhibiting tumourigenesis in a Lewis lung xenograft mouse model, whose therapeutic mechanisms might relate with the downregulation of angiogenesis factor and tumour inflammatory cytokines levels.


Introduction
Lung carcinoma is the most frequent lethal malignancy and the leading cause of cancer death worldwide, harbouring high incidence in Asian countries [1,2]. Due to the lack of specific early symptoms, the patients with lung cancer are neglected in initial stage and are frequently diagnosed in an advanced stage. Management of lung cancer usually entails radiotherapy, chemotherapy, targeted therapy, and immunotherapy following surgery [3]; however, the five-year survival rate is only approximately 25-30% [4,5]. And, thus, it is imperative to develop novel antitumour drugs or methods for alleviating or postponing the progress of lung cancer. Traditional Chinese medicine (TCM) has been shown to be effective in the treatment of lung cancer, especially in decreasing the risk of tumour recurrence and metastasis [6,7]. Using systematic reviews and meta-analyses, scholars have recently reported that Chinese herbal medicine could improve survival, quality of life, and immune suppression and reduce tumour recurrence and metastasis in lung cancer patients by alleviating immune suppression [7,8].
According to TCM theory, the basic pathogenesis of lung cancer involves the deficiency of qi and yin, damage of collaterals in lung, and turbid phlegm obstruction of lung. Therefore, a TCM formula named Fei-Liu-Ping (FLP) ointment (FLP ointment, currently named Yifei Qinghua granules), which was created by Prof. Bingkui Piao, has been used for more than 30 years [9]. FLP ointment is a yinnourishing, toxin-removing, and blood-activating formula of TCM. Previous clinical studies have shown that FLP ointment possessed many marked anticancer properties, such as relieving the manifested symptoms, ameliorating the side effects of radiotherapy and chemotherapy, and reducing the dose of medication required, thus improving the quality of life in lung cancer patients [8][9][10]. FLP ointment is produced by the Beijing Huashen Pharmaceutical Co., LTD. as a hospital preparation (SFDA approval number Z20050851).
Although previous studies indicated that FLP ointment inhibits angiogenesis, the role of exosomes with FLP treatment, especially circulating exosomes in mediating tumour vasculogenesis, is unclear. To elucidate the mechanisms of FLP inhibiting vasculogenesis and to address the role circulating and carcinomatous tissue exosomes play, the values of vascular endothelial growth factor (VEGF), platelet-derived growth factor (PDGF), and PDGFR which were closely related to angiogenesis, were estimated in serum and carcinomatous tissue exosomes. The above proteins were also detected in carcinomatous tissue. Common inflammatory factors, such as IL-6, IL-1β, TNF-α, and TGF-β in serum exosomes, were also measured in this study.
Extracellular microenvironment around lung carcinoma can contribute to the increase in the tumour metastasis and the grade malignancy. Exosomes are lipid-bilayer membrane vehicles that range from 30 to 100 nm in size and can be released into extracellular microenvironment by almost all cell types [11]. Although exosomes were previously considered as dumped vehicles from adjacent cells, their intercellular communication abilities have been proved, especially in carcinoma [12]. In the past decade, extracellular vesicles including apoptotic small body, microvesicles, and exosomes emerged as critical players in cell to cell communications in physiology and pathology [13]. Exosomes are tens nm in size, which allows genetic and molecular exchanges, including the transfer of proteins such as VEGF, PDGF, and PDGFR at a distance to target cells.  (Table 1). Briefly, all herbs were provided by the Guang'anmen Hospital and decocted twice with eightfold volume of distilled water for 1 h. The decoction was collected, filtered, merged, concentrated to 2 g/mL (equivalent to crude herb materials), and stored at 4°C for oral use. To ensure the quality and stability of FLP ointment, ultra-high-performance liquid chromatography combined with LTQ-Orbitrap mass spectrometry (UHPLC-Orbitrap MS, Thermo Fisher Scientific, San Jose, USA) was used to identify the active ingredients.

UHPLC-Orbitrap MS Analysis of FLP Ointment.
The analysis of the extract was performed with a UHPLC-Orbitrap MS with a Thermo BDS HYPERSIL C18 column (2.1 mm × 150 mm, ID 3 μm). The column temperature was set at 30°C. The mobile phase was water (0.1% formic acid, phase A) and acetonitrile (phase B) with a gradient programme as follows: 0.0-25.0 min, 5%-30% B; 25.0-35.0 min, 30%-55% B; 35.0-40.0 min, 55%-75% B; 40.0-45.0 min, 75%-90% B; and 45.0-50.0 min, 90% B. The flow rate was 0.3 ml/min. All solvents were filtered through a 0.45 μm filter before use. The injection volume was 10 μL. The MS spectra were acquired in the positive ion mode with an electrospray ionization source (ESI+). The ESI source conditions were as follows: heater temp 350°C; sheath flow rate 35 arb; aux gas flow rate 10 arb; spray voltage 3.5 kV; capillary temp 350°C; and resolution 30000. Full-scan spectra were acquired in the mass range of m/z 50-1000. The main chemical structures and Mass Spectrogram which may have pharmacoactivity of FLP ointment were shown in Figure 1. Components of FLP ointment identified and HPLC chromatogram were shown in Table 2 and Figure 2, respectively. Additionally, the

Model Establishment and Experimental Treatment.
After 5 days of acclimation, the mice were randomly divided into 4 groups: the normal control (NC), model (MO), CTXtreated MO, and FLP-treated MO groups (n � 10 mice per group). 12 g/kg/day dose of FLP ointment used in the present study was shown to have more antitumour effects in Lewis lung carcinoma xenograft mouse in our previous study [14,15]. A subcutaneous injection suspended in 0.2 mL PBS with 2 × 10 5 Lewis cells was implanted into the right flank of each MO mouse. The mice in the CTX-treated MO group were treated with 60 mg/kg CTX 24 h after the subcutaneous injection of the Lewis cells, as described below.
The mice in the FLP-treated MO group were treated with suspensions of 12 g/kg/day FLP by gavage. The mice in the NC and MO groups were treated with normal saline. The CTX mice were injected only once, and the NC, MO, and FLP-treated mice were treated once per day for 21 days.

Sample Collection.
The mice were sacrificed on day 21. Briefly, peripheral blood was collected via retro-orbital eye bleeds. Serum from each mouse was collected into 1.5 ml sterilised micro-tubes, kept on ice for 2 h, centrifuged at 4°C at 3000 rpm for 15 min, and stored at −80°C until further analysis. The lung and tumour tissues were immediately frozen for WB analysis, and 4% paraformaldehyde was used to fix the lung and tumour tissues for haematoxylin-eosin (H&E) staining, immunofluorescence (IF), and immunohistochemical (IHC) analysis. The volume of tumours was measured daily from day 8 until the mice were sacrificed and the tumour volume was calculated using the formula π * (length) * (width) 2 /6.     2.11. WB Analysis. The exosome proteins were extracted after homogenization with protein extraction agent (50 mmol/L Tris-HCl, pH 7.4; 10% SDS; 1 mmol/L PMSF, 0.25% deoxycholic acid sodium) for 30 min on ice. The samples were then centrifuged at 12,000 rpm at 4°C for 10 min, supernatants were collected, and protein concentrations were measured by a BCA assay. The lung and tumour proteins were extracted and measured as described above. Protein bands were separated by SDS-polyacrylamide gel electrophoresis (PAGE) and subsequently transferred onto nitrocellulose membranes (Millipore). Membranes were probed with primary rabbit anti-CD63 (1 : 2,000), rabbit anti-Alix (1 : 2,000) antibody, rabbit anti-CD31 (1 : 5,000) antibody, rabbit anti-VEGF antibody (1 : 5,000), rabbit anti-PDGF antibody (1 : 5,000), rabbit anti-PDGFR antibody (1 : 5,000), rabbit polyclonal anti-GAPDH antibody (1 : 10,000), or rabbit polyclonal anti-actin antibody (1 : 10,000) in blocking solution at 4°C overnight. Then, the membranes were incubated with secondary goat anti-rabbit IgG (1 : 7,000) for 60 min and electrochemiluminescence (ECL) reagent for 30 sec to 2 min. The membranes were exposed to X-ray film and photographed with a BIO-RAD ChemiDoc XRS gel imaging system. The pictures were exported by using Quantity One software and were analysed by using Image-Pro Plus 4.5 software (Media Cybernetics, Bethesda, MD, USA).

ProcartaPlex
Immunoassays. ProcartaPlex multifactor detection was performed using magnetic beads that could be measured by means of fluorescent staining. The magnetic beads for different detection substances were mixed, and then the samples were added. The capture molecules and the target molecules were specifically bound in the suspension. The plates were read by a Luminex instrument for classification and quantification.

Statistical
Analysis. The data were expressed as the mean standard error (SE) and were analysed with SPSS version 21.0 (SPSS Inc., Chicago, IL, USA). The test of normality was conducted by one-way ANOVA with a post hoc LSD test; otherwise, a rank-sum test was performed. A value of P < 0.05 was considered to indicate statistical significance.

The Tumour Volume of Mice with a Lewis Lung Xenograft.
The effect of FLP ointment was tested in a Lewis lung xenograft mouse model generated by Lewis cell line. The mice were treated for 21 days from the day Lewis cells were implanted in all MO mice. The volume of the tumours was measured daily from day 8 until the mice were sacrificed. On day 21, the mice injected with Lewis cell were sacrificed; the mean volume of the tumours in the CTX-treated group was 130.28 ± 14.82 mm 3 and in the FLP-treated group was 147.21 ± 14.33 mm 3 , significantly lower than that in the control group which was 182.28 ± 17.72 mm 3 (P < 0.05, Figure 3).

Treatment with FLP Ointment Inhibits Tumour Growth of Lewis Lung Cancer Xenografted Mice.
The mice were sacrificed on day 21, and tumour and spleen tissues were measured and weighed. The average tumour weight of MO mice was 1.00 ± 0.15 g. Compared to that of MO mice, the average tumour weight was significantly lower in FLPtreated and CTX-treated mice (P < 0.05, P < 0.01; Figure 4(a)). The average spleen weight of MO mice was 0.24 ± 0.01 g. Compared to that of MO mice, the average spleen weight was significantly lower in FLP-treated mice (P < 0.05; Figure 4(a)). The FLP inhibition rate was 39.31%, and the CTX inhibition rate was 57.29%. Based on the abovementioned results, we found that FLP ointment could inhibit the tumour growth of Lewis lung cancer cells.  Evidence-Based Complementary and Alternative Medicine

Pathological Changes in Tumour
Tissues. The pathological changes in the tumours of mice were observed by light microscopic evaluation of H&E-stained tissues. The tumour cells were large and arranged in a disordered manner. The nuclei were intensely stained, and most of the cells were divided in the MO group. In addition, MO mice showed large tumour cells that were arranged in disordered structures and a large number of cellular atypia structures. Conversely, treatment with CTX increased the number of necrotic cells. FLP-treated MO mice exhibited sporadic necrotic cells, cell membranes dissolved, and the nucleolus disappeared ( Figure 5).

IF and WB Analysis of CD31 Proteins in Tumour Tissues.
Microangiogenesis plays an important role in tumour growth and angiogenesis. CD31 was expressed in vascular endothelial cells, which could be considered as a potential marker of microangiogenesis. The expressions of CD31 in tumour tissues were detected by IF and WB. The fluorescence picture showed the high expression of CD31 in the tumours of MO mice. And when the CTX-and FLP-treated MO mice were observed, the expressions were significantly reduced. For the further study, we detected CD31 by WB. Results showed that CD31 protein expression was obviously decreased in tumour tissues of CTX-and FLP-treated MO mice, compared to that of MO mice (P < 0.001 and P < 0.05, respectively, Figure 6).

WB Analysis of CD63 and Alix Proteins in Serum
Exosomes. Exosomes are a type of membranous vesicles that form from late endosomal or cytosolic budding. Exosomes usually carry proteins and miRNAs from the parent cells. As the marker proteins of exosomes, CD63 and Alix proteins were analysed by WB. The results showed which proteins the NC-EXO, MO-EXO, and FLP-EXO expressed. Compared to that of NC mice, CD63 and Alix protein expressions were increased in model mice and FLP mice (Figure 7). No significant correlation of CD63 and Alix protein expression levels were detected between MO and FLP mice.

ELISA Analysis of VEGF, PDGF, and PDGFR in Serum
and Serum Exosomes. Tumour growth, invasion, and metastasis rely on angiogenesis. VEGF is a crucial angiogenic growth factor, and its level is a critical marker of angiogenesis. The PDGF/PDGF receptor signalling pathway plays a pivotal role in angioblast differentiation and tube formation; therefore, the expression of VEGF, PDGF, and PDGFR was analysed.
ELISA analysis demonstrated that VEGF, PDGF, and PDGFR protein expression in serum were significantly increased in MO mice compared to those of NC mice (P < 0.01 and P < 0.05, respectively). In contrast, the expression levels of these proteins were significantly reduced in serum in FLP-treated MO mice compared to those of MO mice (P < 0.001, P < 0.01 and P < 0.05, respectively). In  Evidence-Based Complementary and Alternative Medicine addition, compared to those of MO mice, PDGF and PDGFR protein expression were obviously decreased in the serum of CTX-treated MO mice (P < 0.01 and P < 0.05, respectively, Figure 8(a)). VEGF, PDGF, and PDGFR protein expression were increased significantly in serum exosomes of MO mice compared to those of NC mice (P < 0.01). Additionally, VEGF and PDGF protein levels were significantly decreased in FLP-treated MO mice compared to those of MO mice (P < 0.05). CTX significantly decreased PDGF and PDGFR protein in serum exosomes of MO mice compared to those of MO mice (P < 0.05, Figure 8(b)).

IHC Analysis of VEGF, PDGF, and PDGFR in Lung and
Tumour Tissues. To further confirm the FLP ointment functions, IHC experiments were carried out. VEGF, PDGF, and PDGFR protein expression were analysed with IHC analysis. The results showed that VEGF and PDGF protein expression were significantly increased in the lungs of MO mice compared to those of NC mice (P < 0.01 and P < 0.001, Figure 9(a)). Treatment with CTX significantly decreased the VEGF, PDGF, and PDGFR levels in the lung compared to those of MO mice (P < 0.05 and P < 0.01, Figure 8(a)). VEGF and PDGFR protein levels in lung were significantly decreased in FLP-treated MO mice compared to those of MO mice (P < 0.01, Figure 9(a)). In addition, compared to those of MO mice, VEGF, PDGF, and PDGFR protein expression were obviously decreased in tumour tissues of CTX-treated and FLP-treated MO mice (P < 0.001, P < 0.01 and P < 0.05, respectively Figure 9(b)).

WB Analysis of VEGF, PDGF, and PDGFR in Lung and
Tumour Tissues. Moreover, VEGF, PDGF, and PDGFR protein expression were analysed with WB. The results showed that VEGF and PDGF protein expression were significantly increased in the lungs of MO mice compared to those of NC mice (P < 0.05 and P < 0.01, Figure 10(a)). Treatment with CTX or FLP significantly decreased VEGF, PDGF, and PDGFR levels in lung and tumour tissues compared to those of MO mice (P < 0.001, P < 0.01 and P < 0.05, Figures 10(a) and 10(b)).
3.9. ProcartaPlex Immunoassay Analysis of IL-6, IL-1β, and TNF-α in Serum Exosomes and Tumours. Compared to that of MO mice, the level of IL-1β was significantly decreased in the tumours of CTX-and FLP-treated MO mice (P < 0.001 and P < 0.01, respectively). TNF-α levels did not exhibit obvious differences in the tumours of MO mice compared to those of CTX-treated mice; however, TNF-α levels were obviously decreased in FLP-treated MO mice compared to those of CTX-treated mice (P < 0.05). There were no significant differences in the IL-6 levels among the 3 groups (Figure 11(a)).
Compared to those of NC mice, the levels of IL-6, IL-1β, and TNF-α were decreased in the serum exosomes of MO mice. The IL-1β and TNF-α levels were markedly increased in serum exosomes of MO mice compared to those of NC mice (P < 0.05). Additionally, IL-1β and TNF-α expression were significantly decreased in CTX-treated and FLP-treated MO mice compared to those of MO mice (P < 0.05). FLP ointment significantly decreased IL-6 levels in serum exosomes of MO mice compared to those of untreated MO mice (P < 0.05, Figure 11(b)).

Discussion
The development of antitumour medicine is a hot field all the time in therapeutic tumour strategy in the last decades [16]. Even though many targeted antitumour medicines have been applied to clinic, the resistance during taking medicine remains to be the critical factor of chemistry therapy failure. And postsurgical adjuvant chemotherapy is commonly used to reduce the risk of lung carcinoma recurrence. However, the biological heterogeneities in lung carcinoma reduce the chemotherapeutic efficiency. FLP ointment, a traditional Chinese herbal formula, could inhibit lung cancer invasion by regulating the tumour inflammatory microenvironment through the NF-κB signalling pathway in the previous study [14]. In combination with celecoxib, FLP ointment enhanced the antitumour growth and antimetastatic effects and may have been associated with the cyclooxygenase (Cox)-2 pathway [15].
The results of this study showed that the tumours of Lewis lung carcinoma xenografted mice were pathologically characterised by large cells with a disordered arrangement, shrinkage and fission, and nuclei that were strongly stained and lysogenic, which is consistent with previous reports [14,15]. Compared with Lewis lung carcinoma xenografted mice, mice treated with FLP ointment manifested reduced tumour weight, and increased necrotic cells in the tumours. Combined with our previous studies, these observations suggested that FLP ointment inhibited tumour growth in Lewis lung xenografted mice.

Evidence-Based Complementary and Alternative Medicine
To characterise the exosomes, CD63 and Alix proteins were measured in the serum. Exosomes are released by a variety of cells, such as tumour cells and immune cells [17].  between cells [18]. Exosomes can enter the lymphatic system and capillaries within the tumour tissue and play roles in immune evasion and the regulation of T cells and NK cells [19,20]. The exosomes interact with tumour suppressors and participate in the regulation of tumour growth [21]. An increasing number of studies have noted that exosomes provide stability and transport proteins and nucleic acids into specific target cells, thus playing a critical role in various aspects of the formation of the tumour microenvironment, invasion, metastasis, and angiogenesis [11,22]. The characteristically expressed exosome proteins were TSG101, Alix, CD9, and CD63. In this study, we examined the expression of CD63 and Alix proteins in the NC-EXOs, MO-EXOs, and FLP-EXOs, which expressed these proteins. The present study found that FLP ointment could increase CD63 and Alix protein levels compared to those of MO mice, which illustrated that FLP ointment could interfere exosomes in serum. FLP ointment may increase the effects of signal transmission through the high levels of exosomes in serum. The endothelial protein CD31, expressed in vascular endothelial cells, could be considered as a potential marker of microangiogenesis [23]. The present study showed that FLP could downregulate the CD31 expression in tumour tissues and this indicated that FLP could reduce neoangiogenesis. As angiogenesis was restricted by FLP treatment, VEGF and PDGF levels in the serum, exosomes, lung, and tumour tissues were measured in the present study. Angiogenesis is mainly regulated by several different growth factors. The VEGF family and its receptors are likely to be the most important tissue factors responsible for angioblast differentiation and lymphangiogenesis, which activates the tyrosine kinase receptors of endothelial cells and induces endothelial cell growth and invasion [24,25]. PDGF and its receptor play an important regulatory role in tumour angiogenesis, stimulating vascular endothelial cell growth and chemotaxis and contributing to tumour angiogenesis and remodelling [26,27]. A previous study found that FLP ointment inhibited angiogenesis and inhibited the expression of VEGF and PDGFR proteins in the lung tissues [15]. The results of that study agreed with our IHC and WB analyses. Treatment with FLP significantly decreased VEGF, PDGF, and PDGFR levels in lung and tumour tissues. In this study, we found that the levels of VEGF, PDGF, and PDGFR in the serum were increased in mice with lung carcinoma xenografts compared to those of NC mice. FLP ointment decreased VEGF, PDGF, and PDGFR protein expression in the serum compared to that of MO mice. Research has shown that exosomes secreted by HepG2 cells activated a receptor for immune cells and promoted human umbilical vein endothelial cell angiogenesis through the transfer of miRNAs [28]. Upon incubation with exosomes from celecoxib-treated lung cancer cells, the levels of VEGF produced by THP-1 cells were increased [29]. Meanwhile, the expression levels of VEGF, PDGF, and PDGFR in the serum exosomes of Lewis lung carcinoma xenografted mice were significantly increased compared with those of normal mice. CTX and FLP ointment both individually decreased VEGF, PDGF, and PDGFR expression. This result implies that FLP ointment inhibited angiogenesis.
In the process of tumourigenesis and progression, chronic inflammation has become a recognised risk factor for epithelial-derived carcinoma. The environment is composed of tumour cells, interstitial cells, immune cells, cytokines, etc., which make up a dynamic microenvironment called the tumour microenvironment (TME) [30]. The TME is closely related to angiogenesis, which mediates the response to antiangiogenesis drugs and increases the risk of tumour recurrence and metastasis after antiangiogenic therapy [31,32]. Cancerous cells can influence the TME by releasing extracellular signals, promoting tumour angiogenesis, and inducing peripheral immune tolerance, while the TME can affect the proliferation and migration of cancerous cells [33][34][35]. Inflammatory cytokines play a key role in shaping the TME [36]. This study examined cytokine levels and found that the levels of inflammation-related cytokines, such as IL-6, IL-1β, and TNF-α, were decreased in FLP-treated MO mice in the tumour tissues compared to those of MO mice. These results were in accordance with previous studies, which showed that FLP ointment could decrease the expressions of IL-6, IL-1β, and TNF-α in the serum. Inflammatory cytokines of IL-6 family including IL-6 have been implicated in the migration and invasiveness of human lung carcinoma. And VEGF as a potent angiogenic vascular sprouting could be promote by IL-6 via activation of transcription factors HIF1α and STAT3 (gp130 Janus-activated kinase/signal transducer and activator of transcription 3 pathway) [37][38][39][40]. IL-1β and TNF-α are another two inflammatory mediators that could activate their responsive receptors (gp130 and TNFR1) to be involved the NF-κB transcription factor [39,41]. NF-κB activates VEGF gene expression binding to NF-κB sites in the promoter regions of VEGF gene. Inhibitors and activators of NF-κB could decrease and increase VEGF levels in carcinoma, respectively [42]. TGF-β, one role in the pathogenesis of carcinoma, could regulate VEGF release in an ALK-5-dependent manner by SMAD2/3 versus SMAD1/5/8 signalling [43]. In this study, inflammatory cytokines that could enhance the tumourigenic process such as IL-6, IL-1β, and TNF-α were decreased in FLPtreated MO mice in the tumour tissues and serum compared to those of MO mice, which is in accordance with previous studies. Furthermore, the present study analysed the levels of IL-6, IL-1β, and TNF-α in exosomes. The expression levels of IL-1β and TNF-α were significantly increased in exosomes of MO mice compared to those of NC mice, while CTX and FLP ointment could markedly reduce the levels of IL-6, IL-1β, and TNF-α compared to those of MO mice. This indicated that FLP ointment significantly decreased the expression of IL-6, IL-1β, and TNF-α in exosomes, serum, lung, and tumour tissues. These results gave a clue that FLP could downregulate tumour inflammatory cytokines levels to lessen the tumourigenic process.
The chemical structures and Mass Spectrogram of ginsenoside Rg3, liquiritin, calycosin, astragaloside VI, and formononetin were shown in Figure 1. Many of the herbal medicine components in FLP ointment have antitumour effects. For example, ginsenoside Rg3 (Radix Panacis Quinquefolii) inhibits tumour migration and invasion through the PI3K-AKT signalling pathway [44] and may improve the sensitivity to cisplatin during lung cancer therapy [45]; liquiritin (Radix Glycyrrhizae) induces apoptotic cell death by upregulating p53 and p21 in A549 non-small cell lung cancer cells [46]; and calycosin, astragaloside VI, and formononetin (Radix Astragali) suppress A549 cell proliferation and migration through the induction of cell cycle arrest and apoptosis via the PKC-α/ERK1/2 pathway and the modulation of macrophage polarization through AMPK signalling [47][48][49]. Thus, these components are the basis for the inhibitory functions in lung cancer.

Conclusions
The present study suggested that FLP could inhibit tumourigenesis in a Lewis lung xenograft mouse model. Downregulation of tumour inflammatory cytokines and VEGF levels in exosomes, serum, and tumour tissues were involved in carcinoma inhibition. The limitation of this study was that angiogenesis and inflammation indicators had been just detected and following studies will help to investigate the relationship between angiogenesis factors and inflammation cytokines, and the mechanisms of FLP on antiangiogenesis with lung carcinoma should be studied more deeply.