Network Pharmacology Analysis of Traditional Chinese Medicine Formula Shuang Di Shou Zhen Tablets Treating Nonexudative Age-Related Macular Degeneration

Objective To analyze the pharmacological mechanism of the treatment of dry age-related macular degeneration (dry AMD) based on a network pharmacological approach of Shuang Di Shou Zhen Tablets (SDSZT) and to provide a new reference for the current lack of effective treatment of dry AMD. Methods The main chemical constituents and their targets of Rehmanniae Radix Praeparata, Ligustrum lucidum, Mori Fructus, Paeonia albiflora, Rhizoma Dioscoreae, Alisma orientale, Schisandra chinensis, Radix Polygoni Multiflori Preparata, Ophiopogon japonicus, and Radix Rehmanniae were obtained from the Traditional Chinese Medicine Systems Pharmacology Database and Analysis Platform (TCMSP) and Traditional Chinese Medicine Integrated Database (TCMID). The active ingredients of traditional Chinese medicine were screened according to Absorption, Distribution, Metabolism, and Excretion (ADME), the gene names of the targets of each active ingredient were obtained from the Uniprot database, the main targets of dry AMD were obtained from GeneCards and DisGeNET database, and the protein interaction analysis was performed on the String database. The Metascape database was used to analyze the “drug-component-target” and the biological processes and networks involved, and then, Cytoscape 3.8.1 was used to construct the “ SDSZT component-dry AMD target-pathway” network. Results The main active ingredients of SDSZT for dry AMD treatment are quercetin, kaempferol, luteolin, β-glutamine, β-carotene, etc. And, the core targets are RAC-alpha serine/threonine-protein kinase (AKT1), prostaglandin G/H synthase 1 (PTGS1), tumor necrosis factor (TNF), transcription factor AP-1 (JUN), apoptosis regulator Bcl-2 (BCL2), caspase-3 (CASP3), phosphatidylinositol 4,5-bisphosphate 3-kinase catalytic subunit gamma isoform (PIK3CG), androgen receptor (AR), apoptosis regulator BAX (BAX), etc. The biological pathways for the treatment of age-related macular degeneration by SDSZT mainly act on pathways in cancer, fluid shear stress and atherosclerosis, and TNF signaling pathway, and the main function of SDSZT is to regulate intracellular cytokine receptor binding. Conclusion This study initially reveals the multiconstituent, multitarget, and multipathway mechanism of action of SDSZT in the treatment of dry AMD and provides the basis for the clinical application of SDSZT.


Introduction
Age-related macular degeneration (AMD) is one of the most common eye diseases in ophthalmology and is one of the leading causes of blindness in the elderly population [1]. It can be divided into nonexudative AMD (dry AMD) and neovascular AMD (wet AMD). Existing anti-vascular endothelial growth factor (anti-VEGF) is known to be effective in the treatment of wet AMD; however, there is a lack of definitive and effective therapeutic measures for dry AMD. In recent years, there have been numerous studies on the treatment of dry AMD, such as complement factors [2], antiinflammatory therapy [3], acupuncture [4], and traditional Chinese medicine [5,6]. However, none of these can be considered definitive and effective treatments. e advantages of traditional medicine in treating AMD should not be overlooked. e treatment mechanism of traditional Chinese medicine is not only to a single target, but is also based on the holistic view of traditional Chinese medicine to provide a multitarget, multifaceted, and comprehensive treatment of the disease, with high safety. Network pharmacology is the use of known drug components, component targets, and known targets of disease to predict the likelihood of a drug's potential to treat a disease, and the network is able to comprehensively respond to possible mechanisms of drug interventions. rough multipathway modulation of signaling pathways, network pharmacology, as a new tool for drug research, can effectively tap into the active ingredients of traditional Chinese medicine, discard toxic ingredients, and treat the corresponding diseases with maximum stability and precision.
Shuang Di Shou Zhen Tablet (SDSZT) is a self-made ophthalmic medicine for Longhua Hospital, Shanghai. For many years, SDSZTs have been used in ophthalmology clinics to treat age-related eye diseases (deficiency of liveryin and kidney-yin type: e deficiency of liver and kidney in traditional Chinese medicine, which means patients with dim eyesight, metamorphosis, drusen, and pigmentary disturbance in macula, accompanied by general symptoms of dizziness, tinnitus, lassitude in loin and legs, blackish tongue with thin coating, and thready thin pulse). In preliminary clinical observations, it was found that treatment with vitamin C and vitamin E combined with SDSZT resulted in a smaller number of Amsler table deformation frames and a smaller extent of central visual field defects in patients with dry AMD compared to vitamin C and vitamin E alone [6,7]. is formula contains 10 herbs: Rehmanniae Radix Praeparata, Ligustrum lucidum, Mori Fructus, Paeonia albiflora, Rhizoma Dioscoreae, Alisma orientale, Schisandra chinensis, Radix Polygoni Multiflori Preparata, Ophiopogon japonicus, and Radix Rehmanniae, of which Rehmanniae Radix Praeparata and Ligustrum lucidum are popular herbs for research on their antioxidant, anti-inflammatory, and lipid metabolizing components in recent years [8][9][10][11][12]. erefore, this study, based on network pharmacology approach, explores the active ingredient targets of SDSZT and the molecular mechanism of its treatment for dry AMD disease and provides a certain theoretical basis for the follow-up study.

Collection of Known AMD-Related Targets.
Using "dry age-related macular degeneration" and "nonexecutive agerelated macular degeneration" as keywords, we explored the GeneCards database [19] (http://www.genecards.org) and DisGeNET database [20] (https://www.disgenet.org) for potential targets related to dry AMD. Although higher scores in the GeneCards database indicate a strong association between the target and the disease, there were too few relevant targets for dry AMD in several databases (1056 relevant targets in the GeneCards database and only 12 relevant targets in the DisGeNET database) and therefore no targets greater than the median were set as potential targets here. After merging the targets in the two databases, the potential targets for dry AMD can be obtained by deleting the duplicate targets.

SDSZT-Dry AMD PPI Data.
To clarify the interactions between SDSZT-associated targets and dry AMD-related targets, we first used the Venny 2.1.0 (https://bioinfogp.cnb. csic.es/tools/venny/) to get intersection targets. e acquired targets are then used to construct a protein-polymer interaction (PPI) network model from the STRING 11.0 database [21,22] (http://string-db.org). e biological genus is set to "Homo sapiens" and the minimum required interaction score is set to "highest confidence" (>0.900); the rest are default settings to generate the PPI network diagram. Using Cytoscape 3.7.1, the network is analyzed by CytoNCA to calculate the betweeness centrality (BC), closeness centrality (CC), degree centrality (DC), and the eigenvector centrality (EC); the mean value is calculated and then filtered.
e PPI network was further analyzed using the MCODE [23] plugin in Cytoscape 3.7.1.

Gene Ontology and Pathway Enrichment Analysis.
e combination of functional enrichment, interactomic analysis, gene annotation, and search in Metascape is an effective tool for comprehensive analysis and interpretation of OMICs-based research in the era of big data [24]. erefore, SDSZT-dry AMD cross-targets were imported into Metascape (http://metascape.org/gp/index.html), the P value < 0.01 was set, and enrichment analysis was performed. Using the Metascape database and the Microbiology Letter Platform (http://www.bioinformatics.com.cn/), we have conducted a number of studies on molecular function (MF), biological process (BP), and cellular component (CC). e top 20 functionally annotated catalogs and signaling pathways are selected for analysis, and the results are presented as bar and bubble plots.
2.5. SDSZT-Dry AMD-Pathway PPI Data. Cytoscape 3.7.1 was used to construct a network diagram of the constituents of SDSZT-dry AMD target-pathway network, and Cytoscape 3.7.1 built-in tools were used to analyze the network topology parameters of the active ingredients and targets, including Degree, BetweennessCentrality, and Close-nessCentrality. e network topology parameters were then used to determine the core targets and the main active ingredients that exerted drug effects.  12 for Ophiopogon japonicu, and 4 for Radix Rehmannia. After merging and deleting duplicate values, a total of 247 drug targets were collated. We then created a network diagram of the SDSZT herbs-active ingredient-drug target ( Figure 1).

Collection of Known Dry AMD-Related
Targets. 1056 dry AMD targets were obtained from the GeneCards database, and due to the small number of targets obtained, the score range was not set for screening. In combination with the DisGeNET database, 11 targets were obtained for nonexudative age-related macular degeneration, and 1059 dry AMD targets were obtained for dry AMD after deleting the duplicates.

SDSZT-Dry AMD PPI Data.
e screened targets of SDSZT active ingredients and dry AMD disease were imported into the online Venny analysis website, and the intersection targets were plotted in the Venny plot, resulting in 107 common targets of the ingredients of SDSZT and dry AMD obtained ( Figure 2). e obtained intersection targets were imported into the STRING 11.0 platform, and the PPI network of SDSZT was obtained, with 2210 interacting edges and an average node degree value of 41.3 (Figure 3(a)). In order to further clarify the potential targets and proteinprotein interactions at key nodes, the CSV files obtained in STRING 11.0 were imported into Cytoscape 3.7.1 software, and the network was analyzed by cytoNCA. e calculated values of BC, CC, DC, and EC were 0.776567196, 0.623529412, 42, and 0.091691375, respectively. Fifty-three core targets were selected based on greater-than-mean values (Figure 3(b)). Among them, ALB, AKT1, IL6, TNF, TP53, VEGFA, CASP3, PTGS2, JUN, and MAPK1 are the top 10 targets of degree value. ere are tightly connected regions in the PPI network, which may represent molecular complexes called module [25], and the MCODE plugin in Cytoscape 3.7.1 is used to mine the interacting tightly connected regions in the PPI network to obtain the modules (Figure 3(c)).

Gene Ontology and Pathway Enrichment Analysis.
e signaling pathways of the targets associated with SDSZT for dry AMD were obtained from the analysis of the Metascape website. e results were then visualized by Cytoscape 3.7.1 software and the Microbiology platform. e result is a target that is inextricably linked to dry AMD.
e main biological processes involved in SDSZT include response to inorganic substance, response to toxic substance, response to lipopolysaccharide, and reactive oxygen species metabolic process, apoptotic signaling pathway, response to wounding, and response to extracellular stimulus (Figure 4(a)). e main pathways involved are pathways in cancer, fluid shear stress and atherosclerosis, and TNF signaling pathway (Figure 4(d)). e enrichment targets of target-pathway enrichment results are shown in Table 2, and the top 20 filtered pathway-target network diagrams produced by Cytoscape 3.7.1 are shown in Figure 4(e).
Relevant targets regulate the function of dry AMD, which is mainly enriched in cytokine receptor binding, transcription factor binding, protein domain specific binding, kinase binding, phosphatase binding, antioxidant activity, and oxidoreductase activity (Figure 4(b)).

3.5.
SDSZT-Dry AMD-Pathway PPI Data. Cytoscape 3.7.1 analyzed the SDSZT-disease target-pathway network, as shown in Figure 5. By backpropagation, 33 compounds of SDSZT acting on dry AMD were obtained, and the network diagram included 725 nodes. e topological parameters of the SDSZT network were analyzed by NetworkAnalyzer, and the core components and core targets were obtained. e analysis showed that D1 quercetin had a degree of 162, a betweenness of 0.4159, and a closeness of 0.608, predicting quercetin to be the main component of SDSZT for the treatment of dry AMD, followed by C2 kaempferol, with a degree of 54, a betweenness of 0.05453, and a closeness of 0.4270, and NZ6 luteolin, with a degree of 32, a betweenness of 0.0705, and a closeness of 0.4318; C1 β-sitosterol has a degree of 19, a betweenness of 0.0075, and a closeness of 0.38; SS3β-carotene has a degree of 17, a betweenness of 0.0259, and a closeness of 0.3897; B1 stigmasterol had a degree of 14, a betweenness of 0.0214, and a closeness of 0.3535 (Table 3).
PTGS2, with a network degree of 29, a betweenness of 0.1304, and a closeness of 0.5084 in the network, was predicted to be the most important target of SDSZT for the treatment of dry AMD. Next, AKT1, PTGS1, TNF, JUN, BCL2, CASP3, PIK3CG, AR, and BAX were also relatively important regulatory targets (Table 4).

Discussion
Age-related macular degeneration is an age-related disease that is currently recognized as a high-risk blinding eye disease, and it is essential to establish an effective treatment measure in an increasingly aging society. Dry AMD is known to be a complex disease with multiple risk factors and molecular mechanisms. In the study of in vivo and in vitro experimental models of AMD, these molecular mechanisms involving AMD can be broadly categorized as oxidative stress-mediated, antioxidant dysregulation, inflammation, dysregulation of lipid metabolism, and angiogenic dysregulation [1,3,26]. For dry age-related macular degeneration (AMD) and advanced geographic atrophy (GA), the treatment options available only slow the progression of the disease. However, there is currently no therapy that can restore degenerated retinal pigment epithelial (RPE) and/or photoreceptor cells. Currently, studies have focused on antibody, gene, and stem cell therapies for the treatment of dry AMD [27,28], which are potentially effective therapies but have not yet been applied in the clinical setting in the absence of long-term validated safety profiles. Shuang Di Shou Zhen Tablets (formerly known as Ziyin Bushen Tablets) have been used in the treatment of dry AMD for 20 years. In previous clinical studies, patients with dry AMD Figure 1: Construction of the SDSZT compound-potential target network. e compound-potential target network was constructed by linking the candidate compounds and their potential targets of the 10 herbs, which are constituents of SDSZT. Squares are for traditional Chinese medicine; hexagons are compound components; abbreviations for compounds can be found in Table 1; diamonds are targets.   Evidence-Based Complementary and Alternative Medicine were found to have improved visual acuity after receiving a combination of SDSZTand supplements such as vitamin E and vitamin C compared to before treatment. As the number of deformation grid in Amsler table was reduced, the range of central vision defect was reduced, and there were also statistical differences on these respects between treatment group and control group (only treated with vitamin E and vitamin C) (P < 0.05) [6,7], but the pharmacological mechanism of action has not been clearly understood. In this study, the active components of SDSZT for the treatment of dry AMD were  Evidence-Based Complementary and Alternative Medicine quercetin, kaempferol, luteolin, β-sitosterol, β-carotene, stigmasterol, etc., based on the preliminary screening by network pharmacology method. Studies on the therapeutic mechanisms of dry AMD have shown that polyphenols such as quercetin can directly scavenge reactive oxygen species (ROS) [29] and it protects ARPE-19 cells against oxidative stress induced by 4hydroxynonene (HNE), an end product of lipid peroxidation [30]. Quercetin protects RPE cells by decreasing mRNA expression of the proinflammatory IL-6 and IL-8 and monocyte chemotactic protein 1 (MCP-1) [31] and activating the Keap1/ Nrf2/ARE pathway [32]. In vivo experiments have shown that kaempferol reduces the apoptosis of rat retinal pigment epithelial cells (RPE cells) caused by NaIO3 by enhancing the expression of vascular endothelial growth factor (VEGF) protein [33]. Luteolin attenuates the IL-1 β-induced increase in interleukin 6 (IL-6), interleukin 8 (IL-8), soluble intercellular adhesion molecule 1 (sICAM-1), and monocyte chemotactic protein 1 (MCP-1) production by inhibiting nuclear transcription factor-kB (NFκ-B) and MAPK signaling pathways to further protect ARPE19 cells [34]. erefore, SDSZT may be used to regulate dry age-related macular degeneration through quercetin, kaempferol, and luteolin. Based on the results of this study, the highest degree value target of SDSZT for dry age-related macular degeneration is PTSG2, while AKT1, PTGS1, TNF, JUN, BCL2, CASP3, PIK3CG, AR, and BAX are also likely to be key therapeutic targets for SDSZT. Gene Ontology analysis and KEGG pathway enrichment analysis showed that these key targets involve multiple pathways, including prostate cancer [35], TNF signaling pathway [36], NF-Kappa B signaling pathway [37], and (e)  JAK-STAT signaling pathway [38], all of which are closely related to the AMD development process. PTGS2, also known as cyclooxygenase (COX2), has been reported to be involved in the AMD process as a key enzyme in the inflammatory response, and is highly expressed in retinal pigment epithelial cells [39]. JUN, also known as transcription factor ap-1, can be activated by 4-hydroxyhexenal-(HHE-) under the control of cyanidin-3-glucoside (C3G) and is associated with AMD inflammation [40]. e expression of phosphatidylinositol 3 kinase (PIK3CG) was upregulated when bevacizumab treated RPE cells in an AMD model [41], suggesting that PIK3CG may be a target for the treatment of AMD. Current information on the modes of RPE death in AMD disease includes apoptosis, necrosis, autophagy, and ferroptosis [42]. e downregulation of the apoptosis suppressor gene Bcl-2 and the upregulation of the proapoptosis gene Bax have been shown to be the main causes of apoptosis in RPE cells in the AMD model [43]. e results of this study indicate that a single monomer compound of SDSZT can regulate different targets, and a single target can interfere with different biological processes and signaling pathways. is embodies the function characteristics of the multipathway, multitarget function of SDSZT. At the same time, it is clear that the whole network can be modulated by modulating a single monomer or an important target or multiple targets in the pharmacological network. is provides a scientific basis for the clinical use of SDSZT for dry age-related macular degeneration and a new direction for exploring the potential mechanisms of SDSZT for dry age-related macular degeneration. However, network pharmacology methods have certain limitations, and validation of the predicted targets and monomer compounds is  Table 1 for Chinese herbal abbreviations. necessary as well to further demonstrate the clinical therapeutic potential of SDSZT and then clarify the exact regulation target of SDSZT.
Data Availability e data supporting this network pharmacology analysis are from previously reported studies and databases, which have been cited. e processed data are available from the corresponding author upon request.

Conflicts of Interest
e authors declare that they have no conflicts of interest.