Abstract

   Growing evidence suggests that perfluorinated compounds (PFCs)
   contribute to reproductive toxicity, with perfluorooctanoic acid (PFOA)
   and perfluorooctane sulfonate (PFOS) being the most extensively
   studied. These chemicals are known to lower testosterone levels and
   compromise the integrity of the blood-testis barrier. However, the
   specific mechanisms of their reproductive toxicity remain largely
   unknown due to research limitations. In this study, we utilized network
   pharmacology to pinpoint the core genes and signaling pathways
   implicated in the reproductive toxicity caused by PFOA and PFOS.
   Molecular docking was employed to validate the interactions between
   these compounds and their targets. Key targets identified include CCL2,
   CXCR4, RPS27A, RPL5, PSMA7, and PSMC1, which are crucial in mediating
   reproductive toxicity. These genes are primarily involved in the
   chemokine signaling pathway, viral protein interactions with cytokines
   and cytokine receptors, and ribosomal functions. This study underscores
   the effectiveness of combining network toxicology and molecular docking
   to analyze the toxicity and molecular mechanisms of mixed environmental
   pollutants.

   Keywords: Perfluorooctanoic acid, Perfluorooctane sulfonate, Network
   pharmacology, Molecular docking, Reproductive toxicity

Graphical abstract

   Image 1
   [31]Open in a new tab

Highlights

     * •
       Network toxicology and rapid molecular docking are effective for
       assessing the toxicity and potential molecular mechanisms of
       perfluorinated compounds.
     * •
       The reproductive toxicity of PFOA and PFOS was analyzed using these
       techniques.
     * •
       Using network toxicology and molecular docking for toxicity
       assessment can address cost and ethical concerns related to animal
       use.

1. Introduction

   Perfluorinated compounds (PFCs) comprise more than 3000 individual
   compounds and generally contain a charged functional moiety, such as a
   sulfonate or carboxylate [[32]1]. A wide range of industrial and
   household products have been pplied to PFCs for their good
   physicochemical properties, including textiles, surfactants, furniture,
   food packaging and lubricants. Despite these advantages, PFCs pose
   serious long-term risks to the environment and human health [[33]2].
   Their toxic effects likely involve multiple targets and pathways,
   disrupting protein and gene networks rather than just altering
   individual proteins or genes [[34]3]. Additionally, since consumers
   often ingest residues of various active substances through their diet,
   it is important to assess the potential combined effects of these
   substances [[35]4]. Therefore, innovative and systematic methods are
   needed to evaluate the health risks of the increasing number of PFCs
   effectively.

   Perfluorooctanoic acid (PFOA) and perfluorooctane sulfonate (PFOS) are
   PFCs widely utilized in various industries [[36]5]. Their persistence
   in the environment and accumulation in the food chain have sparked
   significant concern [[37]6,[38]7]. PFOA and PFOS have been found in the
   tissues and plasma of numerous animal species across almost all
   continents, including polar bears, fish, and birds in North America
   [[39]8]. Based on epidemiological studies, some scholars have developed
   health-based guidance values for PFOA and PFOS in the Chinese
   population, which are 1.52 and 1.56 ng/kg/day, respectively [[40]9].
   Despite new alternatives, similar environmental problems continue to
   arise. The elimination half-life of PFOA in humans is around four
   years, and for PFOS, it is approximately five years [[41]10,[42]11].
   These compounds enter the body mainly through skin contact and
   ingestion, accumulating in various organs and causing toxicity
   [[43]12]. Mammalian studies show that rodents exposed to PFOA and PFOS
   experience weight loss and lipid metabolism issues [[44]13].

   The impact of environmental pollutants on reproduction is a significant
   concern due to its effects on human development. Studies show that PFCs
   can damage testicular function and lower sperm count [[45]14]. Based on
   epidemiological studies, some scholars have developed health-based
   guidance values for PFOA and PFOS in the Chinese population, which are
   1.52 and 1.56 ng/kg/day, respectively [[46]15]. Epidemiological
   research links maternal exposure to PFOA during pregnancy with poor
   semen quality in offspring [[47]16]. In experiments with white-footed
   mice, high prenatal doses of PFOS resulted in neonatal death, while
   lower doses impaired growth and development [[48]17]. For females, PFOA
   and PFOS may produce reproductive toxicity in the ways of damage to
   oocytes through oxidative stress and Inhibition of steroid hormone
   synthesis. Additionally, PFOA and PFOS are linked to reduced
   testosterone levels and compromised blood-testis barrier integrity in
   humans [[49]18]. In women, these chemicals are associated with hormone
   imbalances and higher infertility risks [[50]19]. Despite extensive
   studies on their reproductive effects in rodents [[51]20,[52]21], the
   exact mechanisms of PFOA and PFOS toxicity are not well understood.

   This study aimed to uncover the core genes and signaling pathways
   involved in PFOA- and PFOS-induced reproductive toxicity using network
   pharmacology. We used molecular docking to validate interactions
   between these compounds and their targets, shedding light on their
   toxic mechanisms. This innovative approach provides a rapid method for
   assessing the reproductive toxicity of PFCs and offers new insights for
   treating diseases linked to these toxic substances.

2. Methods

2.1. Collection of potential targets for PFOA and PFOS

   Limiting the analysis to "Homo sapiens," we identified potential
   targets for PFOA and PFOS from the ChEMBL database
   ([53]https://www.ebi.ac.uk/chembl/) [[54]22]. Obtaining the SMILE nodes
   of PFOA and PFOS from the PubChem database
   ([55]https://pubchem.ncbi.nlm.nih.gov/) [[56]23], we then inputted
   these nodes into the Swiss Target Prediction database
   ([57]http://www.swisstargetprediction.ch/) to gather overloaded targets
   [[58]24]. Integrating and deduplicating the Uniprot IDs of all targets
   from the ChEMBL and Swiss Target Prediction databases, we standardized
   the potential targets using the Uniprot database
   ([59]https://www.uniprot.org/id-mapping/) [[60]25]. This process
   facilitated the construction of a potential target library for PFOA and
   PFOS.

2.2. Screening for reproductive toxicity targets

   We searched for related targets using “reproductive toxicity” as the
   key word in the GeneCards ([61]https://www.genecards.org/) and OMIM
   ([62]https://www.omim.org/) databases [[63]26], selecting genes with
   scores above the median to ensure strong correlation. Standardizing all
   targets using the Uniprot database specified "Homo sapiens" as the
   species for target and gene name conversion. Next, we identified common
   targets between PFOA/PFOS and reproductive toxicity by intersecting
   their respective target sets. Using the Venny 2.1.0 online platform
   ([64]https://bioinfogp.cnb.csic.es/tools/venny/), we created a Venn
   diagram to visualize the overlap.

2.3. Construction of protein-protein interaction (PPI) network

   Using the PPI network, we visualized the co-localization, neighborhood,
   and co-expression of interactions among potential target genes and
   predicted genes. Genes representing potential targets of
   PFOA/PFOS-induced reproductive toxicity were analyzed in the String
   database ([65]https://string-db.org) to gain deeper insights into the
   toxic mechanisms of PFOA and PFOS. To ensure PPI accuracy and quantity,
   a confidence score exceeding 0.4 was required, with species limited to
   "Homo sapiens" [[66]27]. The resulting PPI network was analyzed using
   Cytoscape software (version 3.10.1) for topology assessment, and core
   targets were identified based on their degree value ranking. This
   visualization method allowed for easy identification of key targets, as
   node size and color depth reflected numerical changes.

2.4. Analysis of target function and pathway enrichment

   We utilized the DAVID online tool for Gene Ontology (GO) analysis and
   Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis to
   identify functional annotations and pathway enrichment related to
   potential genes. Following this, we performed visual analysis on the
   Bioinformatics online platform ([67]https://www.bioinformatics.com.cn/)
   to interpret and present the results of the GO and KEGG analyses.

2.5. Verification of interaction by molecular docking

   A molecular docking approach was utilized to forecast the binding
   affinity between PFOA/PFOS and their identified core targets. The
   crystal structures of these core proteins were acquired from the RCSB
   Protein Data Bank ([68]http://www.rcsb.org). The core target protein
   underwent modifications including dehydration, hydrogenation, and
   charge balancing using AutoDockTools1.5.7 software, followed by the
   docking experiment.

3. Results

3.1. Identification of targets of PFOA/PFOS-induced reproductive toxicity

   244 potential target proteins of PFOA and PFOS were sourced from the
   ChEMBL and Swiss Target Prediction databases, integrated through
   Uniprot. Additionally, 3403 targets highly relevant to reproductive
   toxicity were screened from GeneCards and OMIM. Utilizing the Venny
   2.1.0 online platform, 82 common targets of PFOA/PFOS and reproductive
   toxicity were identified, detailed in [69]Fig. 1. These 82 targets may
   play a significant role in PFOA/PFOS-induced reproductive toxicity.

Fig. 1.

   [70]Fig. 1
   [71]Open in a new tab

   Venn diagram of potential targets of PFOA/PFOS and reproductive
   toxicity.

3.2. PPI network construction and core target screening

   Eighty-two potential targets related to reproductive toxicity from PFOA
   and PFOS were analyzed using the String database, generating a PPI
   network with 436 edges and an average node degree of 10.6. This
   network's structure allowed for an in-depth examination of how PFOA and
   PFOS induce reproductive toxicity. The results are shown in [72]Fig. 2,
   where node size and color indicated different aspects of node
   importance.

Fig. 2.

   [73]Fig. 2
   [74]Open in a new tab

   PPI network of the potential action targets of PFOA and PFOS induced
   reproductive toxicity. The darker colors, larger fonts, and larger
   shapes of genes indicate higher betweenness, closeness, and degree
   values.

   In the PPI network, the darker the color and the larger the node area,
   which indicate the node is more important. Six core targets crucial for
   PFOA/PFOS-induced reproductive toxicity were identified through
   topological analysis based on their degree values. These targets
   include Ribosomal protein S27a (RPS27A), C-X-C chemokine receptor type
   4 (CXCR4), C-C motif chemokine 2 (CCL2), Ribosomal Protein L5 (RPL5),
   Proteasome 26S subunit ATPase 1 (PSMC1), and proteasome subunit alpha
   type-7 (PSMA7) ([75]Table 1). They are expressed across various tissues
   and organs, playing critical roles in regulating transcription, protein
   modification, and inflammatory responses.

Table 1.

   Core targets screened from PPI network.
   Gene Degree Closeness centrality Betweenness centrality Topological
   coefficient
   RPS27A 28 0.506666 0.129092 0.293126
   CXCR4 25 0.550724 0.162562 0.199384
   CCL2 24 0.493506 0.080804 0.258986
   RPL5 22 0.434285 0.039878 0.390442
   PSMC1 22 0.431818 0.011701 0.449631
   PSMA7 21 0.493506 0.073648 0.320987
   [76]Open in a new tab

3.3. Functional and pathway enrichment analysis of core targets

   In this study, 82 key targets were analyzed for GO enrichment using the
   David online tool, revealing 133 biological processes (BP), 43 cellular
   components (CC), and 44 molecular functions (MF) in GO entries.
   [77]Fig. 3 presents the top 10 entries of −lgP values for BP, CC, and
   MF. Biological processes related to reproductive toxicity induced by
   PFOA and PFOS primarily involve mRNA stability regulation and
   calcium-mediated signaling. The analysis of cellular components showed
   significant enrichment of genes related to the secretory granule lumen
   and nucleoplasm. Molecular function analysis revealed close association
   with virus receptor activity and metalloaminopeptidase activity.

Fig. 3.

   [78]Fig. 3
   [79]Open in a new tab

   GO enrichment analysis of the core genes of PFOA and PFOS induced
   reproductive toxicity.

   Additionally, 82 potential targets underwent KEGG analysis using the
   David database to identify their involvement in specific signaling
   pathways. This analysis yielded 30 KEGG pathway enrichment analyses,
   with the top 20 items shown in the bubble plot of [80]Fig. 4 based on
   the −lgP value. Key targets were primarily enriched in pathways such as
   the chemokine signaling pathway, ribosome, viral protein interaction
   with cytokine and cytokine receptor, and proteasome. The proteasome
   pathway is particularly significant as the main degradation pathway for
   misfolded proteins and those undergoing proteolysis during protein
   synthesis. Its functions include cell cycle control, apoptosis, and
   inflammation, all relevant to reproductive toxicity. Continuous
   observation of the entire life cycle of animals is crucial for
   understanding reproductive toxicity, highlighting the importance of
   proteasomes in this process. A network relationship diagram of KEGG
   pathways and core targets was created using the top 10 KEGG enrichment
   pathways and 6 core targets, illustrating the relationship between each
   pathway and genes ([81]Fig. 5).

Fig. 4.

   [82]Fig. 4
   [83]Open in a new tab

   KEGG enrichment analysis of the core genes of PFOA and PFOS induced
   reproductive toxicity.

Fig. 5.

   [84]Fig. 5
   [85]Open in a new tab

   Network relationship diagram of KEGG pathway and core genes.

3.4. Molecular docking verification

   Molecular docking analysis examined how PFOA/PFOS interact with six
   core targets (CCL2, CXCR4, RPS27A, RPL5, PSMA7, and PSMC1) identified
   in the PPI network. And the scoring equation included the sum of
   several effects such as electrostatic interaction, van der Waals’
   force, hydrogen bonding contributions and intermolecular conflicts.
   Studies have shown that stable complexes are formed when the free
   binding energy is negative. A lower binding energy suggests a higher
   probability of ligand receptor binding. The results indicated strong
   binding between PFOA/PFOS and all six core target proteins, with
   binding energies below 0 ([86]Fig. 6). For example, among them, the
   binding affinity of PFOA to CXCR4 was the strongest, with a value of
   −13.16 kcal/mol. PFOA formed 4 hydrogen bonds with ASP-122 and ASP-120
   in the docking pocket of CXCR4. These core targets may play a critical
   role in the molecular mechanism of reproductive toxicity induced by
   PFOA and PFOS.

Fig. 6.

   [87]Fig. 6
   [88]Open in a new tab

   Molecular docking analysis of PFOA/PFOS and the core genes. (A) The
   results of ligand-target binding energy values. (B) PFOA-CCL2, (C)
   PFOA-CXCR4, (D) PFOA-RPS27A, (E) PFOA-RPL5, (F) PFOA-PSMA7, (G)
   PFOA-PSMC1, (H) PFOS-CCL2, (I) PFOS-CXCR4, (J) PFOS-RPS27A, (K)
   PFOS-RPL5, (L) PFOS-PSMA7, (M) PFOS-PSMC1.

4. Discussion

   PFOA and PFOS, widely used in consumer and industrial products, have
   raised concerns due to their persistence in the environment and harmful
   effects on humans and animals [[89]28]. Evidence increasingly connects
   PFC exposure to reproductive toxicity, including reduced testosterone
   levels akin to those seen in infertile men [[90]29]. However, current
   research on PFOA and PFOS is constrained, focusing on their
   toxicological effects rather than fully understanding their
   reproductive toxicity mechanisms. Using network pharmacology, our study
   identified six core genes—CCL2, CXCR4, RPS27A, RPL5, PSMA7, and
   PSMC1—as potential targets of PFOA/PFOS-induced reproductive toxicity,
   confirmed by molecular docking.

   In this study, we analyzed the GO molecular functions and KEGG pathway
   of the reproductive toxicity-related targets and validated them by
   molecular docking techniques. Enrichment analysis revealed several
   important pathways, including chemokine signaling pathway, ribosome,
   viral protein interaction with cytokine and cytokine receptor, and
   proteasome. Regarding the chemokine pathway, the core targets we
   obtained, CCL2 and CXCR4, also belong to chemokines. And for the
   ribosome pathway, the core targets RPS27A and RPL5 are also associated
   with ribosomes. Moreover, the research indicates that core genes
   related to PFOA and PFOS-induced reproductive toxicity are also closely
   associated with chemokines and protein synthesis mediated by different
   molecules, such as core targets RPS27A and RPL5.

   Ribosomes, essential for protein synthesis, contain RPS27A, a subunit
   of ribosomal protein 40S. RPS27A plays key roles in ribosome formation,
   protein synthesis regulation [[91]30], transcriptional control, and
   protein modifications [[92]31], contributing to sperm vitality and
   potentially impacting reproduction. RPL5, involved in various cancers,
   is linked to ribosome defects associated with cancer development
   [[93]32]. In many tumors, RPL5 heterozygote inactivation rates are
   high, with evidence suggesting an anti-tumor effect in breast cancer,
   which is influenced by reproductive factors [[94]33].

   CXCR4 and CCL2, core targets within the chemokine class, guide cell
   movement by binding to and activating cell surface receptors,
   initiating signal transduction toward the chemokine's gradient
   [[95]34]. Initial studies on the chemokine stromal derived factor 1
   (CXCL12) were proposed to be enhanced in several diseases including
   those which affect the female reproductive tract, While its well-known
   receptor was CXCR4. These include endometriosis, Asherman's syndrome,
   endometrial cancers, and ovarian cancers. Being stimulated by the
   chemokine CXCL12, the CXCR4/CXCR7 cascade is involved in tumor
   proliferation, migration, and metastasis [[96]35]. This migration is
   crucial for many biological processes, with CXCR4 recruiting immune
   cells and contributing to inflammation, while CCL2 stimulates tumor
   cell proliferation [[97]36], impacting breast development and cancer
   progression [[98]37].

   PSMA7, a core subunit of the 20S proteasome's α subunit, regulates
   intracellular protein degradation via the ubiquitin proteasome system.
   Reduced PSMA7 and acylaminoacyl peptide hydrolase (APEH) expression
   decreases proteasome activity, impairing protein degradation and
   inhibiting high-grade serous ovarian cancer growth [[99]38]. In
   contrast, the 26S proteasome subunit ATPase (PSMC) family, including
   PSMC1-6, forms a heterohexamer loop and is part of the 19S regulatory
   particles. Silencing PSMC genes can impede prostate cancer development
   and metastasis by affecting processes such as promotion, apoptosis, and
   migration [[100]39].

   Exposure of developing oocytes to PFCs may disrupt embryonic
   development and reproductive outcomes. Environmental endocrine
   disruptors have been linked to adverse effects on the female
   reproductive system, potentially increasing the risk of tumors in
   hormone-sensitive organs such as the breast, ovaries, and endometrium
   [[101]40]. The majority of the six core targets identified through PPI
   network analysis are linked to tumor development. KEGG analysis has
   identified three significant cancer-related pathways: the chemokine
   signaling pathway, ribosome pathway, and viral protein interaction with
   cytokine and cytokine receptor pathway. These results suggest a strong
   association between PFOA/PFOS-induced reproductive toxicity and cancer,
   particularly in hormone-sensitive organs.

   The combination of network toxicology and molecular docking utilizes
   advances in bioinformatics, genomics, and big data to explore the
   mechanisms of mixed environmental pollutants. This methodology enhances
   the efficiency of ecotoxicological research. However, current
   investigations, including our own, are constrained by the reliance on
   computational techniques to understand the molecular mechanisms behind
   the reproductive toxicity of PFOA and PFOS. These mechanisms require
   further validation through pharmacological and clinical studies.
   However, research based on network pharmacology and molecular docking
   technology only qualitatively predicts the toxic targets of
   ingredients, and clear physiological effects still need to be verified
   through animal experiments or even clinical trials. Given these
   inherent limitations, we believe that the combined molecular docking
   method of network pharmacology is only suitable to evaluate certain
   pollutants that directly interact with proteins expressed by certain
   genes.

5. Conclusions

   In this study, a novel approach combining network pharmacology and
   molecular docking was used to explore the targets and pathways involved
   in PFOA/PFOS-induced reproductive toxicity. Core targets such as CCL2,
   CXCR4, RPS27A, RPL5, PSMA7, and PSMC1 are implicated in mediating
   reproductive toxicity, primarily through pathways like chemokine
   signaling, viral protein interaction with cytokines, and ribosomal
   pathways. Molecular docking further supported the association between
   PFOA/PFOS and these targets. This study represents a significant step
   forward in analyzing the toxicity and molecular mechanisms of potential
   mixed environmental pollutants using network toxicology and molecular
   docking.

CRediT authorship contribution statement

   Liang Chen: Writing – original draft, Funding acquisition, Formal
   analysis, Conceptualization. Shanshan Liang: Investigation, Formal
   analysis, Conceptualization. Jiaxin Li: Investigation. Qian Li:
   Conceptualization. Qingwen Sun: Writing – review & editing, Funding
   acquisition.

Data availability statement

   Data will be made available on request.

Funding statement

   The work was supported by the National Natural Science Foundation of
   China (82160803), and High level Innovative Talents Project of Guizhou
   Province (Guizhou Science and Technology Cooperation Platform
   Talents-GCC [2023] 077).

Declaration of competing interest

   The authors declare the following financial interests/personal
   relationships which may be considered as potential competing
   interests:Liang Chen reports financial support, article publishing
   charges, equipment, drugs, or supplies, and writing assistance were
   provided by National Natural Science Foundation of China. Qingwen Sun
   reports financial support, article publishing charges, and equipment,
   drugs, or supplies were provided by High level Innovative Talents
   Project of Guizhou Province. If there are other authors, they declare
   that they have no known competing financial interests or personal
   relationships that could have appeared to influence the work reported
   in this paper.

References