Metabolomic analysis identifies alterations of amino acids in the tears and plasma of patients with dry eye disease with ocular pain

Abstract

Objectives:

Identifying the amino acid in tears and plasma and investigating the correlation between amino acid concentrations and the presence of dry eye disease (DED) with ocular pain.

Methods:

15 participants in the DED with ocular pain group, 19 in the DED group, and 16 in the control group were enrolled and underwent DED examinations. Evaluation parameters included the Ocular Surface Disease Index score, Numeric Rating Scale score, clinical ocular parameters, as well as measurements from in vivo laser confocal microscopy. Amino acid concentrations were analyzed using the Waters ACQUITY UPLC I-Class/Xevo TQ-S micro system.

Results:

A total of 29 and 36 distinct amino acids were identified in tear fluid and plasma, respectively. The results showed significantly higher levels of methionine (MET) in tear fluid and 1-methyl-L-histidine (1-MEHIS) in plasma in people who developed DED with ocular pain compared to those who did not (p = 0.003, q-value = 0.044; p < 0.001, q-value < 0.001), with receiver-operating characteristic analysis yielding an AUC of 0.686 and 0.869.

Conclusion:

MET in tear fluid and 1-MEHIS in plasma were significantly associated with DED with ocular pain, showing promise as potential biomarkers for DED with ocular pain. Further validation through large-scale studies is warranted.

1 Introduction

Dry eye disease (DED) is a chronic ocular surface disorder influenced by multifactorial etiologies, often accompanied by inflammatory responses, tissue damage, and neural abnormalities, which collectively contribute to a spectrum of uncomfortable ocular symptoms and visual dysfunctions (1). With the increasing prevalence of digital device usage, worsening air pollution, and the progression of population aging, the incidence of DED has been steadily rising. Current data suggest an annual increase in incidence rates ranging from approximately 5 to 50% (2). Clinical manifestations of DED often include sensations such as foreign body presence, burning, and prickling discomfort, and ocular pain affects 89% of patients with DED, significantly impairing their quality of life and contributing to a substantial economic burden on society (3).

In recent years, there has been growing clinical interest in understanding the role of ocular pain in DED. Literature suggests that DED with ocular pain is potentially associated with unstable tear film, abnormal tear secretion, ocular surface inflammation, and reduced corneal nerve density (4–6). However, the precise pathophysiological mechanisms remain unclear and treatment strategies for DED with ocular pain remain limited. Further in-depth investigations are critical for developing targeted therapeutic approaches.

Amino acids, as essential nutrients and regulators of various metabolic processes, play a fundamental role in maintaining cellular homeostasis. Changes in amino acid concentrations in body fluids offer a promising avenue for the development of diagnostic and prognostic biomarkers. Specifically, tear fluid markers have demonstrated potential for early diagnosis and monitoring of ocular disease progression (7–9). Previous studies employing mass spectrometry have revealed that DED pathogenesis is primarily driven by immune and inflammatory mechanisms. Key metabolic pathways, including complement and coagulation cascades, glycolysis/gluconeogenesis, and amino acid metabolism, have been implicated in the disease process (10–12). Alterations in amino acid metabolism are known to modulate oxidative stress on the ocular surface, exacerbating inflammation and contributing to tissue damage in affected individuals (9). Despite these findings, there is a notable gap in research addressing whether specific amino acids in the tears or blood of patients with DED with ocular pain might influence disease onset, progression, or diagnostic outcomes.

This study primarily detects amino acids in the tears and plasma of patients with DED experiencing ocular pain to investigate potential amino acids associated with the condition.

2 Participants and methods

2.1 Participants and study design

This study employed a cross-sectional, observational design conducted during a single visit, with participants stratified into three groups: the DED with ocular pain group, the DED group, and the control group. Both the left and right eyes of participants were randomly selected for inclusion. Data collection was comprehensive, encompassing demographic information, the Ocular Surface Disease Index (OSDI) score, Numeric Rating Scale (NRS) scores, clinical ocular indicators, as well as the collection of tear fluid and plasma samples. Additionally, in vivo laser confocal microscopy (IVCM) examinations were performed. Subsequently, amino acid concentrations in both tear fluid and plasma were quantified.

The diagnostic criteria for DED with ocular pain were as follows (6): OSDI ≥ 13 points, Schirmer I < 10 mm/5 min, tear film breakup time (FBUT) ≤ 7 s, and NRS ≥ 2 points. The diagnostic criteria for DED: OSDI ≥ 13 points, Schirmer I < 10 mm/5 min, FBUT ≤ 7 s, and NRS < 2 points.

Inclusion criteria were as follows: (1) Voluntary participation in the study, with signed informed consent; (2) Age between 18 and 70 years; (3) No history of contact lens use within the past 6 month; (4) No participation in other clinical trials within the past month; (5) No current use of eye or systemic medications, or if used, cessation for more than one month prior to the study.

Exclusion criteria were as follows: (1) Allergy to any component of the test reagents; (2) Active ocular surface infection or recent ocular surface infection (within the past month); (3) Ocular surgery or trauma within the past 6 month; (4) Severe systemic diseases, including those affecting the cardiovascular, cerebrovascular, liver, kidney, thyroid, or hematopoietic systems; (5) Recent coronary stent implantation or history of coronary artery disease; (6) Diagnosed glaucoma or diabetic retinopathy; (7) Use of vitamin and/or folic acid supplements; (8) Pregnant or lactating women; (9) Any individual deemed unsuitable for participation by the researcher, such as those with mental disorders, substance abuse history, or poor compliance during exams.

The study was approved by the Ethics Committee of Wuhan Aier Hankou Eye Hospital (approval number: HKAIER2023IRB-09-01) and was conducted in accordance with the Declaration of Helsinki. Additionally, the study is registered with the Chinese Clinical Trial Registry (registration number: ChiCTR230007634). All participants provided written informed consent prior to inclusion in the study.

2.2 Methods

2.2.1 Basic information

The collected data encompassed the following variables: participant name, gender, age (in years), race, history of drug allergies, past medical and surgical histories, duration of DED (in months) and average daily sleep duration (in hours).

2.2.2 The OSDI and NRS score

The OSDI score (13) consists of three sections. The first section evaluates ocular symptoms (questions 1–5, with a scoring range of 0–20). The second section assesses the impact of ocular discomfort on daily life and work (questions 6–9, with a scoring range of 0–16). The third section examines the contribution of environmental factors to ocular discomfort (questions 10–12, with a scoring range of 0–12). The total score, which ranges from 0 to 100, is calculated by the formula: (Total Score / Number of Questions Answered) × 25.

The NRS (14) is represented as a 10-cm visual analog scale, divided into 100 equal graduations. The endpoints are designated as “0” (no pain, marked with a smiling face) and “10” (intolerably severe pain, marked with a crying face, representing pain so extreme it would necessitate the removal of the eyeball). Participants are asked to mark the point on the scale that corresponds to their pain level, with the total score ranging from 0 to 10.

2.2.3 Ocular examination

2.2.3.1 Corrected visual acuity

Participants stood 5 meters away from the standard logarithmic visual acuity chart and read the visual targets sequentially from top to bottom until they could no longer discern them clearly. Corrected visual acuity was recorded.

2.2.3.2 Intraocular pressure (IOP)

IOP for each eye was assessed using a non-contact tonometer (NT-510, NIDEK, Japan), with the average value calculated from three separate measurements.

2.2.3.3 FBUT

The time interval between blinking and the first tear film rupture was measured three times, with the average value used for analysis. FBUT ≤ 7 s was considered abnormal (15).

2.2.3.4 Schirmer I test

A Schirmer test strip (Liaoning Meizilin Pharmaceutical Co., Ltd., China) was placed in the outer one-third of the lower temporal conjunctival fornix without topical anesthetics. Participants were instructed to gently close their eyes and refrain from moving their eyeballs. After 5 min, the length of the moistened portion of the strip was measured in millimeters (mm) (15).

2.2.3.5 National Eye Institute (NEI) score

The Oxford grading system (0–5 areas) was used to assess the integrity of the ocular surface following corneal fluorescein staining (Tianjin Jingming New Technology Development Co., Ltd., China), with each of the five areas receiving a score between 0 and 3, resulting in a total score ranging from 0 to 15. A higher score indicates a greater degree of ocular surface abnormality (15, 16).

2.2.4 Preparation of tear fluids samples

Fifty microliters of 0.9% sodium chloride injection (Tangpa Medical Co., Ltd., China) were administered into the lower conjunctival fornix using a pipette (17). Participants were instructed to gently close their eyes and tilt their heads to one side. Tear fluid flushed from the conjunctival fornix was collected using a 20 μL microcapillary (Jiangsu Kangjianhua Medical Co., Ltd., China) within a 1-min collection window. The collected tear fluid was transferred from the microcapillary to a 100 μL centrifuge tube (Wuhan Servicebio Technology Co., Ltd., China), then centrifuged, and stored at −80 °C for subsequent analysis. All participants’ tear samples were collected by the same examiner on the second day of enrollment, between 8:00 and 11:00 daily in the same indoor room, with the room temperature controlled at approximately 20 °C and the relative humidity maintained at around 35%.

2.2.5 Preparation of plasma samples

Participants refrained from food and fluid intake for 8 h prior to sample collection. Peripheral venous blood was drawn into an anticoagulant tube (HuBei JinXing Technology Development Co., Ltd., China), inverted gently, and centrifuged at 2750 × g using a centrifuge (320A, Beijing Baiyang Medical Equipment Co., Ltd.) for 10 min at room temperature (18). The resulting supernatant was transferred to a pre-chilled 100 μL centrifuge tube (Beijing Labgic Technology Co., Ltd., China) using a pipette, vortexed for 1 min (Haimen Kylin-Bell Lab Instruments Co., Ltd., China), and stored at −80 °C until analysis. The entire process, from blood collection to vortexing, was completed within 30 min.

2.2.6 IVCM and image analysis

The corneal subbasal nerve plexus was examined using in vivo confocal microscopy (IVCM) (Heidelberg Engineering, GmBH, Heidelberg, Germany) in the central cornea and at four quadrants approximately 3 mm from the center of each participant’s eye (19). For each orientation of the eye, five images were selected (a total of 25 images per eye). Only images with a nerve coincidence rate not exceeding 1/10 were included for analysis. The 25 images from each eye were grouped into a single folder, and the ACC Metrics software (The University of Manchester, United Kingdom) was employed to analyze various parameters for each image within the folder. The analyzed parameters included: Corneal Nerve Fiber Density (CNFD) (number of fibers per mm²), Corneal Nerve Branch Density (CNBD) (number of branch points on the main fibers per mm²), Corneal Nerve Fiber Length (CNFL) (total length of nerves in mm per mm²), Corneal Total Branch Density (CTBD) (total number of branch points per mm²), Corneal Nerve Fiber Area (CNFA) (total nerve fiber area in mm² per mm²), Corneal Nerve Fiber Width (CNFW) (average nerve fiber width in mm per mm²), and Corneal Nerve Fiber Fractal Dimension (CNFracDim) (spatial loss of nerves). The average values for all parameters derived from the 25 images of each eye were subsequently calculated and used for further analysis.

2.2.7 Detection of amino acid profiles in tears and plasma by the Waters ACQUITY UPLC I-Class/Xevo TQ-S micro system (Waters Corporation)

The processed plasma, tear fluid samples, and standards (Supplementary material) were subjected to chromatographic separation using a Waters ACQUITY UPLC I-CLASS chromatography system (chromatographic column: Waters UPLC HSS T3 (1.8 μm, 2.1 mm × 150 mm); mobile phase: Phase A [water (W6-4, Fisher Scientific, Waltham, MA), 0.1% formic acid (A117-50, Fisher Scientific, Waltham, MA)], Phase B [acetonitrile (A955-4, Fisher Scientific, Waltham, MA)]; elution gradient: Supplementary Table S1; flow rate: 0.5 mL/min; injection volume: 5.0 μL; column temperature: 50 °C). Mass spectrometry analysis was subsequently performed using a tandem quadrupole mass spectrometer (XEVO TQ-S Micro, Waters Corp, Milford, MA, USA) (ion source voltage 1.5 kV, cone voltage 20 V; desolvation temperature 600 °C, desolvation gas flow rate 1,000 L/h; cone gas flow rate 10 L/h). The peak areas were then integrated using Masslynx software for mass spectrometry data collection (TargetLynx, Waters Corp, Milford, MA, USA), with a retention time tolerance of 15 s. The final concentrations were calculated using the standard curve method (20).

Thirty-eight amino acids were detected in both plasma and tears: L-histidine (HIS), 1-methyl-L-histidine (1-MEHIS), 3-MEHIS, L-hydroxyproline (OHPRO), L-carnosine (CAR), L-arginine (ARG), L-asparagine (ASN), O-phosphoethanolamine (PHETH), L-anserine (Ans), taurine (TAU), L-glutamine (GLN), L-serine (SER), ethanolamine (ETH), glycine (GLY), L-aspartic acid (ASP), L-citrulline (CIT), L-sarcosine (SAR), L-glutamic acid (GLU), β-alanine (beta-ALA), L-threonine (THR), L-alanine (ALA), γ-aminobutyric acid (gamma-GABA), δ-D, L-hydroxylysine (OH-LYS), L-α-aminoadipic acid (AMADP), L-proline (PRO), D, L-β-aminoisobutyric acid (beta-GABA), L-ornithine (ORN), L-lysine (LYS), L-α-aminobutyric acid (alpha-GABA), L-cystine (CYS), cystathionine (CYST), L-tyrosine (TYR), L-methionine (MET), L-valine (VAL), L-isoleucine (ILE), L-leucine (LEU), L-phenylalanine (PHE), and L-tryptophan (TRP).

2.2.8 Statistical analysis

Sample size calculation was performed based on previous literature (21). The following parameters were used: test power (1-β) = 0.9, significance level (α) = 0.05 (two-sided test), number of groups = 3, group allocation ratio = equal, and means multiplier = 1. The means ± σ values for the 23 amino acids detected were inputted into the calculation. The sample size was determined using PASS 15 software, yielding a total required sample size of 42 participants, with 14 participants per group, achieving a test power of 0.9204.

Data analysis was conducted using IBM SPSS Statistics 26.0. Missing values were imputed using multiple imputation. Quantitative variables with a normal distribution were presented as means ± standard deviation (), with intergroup comparisons performed using Analysis of Variance (ANOVA). Post-hoc pairwise comparisons were conducted using the Student–Newman–Keuls (SNK) method. For quantitative variables that did not conform to a normal distribution, data were expressed as [M (IQR)], and intergroup comparisons were performed using the rank sum test, followed by post-hoc pairwise comparisons with the Bonferroni correction method. Qualitative data were expressed as counts (%), with intergroup comparisons made using Fisher’s exact probability test. Pearson correlation was used to analyze relationships between normally distributed quantitative variables, while Spearman correlation was applied for associations involving non-normally distributed or qualitative variables. False Discovery Rate (FDR) correction was used to account for the multiple amino acids tested. Receiver operating characteristic (ROC) curve analysis were used to assess the predictive ability of MET in tear fluid and 1-MEHIS in plasma, with cut-off values determined by the Youden Index. A p-value < 0.05 was considered statistically significant.

3 Results

3.1 Demographics

A total of 50 participants (50 eyes) were enrolled in this study, comprising 15 individuals with DED with ocular pain, 19 with DED, and 16 in the control group (Table 1). The gender distribution was 24% male and 76% female, with an average age of 32.38 ± 8.14 years. The average duration of DED was 24.00 (0.00, 45.00) months, and the average sleep duration was 7.00 (7.00, 8.00) hours per day. No significant difference was observed between the DED with ocular pain [36.00 (12.00, 64.00) months] and the DED groups [24.00 (24.00, 60.00) months] (H = 33.409, p < 0.001) in the duration of DED.

3.2 Ocular examination and the OSDI and NRS scores of participants

The FBUT [2.00 (2.00, 3.00)] in the DED group was significantly lower than that in the DED with ocular pain group [3.00 (2.00, 3.00)], and the NEI score [1.00 (0.00, 3.00)] in the DED group was significantly higher than that in the control group [0.00 (0.00, 0.50)], with significant differences (all p < 0.05). The NRS score was significantly negatively correlated with Schirmer I test (r = −0.329, p < 0.05), while it was positively correlated with OSDI score (r = 0.446, p = 0.001), but not significantly correlated with NEI score and FBUT (both p > 0.05) (Table 2).

3.3 IVCM parameters in participants

Compared to the control group, both the DED with ocular pain and DED groups exhibited reductions in IVCM parameters (Table 3). Specifically, the DED with ocular pain group demonstrated significantly lower values in CNFD (11.43 ± 2.42), CNBD (11.30 ± 5.60), CNFL (10.00 ± 1.46), and CTBD (25.72 ± 9.75) compared to the control group (CNFD: 16.12 ± 5.39, CNBD: 18.87 ± 8.62, CNFL: 12.23 ± 2.50, CTBD: 36.73 ± 12.74), with all comparisons showing statistical significance (all p < 0.05) (Figure 1).

Figure 1

3.4 The detection results of amino acids in tear fluid

A total of 29 amino acids were detected in the tear fluid (Table 4; Figure 2). Significant differences were noted in the levels of ETH (H = 8.785, p = 0.012), OH-LYS (H = 36.684, p < 0.001), and MET (F = 6.396, p = 0.003) across the groups. The MET concentration in the DED with ocular pain group (0.97 ± 0.92) was significantly higher than that in both the DED group (0.34 ± 0.22) and the control group (0.36 ± 0.37) (all p < 0.05). In contrast, OH-LYS levels in the DED group [0.10 (0.10, 0.11)] were lower than those observed in the other two groups [DED with ocular pain group: 0.19 (0.14, 0.30); control group: 0.18 (0.13, 0.21)], and ETH levels [1.58 (1.07, 2.81)] were lower in the DED group than in the control group [3.70 (2.54, 4.89)], with all differences reaching statistical significance (all p < 0.05).

Figure 2

After applying FDR correction, tear fluid MET also remained a significant biomarker after correction (q-value = 0.044), though with a more moderate effect size.

3.5 The detection results of amino acids in plasma

A total of 36 amino acids were detected in the plasma (Table 5; Figure 3). Significant differences were found in HIS (H = 9.504, p = 0.024), 1-MEHIS (H = 17.596, p < 0.001), and CYST (H = 9.504, p = 0.008) levels across the three groups. The 1-MEHIS concentration in the DED with ocular pain group [4.99 (4.46, 6.48)] was significantly higher than that in the other two groups [DED group: 3.66 (2.84, 4.47); control group: 3.15 (1.98, 3.90)], with all differences being statistically significant (all p < 0.05). Furthermore, HIS levels in the DED group (74.26 ± 7.77) were higher than those in the DED with ocular pain group (73.95 ± 6.02), and CYST levels [2.21 (1.78, 2.39)] were higher in the DED group than in the control group [1.81 (1.42, 1.84)], with all differences statistically significant (all p < 0.05).

Figure 3

After applying FDR correction, plasma 1-MEHIS continued to demonstrate a highly significant association with DED with ocular pain (q-value < 0.001), indicating superior diagnostic potential.

3.6 The correlation between MET in tear fluid, 1-MEHIS in plasma, other clinical variables, and the DED with ocular pain

The results of the correlation analysis (Table 6) showed that the MET in tear fluid exhibited negative correlations with CNFD (r = −0.430, p < 0.05), CNBD (r = −0.379, p < 0.05), CNFL (r = −0.399, p < 0.05), CTBD (r = −0.345, p < 0.05) and CNFA (r = −0.312, p < 0.05), while positively correlated with age (r = 0.301, p < 0.05). In plasma, 1-MEHIS was positively correlated with DED duration (r = 0.346, p < 0.05), OSDI score (r = 0.396, p < 0.05), and NRS score (r = 0.556, p < 0.05), while negatively correlated with Schirmer I test results (r = −0.302, p < 0.05). To mitigate potential gender bias, t-tests revealed no statistically significant differences in MET and 1-MEHIS levels between male and female participants (all p > 0.05).

3.7 ROC curve analysis for MET in tear fluid and 1-MEHIS in plasma

As MET in tear fluid and 1-MEHIS in plasma showed statistically significant differences among the three groups, ROC analyses were conducted (Figure 4). The area under the ROC curve (AUC) was 0.686 for MET and 0.869 for 1-MEHIS. The 95% confidence interval (CI) for the AUC of MET ranged from 0.495 to 0.877 (p = 0.039), whereas that of 1-MEHIS ranged from 0.770 to 0.967 (p < 0.001). Using the Youden index, the optimal cut-off value for MET was determined to be 0.770, yielding a sensitivity of 53.3% and a specificity of 94.3%. For 1-MEHIS, the optimal cut-off value was 3.779, achieving 100% sensitivity and 62.9% specificity.

Figure 4

4 Discussion

Metabolomics analysis of amino acids in tear fluid and plasma was conducted using ultra-performance liquid chromatography in this study. Notably, for the first time, significant elevations of MET in tears and 1-MEHIS in plasma were identified in patients with DED with ocular pain. After our rigorous statistical analysis, which included FDR correction, confirms 1-MEHIS in plasma as a robust biomarker candidate, showing remarkable resilience to multiple testing corrections. The significance of MET in tear fluid, while slightly attenuated after correction, still supports its potential role, albeit one that may require further investigation in larger cohorts to fully elucidate.

Amino acids are essential components for forming various ocular structures and maintaining their physiological functions. For example, collagen fibers in the corneal stroma, primarily composed of proline, lysine, glycine, and leucine, provide strength and transparency to the cornea (22). Abnormalities in amino acid metabolism or deficiencies in specific amino acids may contribute to ocular diseases (7, 8, 10, 23). Our study revealed that the concentrations of amino acids in tear fluid were lower than in plasma, a finding consistent with prior research (21). However, 29 amino acids were identified in tear fluid and 36 in plasma, surpassing previous studies in terms of the number of amino acids detected. This improvement can be attributed to the enhanced precision and sensitivity of modern mass spectrometry techniques, as well as a growing recognition of the link between amino acids and various diseases (21). Given that autologous serum eye drops closely resemble the physiological composition of human tears, they have been explored as a therapeutic option for DED (24, 25). Our results indicate that certain amino acids in plasma are present at much higher concentrations than in tear fluid, which may explain the clinical efficacy of plasma-based therapies for ocular surface disorders. The therapeutic effects may be attributed to the regenerative and anti-inflammatory properties of specific amino acids, which promote tissue repair and reduce inflammation (25, 26).

MET, or L-MET, is an essential amino acid vital for normal human growth and development. MET plays multiple critical roles in protein metabolism (27–31). One key function involves its conversion to S-adenosylmethionine (SAM) via the transmethylation reaction, with SAM serving as the principal methyl donor for most methylation reactions in the body. Disruption of this process can lead to homocysteine accumulation, which impaired ganglion cell apoptosis, reduced lysyl oxidase activity, and increased oxidative stress (30, 31). In addition, MET participates in the production of L-cysteine through transsulfuration. Cysteine acts as the rate-limiting substrate for GSH synthesis, combining with glutamate and glycine to form the tripeptide GSH. Dysfunction of the antioxidant pathway regulated by glutathione can lead to oxidative stress, exacerbating inflammatory responses and contributing to the progression of DED (32, 33). Studies have also demonstrated that administering MET at twice the recommended dietary dose for seven consecutive days in adult mice enhances microglial activation and inflammation in the brain while reducing hippocampal neurogenesis (34).

In our study, statistical differences were observed in CNFD, CNBD, CNFL and CTBD, particularly between the DED with ocular pain group and the control group. Our results also suggest that patients with DED with ocular pain may have a certain dissociation between symptoms (NRS) and signs (NEI score), which is consistent with neuropathic features and may indicate non-noceptive (e.g., neuropathic) mechanisms. These findings are consistent with previous studies (35, 36). In addition, our correlation analysis revealed that MET in tears is particularly negatively correlated with CNFD, CNBD, CNFL, CTBD and CNFA. These findings suggest that elevated MET levels in tears may disrupt amino acid metabolism in the ocular surface, contributing to corneal nerve damage and a reduction in corneal nerve density, fiber length, and area, which may ultimately lead to ocular pain. ROC curve analysis further supported its clinical relevance, yielding an AUC of 0.686, with an optimal cut-off value of 0.770, demonstrating 53.3% sensitivity and 94.3% specificity. Despite limited sensitivity, high concentrations of MET in tears are a highly specific indicator of DED with ocular pain. When elevated concentrations are detected, they strongly suggest the presence of a pathologic condition called ocular pain. This suggests that MET may be involved in specific biological pathways involved in the development of ocular pain. The underlying mechanisms of amino acid regulation in the tears of patients with DED with ocular pain remain poorly understood. Detailed information on the mechanisms of amino acid transport in corneal epithelial cells or tears is lacking in patients with DED or ocular pain. Thus, further experimental studies are needed to elucidate the role of MET in tear fluid in the pathophysiology of DED with ocular pain. Besides, the collection of tears for MET detection is more convenient and straightforward compared to IVCM examination, which has drawbacks such as the elevated requirement of patient cooperation and long duration. It can play a surveillance role in assessing changes in corneal nerves in patients with DED with ocular pain patients, and whether the disease progresses from DED to DED with ocular pain or even DED with ocular neuropathic pain.

In plasma, the content of 1-MEHIS in the DED with ocular pain group was significantly higher than in the other two groups. 1-MEHIS, a component of the dipeptide anserine (β-alanyl-N1-methyl-histidine), is primarily synthesized in animal muscle tissue, which explains its detection in plasma but not in tear fluid. Research on 1-MEHIS is predominantly focused on nutrition, with existing studies confirming a positive correlation between plasma levels of 1-MEHIS and the intake of animal proteins such as meat and fish (37, 38). In our study, ROC curve analysis of 1-MEHIS yielded an AUC of 0.877, with 100% sensitivity and 62.9% specificity, indicating a significant association between 1-MEHIS and DED with ocular pain. The high sensitivity and relatively low specificity observed may be attributed to its biological susceptibility to variations in dietary protein intake. Due to the lack of data on participants’ recent dietary intake of meat or fish, it is unclear whether the increased 1-MEHIS levels in plasma are related to recent animal protein consumption or the disease itself. Given that DED is a chronic condition with pathogenesis closely linked to diet, previous studies have shown that diets like the Mediterranean diet, characterized by lower meat consumption, can help mitigate DED (39, 40). Our results also suggest that plasma 1-MEHIS levels and NRS score influence each other, and the reason may be that ocular pain caused by chronic DED and its severity may be related to long-term high animal protein intake. Although the detection of plasma 1-MEHIS is an invasive procedure with certain limitations for clinical application, its 100% diagnostic sensitivity warrants further evaluation in prospective studies under strictly controlled dietary conditions.

Notably, MET in tear fluid and 1-MEHIS in plasma demonstrate complementary diagnostic performance. This observation suggests the potential development of a two-step biomarker strategy: first, using the high sensitivity of 1-MEHIS for initial screening to minimize false-negative results, followed by the high specificity of MET to confirm positive cases and reduce false-positive outcomes. Such a combined approach may significantly improve overall diagnostic accuracy compared to the use of a single biomarker.

This study has several limitations. First, although this study provides the most comprehensive amino acid profiling to date for patients with DED with ocular pain, the small sample size limits its conclusions and renders it exploratory in nature. Secondly, the Cochet-Bonnet or noncontact air-jet esthesiometry, which is not yet widely available in China, was not used in this study, preventing the assessment of the correlation between corneal sensitivity and amino acid concentrations. Thirdly, our study was not conducted in a Controlled Environment Laboratory (CELab), potentially introducing variability in amino acid detection.

In summary, our study has expanded the amino acid profiles in the tear fluid and plasma of patients with DED with ocular pain. A preliminary observation revealed a significant elevation in levels of MET in tears and 1-MEHIS in plasma, which were correlated with changes in corneal nerve morphology and NRS score, respectively, serving as possible biomarkers for DED with ocular pain. Future large-scale studies are still required to validate this conclusion.

References

• 1.

  Craig JP Nichols KK Akpek EK Caffery B Dua HS Joo CK et al . TFOS DEWS II definition and classification report. Ocul Surf. (2017) 15:276–83. doi: 10.1016/j.jtos.2017.05.008,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 2.

  Yang W Wang H Li J Chen Y Zhang Y Niu X et al . Demographic and lifestyle factors associated with dry eye disease in China: a cross-sectional multi-center study. Ocul Surf. (2024) 34:146–55. doi: 10.1016/j.jtos.2024.07.007

  • CrossRef
  • Google Scholar

• 3.

  Stapleton F Alves M Bunya VY Jalbert I Lekhanont K Malet F et al . TFOS DEWS II epidemiology report. Ocul Surf. (2017) 15:334–65. doi: 10.1016/j.jtos.2017.05.003,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 4.

  Hoffmann M Farrell S Colorado LH Edwards K . Discordant dry eye disease and chronic pain: a systematic review and meta-analysis. Contact Lens Anterior Eye. (2024) 47:102248. doi: 10.1016/j.clae.2024.102248,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 5.

  Ruiz-Lozano RE Soifer M Zemborain ZZ Azar NS Quiroga-Garza ME Murillo S et al . Deep-learning based analysis of in-vivo confocal microscopy images of the subbasal corneal nerve plexus' inferior whorl in patients with neuropathic corneal pain and dry eye disease. Ocul Surf. (2024) 34:241–6. doi: 10.1016/j.jtos.2024.08.002,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 6.

  Blanco-Vázquez M Vázquez A Fernández I Novo-Diez A Martínez-Plaza E García-Vázquez CI et al . Inflammation-related molecules in tears of patients with chronic ocular pain and dry eye disease. Exp Eye Res. (2022) 219:109057. doi: 10.1016/j.exer.2022.109057,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 7.

  Lim EW Fallon RJ Bates C Ideguchi Y Nagasaki T Handzlik MK et al . Serine and glycine physiology reversibly modulate retinal and peripheral nerve function. Cell Metab. (2024) 36:2315–2328.e6. doi: 10.1016/j.cmet.2024.07.021,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 8.

  Hu G Gu L Wang R Jian Q Lv K Xia M et al . Ethanolamine as a biomarker and biomarker-based therapy for diabetic retinopathy in glucose-well-controlled diabetic patients. Sci Bull. (2024) 69:1920–35. doi: 10.1016/j.scib.2023.12.053,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 9.

  Chen X Zhang C Tian L Wu L Jie Y Wang N et al . In situ metabolic profile and spatial distribution of ocular tissues: new insights into dry eye disease. Ocul Surf. (2022) 24:51–63. doi: 10.1016/j.jtos.2021.12.013,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 10.

  Chen X Rao J Zheng Z Yu Y Lou S Liu L et al . Integrated tear proteome and metabolome reveal panels of inflammatory-related molecules via key regulatory pathways in dry eye syndrome. J Proteome Res. (2019) 18:2321–30. doi: 10.1021/acs.jproteome.9b00149,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 11.

  Perumal N Funke S Pfeiffer N Grus FH . Proteomics analysis of human tears from aqueous-deficient and evaporative dry eye patients. Sci Rep. (2016) 6:29629. doi: 10.1038/srep29629

  • CrossRef
  • Google Scholar

• 12.

  Aqrawi LA Galtung HK Guerreiro EM Øvstebø R Thiede B Utheim TP et al . Proteomic and histopathological characterisation of sicca subjects and primary Sjögren's syndrome patients reveals promising tear, saliva and extracellular vesicle disease biomarkers. Arthritis Res Ther. (2019) 21:181. doi: 10.1186/s13075-019-1961-4,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 13.

  Miller KL Walt JG Mink DR Satram-Hoang S Wilson SE Perry HD et al . Minimal clinically important difference for the ocular surface disease index. Arch Ophthalmol. (2010) 128:94–101. doi: 10.1001/archophthalmol.2009.356,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 14.

  Satitpitakul V Kheirkhah A Crnej A Hamrah P Dana R . Determinants of ocular pain severity in patients with dry eye disease. Am J Ophthalmol. (2017) 179:198–204. doi: 10.1016/j.ajo.2017.05.009,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 15.

  Vázquez A Martínez-Plaza E Fernández I Sobas EM González-García MJ Enríquez-de-Salamanca A et al . Phenotypic characterization of patients developing chronic dry eye and pain after refractive surgery: a cross-sectional study. Ocul Surf. (2022) 26:63–74. doi: 10.1016/j.jtos.2022.07.010,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 16.

  Bron AJ Evans VE Smith JA . Grading of corneal and conjunctival staining in the context of other dry eye tests. Cornea. (2003) 22:640–50. doi: 10.1097/00003226-200310000-00008,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 17.

  Markoulli M Papas E Petznick A Holden B . Validation of the flush method as an alternative to basal or reflex tear collection. Curr Eye Res. (2011) 36:198–207. doi: 10.3109/02713683.2010.542867,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 18.

  Suhre K Arnold M Bhagwat AM Cotton RJ Engelke R Raffler J et al . Connecting genetic risk to disease end points through the human blood plasma proteome. Nat Commun. (2017) 8:14357. doi: 10.1038/ncomms14357,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 19.

  Liu C Lin MT Lee IXY Wong JHF Lu D Lam TC et al . Neuropathic corneal pain: tear proteomic and neuromediator profiles, imaging features, and clinical manifestations. Am J Ophthalmol. (2024) 265:6–20. doi: 10.1016/j.ajo.2024.03.015,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 20.

  Gray N Zia R King A Patel VC Wendon J McPhail MJ et al . High-speed quantitative UPLC-MS analysis of multiple amines in human plasma and serum via precolumn derivatization with 6-aminoquinolyl-N-hydroxysuccinimidyl carbamate: application to acetaminophen-induced liver failure. Anal Chem. (2017) 89:2478–87. doi: 10.1021/acs.analchem.6b04623,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 21.

  Nakatsukasa M Sotozono C Shimbo K Ono N Miyano H Okano A et al . Amino acid profiles in human tear fluids analyzed by high-performance liquid chromatography and electrospray ionization tandem mass spectrometry. Am J Ophthalmol. (2011) 151:799–808.e1. doi: 10.1016/j.ajo.2010.11.003,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 22.

  Rusciano D Roszkowska AM Gagliano C Pezzino S . Free amino acids: an innovative treatment for ocular surface disease. Eur J Pharmacol. (2016) 787:9–19. doi: 10.1016/j.ejphar.2016.04.029,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 23.

  Zhang Z Li L Zhang C Zhang P Fang Z Li J et al . Relationship between plasma amino acid and carnitine levels and primary angle-closure glaucoma based on mass spectrometry metabolomics. Exp Eye Res. (2023) 227:109366. doi: 10.1016/j.exer.2022.109366,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 24.

  Aggarwal S Colon C Kheirkhah A Hamrah P . Efficacy of autologous serum tears for treatment of neuropathic corneal pain. Ocul Surf. (2019) 17:532–9. doi: 10.1016/j.jtos.2019.01.009,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 25.

  Wang M Yennam S McMillin J Chen HH de la Sen-Corcuera B Hemmati R et al . Combined therapy of ocular surface disease with plasma rich in growth factors and scleral contact lenses. Ocul Surf. (2022) 23:162–8. doi: 10.1016/j.jtos.2021.09.003,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 26.

  Anitua E Muruzabal F de la Fuente M Riestra A Merayo-Lloves J Orive G . PRGF exerts more potent proliferative and anti-inflammatory effects than autologous serum on a cell culture inflammatory model. Exp Eye Res. (2016) 151:115–21. doi: 10.1016/j.exer.2016.08.012,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 27.

  Elango R . Methionine nutrition and metabolism: insights from animal studies to inform human nutrition. J Nutr. (2020) 150:2518S–23S. doi: 10.1093/jn/nxaa155,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 28.

  Ajith TA Ranimenon . Homocysteine in ocular diseases. Clin Chim Acta. (2015) 450:316–21. doi: 10.1016/j.cca.2015.09.007,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 29.

  Tawfik A Elsherbiny NM Zaidi Y Rajpurohit P . Homocysteine and age-related central nervous system diseases: role of inflammation. Int J Mol Sci. (2021) 22:6259. doi: 10.3390/ijms22126259,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 30.

  Koklesova L Mazurakova A Samec M Biringer K Samuel SM Büsselberg D et al . Homocysteine metabolism as the target for predictive medical approach, disease prevention, prognosis, and treatments tailored to the person. EPMA J. (2021) 12:477–505. doi: 10.1007/s13167-021-00263-0,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 31.

  Xiao H Wang J Barwick SR Yoon Y Smith SB . Effect of long-term chronic hyperhomocysteinemia on retinal structure and function in the cystathionine-β-synthase mutant mouse. Exp Eye Res. (2022) 214:108894. doi: 10.1016/j.exer.2021.108894,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 32.

  Bu J Liu Y Zhang R Lin S Zhuang J Sun L et al . Potential new target for dry eye disease-oxidative stress. Antioxidants. (2024) 13:422. doi: 10.3390/antiox13040422,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 33.

  Haddad JJ Harb HL . L-gamma-Glutamyl-L-cysteinyl-glycine (glutathione; GSH) and GSH-related enzymes in the regulation of pro- and anti-inflammatory cytokines: a signaling transcriptional scenario for redox(y) immunologic sensor(s)?Mol Immunol. (2005) 42:987–1014. doi: 10.1016/j.molimm.2004.09.029,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 34.

  Alachkar A Agrawal S Baboldashtian M Nuseir K Salazar J Agrawal A . L-methionine enhances neuroinflammation and impairs neurogenesis: implication for Alzheimer's disease. J Neuroimmunol. (2022) 366:577843. doi: 10.1016/j.jneuroim.2022.577843,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 35.

  Moein HR Akhlaq A Dieckmann G Abbouda A Pondelis N Salem Z et al . Visualization of microneuromas by using in vivo confocal microscopy: an objective biomarker for the diagnosis of neuropathic corneal pain?Ocul Surf. (2020) 18:651–6. doi: 10.1016/j.jtos.2020.07.004,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 36.

  Zhang Y Wu Y Li W Huang X . Semiautomated and automated quantitative analysis of corneal sub-basal nerves in patients with DED with ocular pain using IVCM. Front Med. (2022) 9:831307. doi: 10.3389/fmed.2022.831307,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 37.

  Kochlik B Gerbracht C Grune T Weber D . The influence of dietary habits and meat consumption on plasma 3-Methylhistidine-a potential marker for muscle protein turnover. Mol Nutr Food Res. (2018) 62:e1701062. doi: 10.1002/mnfr.201701062,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 38.

  Solvik BS Øyen J Kvestad I Markhus MW Ueland PM McCann A et al . Biomarkers and fatty fish intake: a randomized controlled trial in Norwegian preschool children. J Nutr. (2021) 151:2134–41. doi: 10.1093/jn/nxab112,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 39.

  Molina-Leyva I Molina-Leyva A Riquelme-Gallego B Cano-Ibáñez N García-Molina L Bueno-Cavanillas A . Effectiveness of Mediterranean diet implementation in dry eye parameters: a study of PREDIMED-PLUS trial. Nutrients. (2020) 12:1289. doi: 10.3390/nu12051289,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 40.

  Markoulli M Ahmad S Arcot J Arita R Benitez-Del-Castillo J Caffery B et al . TFOS lifestyle: impact of nutrition on the ocular surface. Ocul Surf. (2023) 29:226–71. doi: 10.1016/j.jtos.2023.04.003,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar