Age-related macular degeneration (AMD) is a multifactorial disorder having risk factors, such as heredity, environment (smoking, light, diet, etc.), age, and race (Mitchell et al., 2018). AMD mainly affects the macular region of the retina. In severe cases, it can lead to vision loss and even blindness. Wet AMD is an advanced stage of AMD that is mainly characterized by choroidal neovascularization (CNV), macular hemorrhage, exudation, and edema. Current clinical diagnosis of AMD is done through funduscopy, fluorescein angiography, optical coherence tomography angiography, and, occasionally, indocyanine green angiography (Told et al., 2018). The first-line treatment for advanced AMD is an intravitreal injection of anti-vascular endothelial growth factor drugs; however, some patients may experience disease relapse after operation and even show no response to treatment. Therefore, AMD pathogenesis needs further exploration. Recently, some researchers have postulated that the intestinal microbiota may be involved in AMD pathogenesis, providing a new avenue for exploring the etiology and therapeutic modalities for AMD (Andriessen et al., 2016; Zinkernagel et al., 2017; Zysset-Burri et al., 2020).

The gut microbiota is composed of bacteria, archaea, viruses, fungi, and eukaryotic microbes. Among these, bacteria are widely studied by researchers. The most common bacterial phyla are Firmicutes, Bacteroidota, Actinobacteria, and Proteobacteria (Lynch and Pedersen, 2016). The intestinal microbiota interacts with the host to maintain the normal functioning of the human body. It plays many roles in immunological regulation, metabolism, and intercellular communication to contribute to the homeostasis of the body (Jandhyala et al., 2015). However, intestinal microbiota abnormalities have been linked to disorders ranging from localized gastroenterology to systemic diseases (Hou et al., 2022). The view that the gut microbiota affects anatomically distant organs is very confusing, especially the regulatory effect of the gut microbiota on immune-privileged sites, like the eyes (Zárate-Bladés et al., 2017). Over the past few years, groundbreaking research on the relationship between intestinal microbiota and the retina (gut-retina axis) has increased (Lin, 2019; Scuderi et al., 2021).

Disturbances in the intestinal microbiota and its metabolites may cause systemic inflammation by stimulating both the innate and adaptive responses of the immune system. This inflammation may be the reason for many diseases, including the ocular diseases, as revealed by human and animal studies (Floyd and Grant, 2020). Ten percent of individuals with inflammatory bowel disease develop eye diseases, such as episcleritis, uveitis, and conjunctivitis (Vavricka et al., 2015). One study (Astafurov et al., 2014) found that low levels of lipopolysaccharide (LPS) administered subcutaneously to induce chronic inflammation in two animal models of glaucoma could increase optic nerve axon loss and expression of Toll-like receptor (TLR) and complement system pathway genes, which could cause microglial activation in the affected tissue. Moreover, microglia participate in the pathology of glaucomatous and other neurodegenerative diseases and express high levels of TLR4 (Amor et al., 2010; Ebneter et al., 2010). Studies in humans have identified an association between TLR4 polymorphisms and normal tension glaucoma (Shibuya et al., 2008). Administrating a TLR4 inhibitor, naloxone, orally was found to protect retinal ganglion cells and optic nerve axons against damage (Astafurov et al., 2014). It is very likely that an increase in the level of LPS in circulation plays a vital role in inducing diabetes-related inflammation. Microbial metabolites and components migrate into the circulation through the damaged intestinal barrier; this could trigger oxidative stress and pro-inflammatory pathways and ultimately lead to vascular dysfunction (Burcelin, 2016; Xu and Chen, 2017). One mechanism that possibly links the imbalance of the gut microbiota to diabetic retinopathy may involve the activation of TLR, the key mediator of innate immunity, to cause systemic inflammatory and abnormal vascular responses (Li et al., 2017). Fernandes et al. found that hyperglycemia was related to an altered intestinal barrier, causing the loss of immune system homeostasis in association with systemic chronic inflammation and even involving retinal vessels. They hypothesized that all these changes contributed to retinal inflammation, breakdown of the blood-retinal barrier, apoptosis, and neovascularization, leading to diabetic retinopathy progression (Fernandes et al., 2019). The combination of LPS and TLR4 was found to trigger a series of reactions, eventually leading to the release of pro-inflammatory molecules, which interfere with glucose and insulin metabolism (Blandino et al., 2016). A study performed on type 1 diabetes patients with and without diabetic retinopathy demonstrated that they all had higher levels of LPS compared to individuals with normal glucose tolerance (Fernandes et al., 2019). A growing number of studies have also shown that type 2 diabetes is associated with an alteration of the intestinal microbiota (Qin et al., 2012; Karlsson et al., 2013). Potential mechanisms underlying this association include increased inflammation, oxidative stress, and vascular permeability (Allin et al., 2015; Arora and Bäckhed, 2016). A study of a transgenic mouse model of uveitis, showed that the intestinal microbiota could activate autoreactive T cells that produced IL-17 and thereby directly triggered uveitis (Horai et al., 2015). The reason behind this direct association between the gut and the eye is that auto-pathogenic T cells in the retina receive a microbiota-dependent activation signal through their clonotypic T cell receptor in the gut (Horai et al., 2015). Another study revealed, using different uveitis animal models, that administration of broad-spectrum antibiotics increased regulatory T cells in the lamina propria and decreased effector CD4+ T cells (Nakamura et al., 2016). In humans, uveitis and related autoimmune diseases are related to human leukocyte antigen-B27, which is known to be associated with intestinal disorders in human and animal models that cause increased intestinal permeability and innate and adaptive immune responses (Nakamura et al., 2016). Although a growing number of studies have found important roles played by intestinal microbiota in many diseases, these roles need to be explored further.

Recent studies employed CNV models to investigate the involvement of intestinal microbiota in the pathogenesis of AMD (Andriessen et al., 2016). These studies explored indirect effects of dietary changes on the retina mediated by the composition of the gut microbiota. However, the pathway underlying the direct involvement of intestinal microbiota in AMD remains to be explored. There have only been a few studies to date on changes in the intestinal microbiota of patients with AMD (Zinkernagel et al., 2017; Zysset-Burri et al., 2020). Moreover, intestinal microbiota vary with geographical location (Vaiserman et al., 2017). As a result, more investigations and validations at other research sites are required to ascertain the effects of changes in gut microbiota on AMD patients. In the present study, to elucidate the mechanisms underlying AMD pathogenesis, we employed 16S rRNA gene sequencing to describe the intestinal microbiota of AMD patients and forecast probable metabolic pathways they may affect. Through this method, we searched for potential microbial biomarkers for the development of AMD and innovative treatment methods for AMD.

The intestinal microbiota is involved in human disease resistance. Since the concept of the gut-retina axis was put forward, the role of intestinal microbiota in regulating eye diseases has been gradually acknowledged (Rowan and Taylor, 2018). Recent studies have revealed the changes and potential applications of gut microbiota in patients with various retinal disorders (Gong et al., 2020; Skondra et al., 2020). A few studies have indicated an association between gut microbiota and AMD (Zinkernagel et al., 2017; Zysset-Burri et al., 2020). To understand this association, we used 16S rRNA gene sequencing in the present study to identify differences in the gut microbiota of patients with AMD and healthy individuals. We found that AMD patients had a different composition of gut microbiota than healthy controls. Moreover, we performed functional predictions of intestinal microbiota in patients with AMD.

In the present study, we found a significant difference in the gut microbiota and specific bacterial taxa between the AMD and control groups. At the phylum level, the AMD group was found to have significantly lower proportions of Firmicutes, while the proportions of Proteobacteria and Bacteroidota were found to be markedly increased. At the genus level, the abundance of Escherichia-Shigella increased notably while the abundance of Blautia and Anaerostipes decreased greatly in the AMD group compared with those in the control group. Enterotype analysis further revealed that Escherichia-Shigella was the main microbiota component in the AMD group, and Blautia was the main one in the control group. Studies of the gut microbiota to date have found that the composition of the gut microbiota differs between AMD patients and healthy individuals. Zinkernagel et al. found that Anaerotruncus, Oscillibacter, Ruminococcus torques, and Eubacterium ventriosum were enriched in AMD patients, and, conversely, Bacteroides eggerthii was enriched in control individuals (Zinkernagel et al., 2017). Another larger investigation of human intestinal microbiota found that AMD patients had a higher level of the class Negativicutes, whereas healthy controls had a higher level of the genus Oscillibacter and species Bacteroides (Zysset-Burri et al., 2020).

Firmicutes, a main component of intestinal flora, can produce short-chain fatty acids (SCFAs), especially butyrate, which contribute toward maintaining epithelial integrity, reducing appetite and inflammation, and preventing weight gain associated with the consumption of a high-fat diet (HFD; Blaak et al., 2020). Short-chain fatty acids have been found to regulate intestinal microbiota by reducing pH; preventing excessive growth of pathogens, such as Escherichia coli; and stimulating the growth of beneficial bacteria of the phylum Firmicutes (Duncan et al., 2009; Ghosh et al., 2011; Zimmer et al., 2012). Blautia, a member of the Firmicutes phylum, has long been recognized for its function in the treatment of inflammatory and metabolic diseases (Liu et al., 2021). The production of bacteriocins enables Blautia to inhibit the colonization of pathogenic bacteria in the gut; this ability has allowed its use in probiotics (Liu et al., 2021). One study found that Blautia was more enriched in the feces of healthy dogs who were fed with more higher concentrations of potato fiber (Panasevich et al., 2015). Dietary fiber added to HFD for mice reduced their body weight and greatly increased the abundance of Blautia in their feces (Yang et al., 2016). A Japanese study revealed that Blautia was the only genus with a significant negative correlation between abundance and visceral fat accumulation (Ozato et al., 2019). Bacteroidota is composed of various genera with broad metabolic potential, and changes in the abundances of these genera are negatively or positively correlated with human health (Pittayanon et al., 2019). Proteobacteria are gram-negative bacteria, and their outer membranes are mainly composed of LPS. Many common human pathogens are found in the Proteobacteria phylum, such as Escherichia-Shigella, which was found to be associated with increased intestinal permeability and was detected in a variety of disorders (Pedersen et al., 2018). Moreover, increased microbiota instability combined with a rise in the quantities of potential pathogens may affect human hosts by raising the likelihood of local or systemic infections. This kind of instability may lead to increased LPS secretion, which might exacerbate metabolic disorders, creating a vicious loop (Frost et al., 2021). Yang et al. showed that Anaerostipes can produce butyric acid, promote the absorption of carbohydrates, and increase the energy utilization efficiency and may thus alleviate the symptoms of irritable bowel syndrome with diarrhea (Yang et al., 2021).

Through phenotypic prediction, we found that microbial phenotypes were different between the AMD and control groups. Escherichia-Shigella, Klebsiella, Kluyvera, Bacteroides, Phascolarctobacterium, Ruminococcus, and Bifidobacterium were obviously more enriched in AMD patients, and Blautia and Subdoligranulum were mainly enriched in the control group. Klebsiella pneumoniae is the main pathogen involved in Gram-Negative bacteremia and the most common cause of healthcare-associated infections overall. The gastrointestinal tract is a major reservoir for Klebsiella pneumoniae bacteremia (Diekema et al., 2019). Kluyvera is a genus of the family Enterobacteriaceae, which caused many kinds of infectious diseases. Many species of this genus have been reported to cause human infection, and Kluyvera ascorbata is the most common species causing infection in humans (Lee et al., 2019). A large meta-analysis found a higher abundance of Proteobacteria, Bifidobacterium, and Phascolarctobacterium in patients with Alzheimer’s disease compared with healthy individuals (Hung et al., 2022). A strong correlation exists between Crohn’s disease (CD) and Ruminococcus gnavus. Some researchers found that an increased relative abundance of Ruminococcus gnavus in the gut microbiota was associated with an increase in CD and various other inflammatory diseases (Hall et al., 2017; Nishino et al., 2018). Another research found that Ruminococcus was independently associated with liver fibrosis (Boursier et al., 2016). Upon examining Subdoligranulum in two cohorts of overweight/obese individuals (from two separate studies), researchers found that the abundance of Subdoligranulum had a positive correlation with the HDL-cholesterol level and a negative correlation with a variety of metabolic risks parameters (Van Hul et al., 2020). Another study revealed that Subdoligranulum improved necrotizing enterocolitis by producing butyrate to regulate gut function (Lin et al., 2021).

A predictive function analysis was next used to better investigate the rich metabolic activities of the gut microbiota. K00979 represents KdsB, also known as CMP-KDO synthetase or KDO cytidylyltransferase. In Gram-Negative bacteria, KdsB is an important enzyme in the biosynthesis of LPS. K03270 represents KdsC, which is also a key enzyme in the synthesis of KDO. Any interruption in the KDO biosynthesis pathway can reduce the pathogenicity of bacteria (Ahmad et al., 2019). Lipid A is a highly conserved component of LPS. Nine essential enzymes, including K00677, K02517, and K02536, were involved in the Lipid A biosynthesis pathway (Emiola et al., 2014). K03561 represents the biopolymer transport protein ExbB. TonB complex, composed of TonB, ExbB, and ExbD, is an energy provider for the active transport of external nutrients by bacteria. A great number of studies have shown that the TonB system is not only involved in nutrient transport but also in bacterial signal transduction and, indirectly, in bacterial virulence (Porcheron et al., 2013). K07277 represents the outer membrane protein insertion porin family or bamA. BamA is the core component of the β⁃barrel assembly machinery (BAM) complex, which plays a key role in the immunogenicity, pathogenicity, and multidrug resistance of bacteria (Noinaj et al., 2013; Babu et al., 2018). K06861 represents the LPS export system ATP-binding protein or LptB. The absence of any Lpt component prevents LPS transport to the outer membrane of gram-negative bacteria (Bowyer et al., 2011). In the present study, KEGG analysis revealed the top seven pathways linked to the abovementioned gut microbiota that were altered in the AMD group. These pathways were also reported to be involved in inflammatory disease in other studies. E. ramulus, of the Firmicutes phylum, has been found to cleave quercetin and produce 3,4-dihydroxyphenylacetate, which inhibits colon cancer cell proliferation (Gao et al., 2006). Moreover, Bacteroides thetaiotaomicron, of the Bacteroidota phylum, is thought to be able to metabolize complex substrates, like mucin; it is thus important to learn whether it contributes to inflammation and infection by weakening the mucosal barrier (Tramontano et al., 2018).

However, there are several limitations to this research as well. First, since this was single-center research with a small sample size, further multi-center studies are required to expand the sample size. Second, our research predicted metabolic functions of gut microbiota in AMD patients and healthy controls; further analysis of bacterial metabolites is needed to determine their potential underlying mechanisms. Third, we found that the proportions of Blautia and Anaerostipes in the AMD group were lower than those in the control group. Contrarily, the abundances of Escherichia-Shigella, Bacteroides, and Bifidobacterium in the AMD group were higher than those in the control group. Whether to change the abundances of these bacteria in the guts of patients with AMD or to use the metabolites produced by these bacteria to treat AMD remains to be further explored.

In conclusion, the present study revealed alterations in the compositions and metabolic functions of intestinal microbiota between AMD patients and healthy individuals. Moreover, it proposed specific bacteria and potential microbial markers-that can be targeted for AMD treatment. These findings clarify the mechanism underlying AMD from the perspective of intestinal microbial dysbiosis. In addition, the findings provide a foundation prospect for improving the balance of the intestinal microenvironment to treat AMD.

The data presented in the study are deposited in the SRA database repository accession number PRJNA799475.

The studies involving human participants were reviewed and approved by The ethics committee of Shanghai Tenth People’s Hospital. The patients/participants provided their written informed consent to participate in this study.

QP, MW, and RD conceived and designed the study. YZ wrote the manuscript. YZ, TW, and ZW performed the experiments. JB and YX analyzed the data. YZ, TW, YX, JB, and ZW reviewed the manuscript. All authors contributed to the article and approved the submitted version.

This work was supported by the National Natural Science Foundation of China (grant numbers 62271337 and 81470029) and Three-Year Action Plan for Promoting Clinical Skills and Clinical Innovation in Municipal Hospitals (grant number SHDC2020CR5014).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.