Tissue-Specific Antioxidant Capacity in Primary Open-Angle Glaucoma Patients

Abstract

Purpose

To evaluate antioxidant capacity in aqueous humor (AH), trabecular meshwork (TM), and lens epithelium (LE) from patients with primary open-angle glaucoma and cataract and to assess associations with race and disease severity.

Methods

In this prospective cross-sectional study, two surgical cohorts undergoing cataract with glaucoma surgery provided either AH or paired TM/LE. AH total reactive antioxidant potential and ascorbic acid were measured. In paired TM and LE from the same eye, qPCR assessed nuclear factor erythroid 2–related factor 2 (NRF2), catalase (CAT), and superoxide dismutase 2. For AH, group comparisons used Welch's t tests or Mann–Whitney tests, and severity-stratified analyses used Race × Severity two-way ANOVA with Šídák correction. For qPCR, within-eye differences were calculated as Δ(TM − LE); race comparisons used Welch's t tests for each marker, with Šídák correction across the three markers.

Results

AH was analyzed from 106 eyes (Black n = 37; White n = 69); no race differences remained after Race × Severity modeling with Šídák adjustment. Paired TM/LE qPCR was analyzed from 42 eyes (Black n = 21; White n = 21). In Black eyes, NRF2 and CAT were lower in TM relative to matched LE (both P < 0.0001), whereas no such paired tissue differences were detected in White eyes. The within-eye difference Δ(TM − LE) was more negative in Black than White eyes for NRF2 (P = 0.0001) and CAT (P = 0.0024). Superoxide dismutase 2 showed no race-associated difference.

Conclusions

AH antioxidant measures showed limited race-associated differences, whereas paired TM/LE profiling revealed tissue- and marker-specific expression differences most evident in Black patients. These compartment-specific signatures warrant validation in larger prospective cohorts.

Primary open-angle glaucoma (POAG) is one of the leading causes of irreversible blindness worldwide, characterized by progressive optic nerve damage and visual field loss.^1–3 Although the precise pathophysiology of POAG remains multifactorial, elevated IOP is a well-established and the only modifiable risk factor.⁴ Increasing evidence suggests that oxidative stress plays a significant role in the progression of glaucoma, particularly in the trabecular meshwork (TM), which regulates aqueous humor (AH) outflow and IOP.^5–8 Reactive oxygen species can impair the TM's ability to maintain normal outflow, leading to increased IOP and subsequent optic nerve damage.^6,9,10 As the disease progresses, oxidative damage may further compromise the integrity of ocular tissues such as the lens epithelium (LE) and the AH itself.^11–13

In recent years, racial disparities in POAG prevalence and severity have been highlighted in glaucoma research. Black individuals are disproportionately affected by POAG, with higher incidence rates, more aggressive disease progression, and worse clinical outcomes compared with their White counterparts.^14–17 However, the underlying biological mechanisms contributing to these disparities remain unclear. One area of interest is the potential variation in defense mechanisms of oxidative stress/damage between racial groups, particularly in ocular tissues critical to POAG pathophysiology, such as the TM and LE.^8,18–20

In our previous study, we measured AH levels of 8-hydroxy-2′-deoxyguanosine (8-OHdG), a marker of DNA oxidative damage. The results showed significantly higher levels of 8-OHdG in Black patients with severe stage of glaucoma, whereas no statistical differences were observed in nonglaucoma (cataract-only) patients or in patients with mild or moderate glaucoma.⁸ This finding prompted us to further investigate the antioxidant status in human eyes.

Antioxidants play a crucial role in neutralizing reactive oxygen species and limiting oxidative damage in anterior segment tissues. Core enzymatic defenses include mitochondrial superoxide dismutase 2 (SOD2) and catalase (CAT), as well as the transcription factor nuclear factor erythroid 2–related factor 2 (NRF2), a master regulator of antioxidant responses.^21–24 In this study, we evaluated five complementary antioxidant-related parameters across three anterior segment components—AH, TM, and LE—in Black and White patients with POAG undergoing combined glaucoma (goniotomy) and cataract surgery.²⁵ Specifically, we measured AH total reactive antioxidant potential (TRAP) and AH ascorbic acid (AsA), and quantified tissue expression of NRF2, CAT, and SOD2 by qPCR in paired TM and LE from the same eye. By integrating soluble antioxidant capacity in AH with tissue-specific antioxidant gene expression, we aimed to determine whether antioxidant defenses differ by race and whether any differences are tissue specific. Such tissue-resolved variation in antioxidant protection could help to clarify biological contributors to POAG disparities and inform links between oxidative stress, TM dysfunction, and cataract-related changes.

Methods

Study Design

This study prospective, cross-sectional analysis aimed to evaluate racial disparities and disease severity in antioxidant capacity among POAG patients. Patients were recruited from a single tertiary care center, and the study was conducted in accordance with the Declaration of Helsinki. Washington University Institutional Review Board approval was obtained, and informed written consent was provided by all participants.

Eligibility criteria included a confirmed diagnosis of POAG according to the patient's visual field and/or optic nerve findings and age of 50 years and older. Exclusion criteria included any previous intraocular surgery, other types of glaucoma, history of ocular trauma, active proliferative diabetic retinopathy, or a history of inflammatory eye disease. IOP measurements from the most recent preoperative visit were recorded. Clinical information such as antiglaucoma medications, cataract types, and cataract grading, as well as systemic diseases, was collected. AH samples were obtained from patients who underwent glaucoma and/or cataract surgery. TM and LE specimens were collected from patients undergoing combined cataract and glaucoma (goniotomy) surgery. The study population was stratified by race (self-reported as Black or White) and glaucoma severity based on clinical evaluation, using the Hodapp–Parrish–Anderson stages: mild, moderate, and severe based on visual field testing as outlined in our previous reports.^8,26

Sample Collection

AH Collection

AH samples were collected immediately prior to the surgery procedures. Under sterile conditions, a ∼100 µL sample of AH was aspirated from the anterior chamber using a 30G cannula on a tuberculin syringe through a corneal paracentesis site, taking care to avoid contamination and minimizing the risk of damage to surrounding ocular structures. The samples were immediately placed on dry ice and subsequently transferred to a liquid nitrogen tank for storage until analysis. Only one eye sample per patient was used for AH analysis.

TM and LE Collection

Paired TM and LE specimens were collected from the same surgical eyes undergoing combined cataract and goniotomy surgery to enable direct comparisons.

We collected TM using ab interno goniotomy with the Kahook Dual Blade (KDB, New World Medical, Rancho Cucamonga, CA, USA) or SION Surgical Instrument (Sight Sciences, Menlo Park, CA, USA). Under direct gonioscopic visualization, a strip of TM tissue was excised, typically ranging from 90 to 120 degrees of the angle, which corresponds with approximately 3 to 4 clock hours, yielding a strip of roughly 4 to 5 mm in length. The identity of these TM specimens was confirmed using morphological and marker-based criteria.²⁷ Figure 1 shows that a representative subset of specimens was examined by light microscopy to verify characteristic trabecular beam/meshwork architecture and specimen orientation with minimal evidence of adjacent tissue contamination.

Figure 1.

Goniotomy surgical specimens. (A) Illustration of goniotomy using the Kahook Dual Blade (KDB, New World Medical). The red circles highlight the dual-blade design, featuring a pointed tip and ramp structure that enables parallel incisions and excision of the TM with minimal residual tissue. (Images modified with permission from New World Medical). (B) Hematoxylin and eosin (H&E) staining of fresh-frozen goniotomy specimens, showing the TM structure in cross-section. (C) H&E staining of en face sections demonstrates the detailed architecture of the excised TM tissue. (D) Immunofluorescence staining of paraffin-embedded sections reveals strong αB-crystallin (green) expression in the juxtacanalicular region, with weaker labeling in the corneoscleral TM. Aquaporin 1 (red) is expressed throughout the TM. DAPI (blue) stains nuclei. The SC side is indicated.

LE specimens were obtained by the capsulorrhexis procedure during cataract extraction. These specimens were consistently of high purity of LE and did not require cell confirmation. Both TM and LE samples were immediately placed in RNeasy lysis buffer (Qiagen, Hilden, Germany) respectively, to preserve RNA integrity and quickly frozen on dry ice in the operating room, then stored at −80°C until further analysis.

Antioxidant Measurements

TRAP and AsA in AH samples

AH antioxidant levels were evaluated using two primary metrics: TRAP and AsA, as fully detailed in our previous study.²⁸ Briefly, the TRAP assay measures the ability of a sample to neutralize free radicals. A solution containing 2,2′-azobis(2-amidinopropane) and luminol is prepared, generating free radicals at a steady rate. Antioxidants in the AH (50 µL) added to the solution reduce light emitted by the reaction with luminol and compared with a reference standard. Results are reported in Trolox units, representing the sample's antioxidant activity. Owing to limited AH volume, each patient sample was assayed once.

AsA levels were quantified based on iron-reduction using colorimetric analysis of 10-µL samples as described in our published study.²⁸ Each AH sample per patient was measured in triplicate using mean levels for statistical analysis.

RNA Extraction and qPCR Analysis for TM and LE Samples

Total RNA was extracted from paired TM and LE samples using the RNeasy Micro Kit (Qiagen). Tissues were homogenized, and RNA was purified using spin columns, with on-column DNase I treatment to minimize genomic DNA contamination. RNA quantity and integrity were measured by a 2100 bioanalyzer (Agilent Technologies, Santa Clara, CA, USA), and only samples with high-quality RNA (RNA integrity number of ≥8) were used for further analysis. Five nanograms of total RNA was reverse transcribed and amplified to cDNA using Ovation Pico WTA System V2 kit (Tecan, Männedorf, Switzerland). TM samples that did not meet this criterion were excluded. Because goniotomy-derived TM tissue yields are limited, three antioxidant-related targets (NRF2, CAT, and SOD2) were selected for expression analysis. Protein-based assessment of phosphorylated NRF2 and protein levels of CAT and SOD2 were not feasible in this cohort owing to limited tissue yield. Gene expression relative to GAPDH of NRF2, CAT, and SOD2 was quantified by qPCR. qRT-PCR amplification was performed using a SYBR Green JumpStart TaqReadyMix (Sigma-Aldrich, Saint Louis, MO, USA) and a CFX^tm96 real-time PCR System (Bio-Rad, Hercules, CA, USA). Reactions were performed in triplicate, and relative gene expression was calculated using the ΔCt method, normalized to GAPDH. NRF2, SOD2, and CAT, and their respective primer designs are listed in Table 1.

Table 1.

Primer Designs for qPCR

NRF2	Forward: 5′ CTCAACCAGCTTGTCATTTTC 3′
	Reverse: 5′ CCCTGTTGATTTAGACGGTA 3′
SOD2	Forward: 5′ TCTAGTCTTTTGCCCCAGTTAC 3′
	Reverse: 5′ CCAGAAGTATCACTGCAGAGG 3′
CAT	Forward: 5′ CTCAGGTGCGGGCATTCT 3′
	Reverse: 5′ CAATGTTCTCACACAGACGTTTCC 3′
GAPDH	Forward: 5′ CAGCCGAGCCACATCGCTCAGACACCAT 3′
	Reverse: 5′ CTTTACCAGAGTTAAAAGCAGCCCTGGTGACCA 3′

Statistical Analysis

All analyses were performed in GraphPad Prism 10 (GraphPad Software, La Jolla, CA, USA) using two-tailed tests with an α of 0.05. Data are presented as mean ± SD. Normality was assessed using the Shapiro–Wilk test. For baseline characteristics (Tables 2, 3), continuous variables were compared by race using unpaired t tests with Welch's correction, and categorical variables (sex; POAG severity-stage distributions) were compared using exact tests (Fisher's exact test; Fisher–Freeman–Halton exact test for 3 × 2 tables).

Table 2.

Demographic and Clinical Characteristics of Patients for TRAP and AsA Analysis of AH Samples

	No.	Black	White	P Value
Total POAG
Age (years)	106	70.9 ± 9.8	74.2 ± 10.2	0.124
Male*		17	22	0.205
Female*		20	47
POAG severity
Mild
Age (years)		78.7 ± 10.9	71.1 ± 8.5	0.102
IOP (mm Hg)		18.3 ± 7.5	19.8 ± 6.0	0.646
Male	33 (31.1)	2	8
Female		6	17
Moderate
Age (years)		69.2 ± 5.6	66.8 ± 9.9	0.389
IOP (mm Hg)		16 ± 5.0	15.3 ± 4.1	0.533
Male	33 (31.1)	5	8
Female		6	14
Severe
Age (years)		68.5 ± 10.0	74.2 ± 10.2	0.085
IOP (mm Hg)		18.4 ± 4.0	16.2 ± 4.4	0.199
Male	40 (37.7)	10	6
Female		8	16

Data are shown for the full cohort (total POAG) and stratified by POAG severity (mild, moderate, or severe). Continuous variables (age, IOP) are presented as mean ± SD; categorical variables (sex) are presented as counts. Race–group comparisons for continuous variables were performed using two-tailed unpaired t tests. Sex distribution between Black and White patients was compared using a two-tailed Fisher's exact test (P value indicated by *).

^*Two-tailed Fisher's exact test for sex (overall) comparing Black vs. White in the full cohort.

Table 3.

Demographic and Clinical Characteristics of Paired-Tissue qPCR Cohort by Race

	Black	White	P Value
Age (years)	69.1 ± 9.0	72.8 ± 4.3	0.106
No.	21	21
Sex*
Male	10 (47.6)	9 (42.9)	1.000
Female	11 (52.4)	12 (57.1)
POAG† severity
Mild	8 (38.1)	3 (14.3)	0.249
Moderate	10 (47.6)	13 (61.9)
Severe	3 (14.3)	5 (23.8)

Patient age is presented as mean ± SD, and categorical variables are shown as number (%). Race-group comparisons for age were performed using a two-tailed unpaired t test. Sex distribution between Black and White patients was compared using a two-tailed Fisher's exact test (P value indicated by *). Differences in the distribution of POAG severity stages between races were assessed using a Fisher–Freeman–Halton exact test for a 3 × 2 contingency table (P value indicated by †).

^*Two-tailed Fisher's exact test for sex (overall).

^†Fisher–Freeman–Halton exact test for POAG severity distribution (overall).

For AH Antioxidants

Overall race comparisons used Mann–Whitney U tests when normality was not met. Within each severity stage, race comparisons were performed using unpaired t tests with Welch's correction to obtain raw P values (p_raw). AH outcomes were also evaluated using two-way ANOVA with factors Race, Severity, and the Race × Severity interaction, followed by Šídák multiple comparisons testing for Black vs. White within each severity stage to obtain adjusted P values (p_adj). Associations between TRAP and AsA were assessed by Pearson correlation. Multivariable regression was used to evaluate potential confounding by age, sex, race, and medication class.

For Paired Tissues

Within-eye differences were computed for each marker as Δ(TM – LE) = (fold change in TM) − (fold change in LE). TM vs. LE comparisons were assessed within each race group using paired t tests for each gene. Between-race differences in Δ(TM – LE) were evaluated using unpaired t tests with Welch's correction, with Šídák adjustment across the three genes (NRF2, CAT, and SOD2).

Results

AH Study: TRAP and AsA

A total 106 AH specimens from 106 patients diagnosed with POAG (Black: n = 37; White: n = 69) were analyzed using TRAP and AsA assays (Table 2). Baseline characteristics were comparable between races in the full cohort, with no significant differences in age (70.9 ± 9.8 years vs. 74.2 ± 10.2 years; P = 0.124) or sex distribution (P = 0.205). After stratification by visual field severity, age and preoperative IOP were also not significantly different between Black and White patients within any subgroup (all P > 0.05).

Normality was evaluated using the Shapiro–Wilk test within each race subgroup for each endpoint. Because distributions were approximately normal within severity subgroups (with one borderline case), between-race comparisons were performed using two-tailed unpaired t tests with Welch's correction. In severe POAG, TRAP was lower in Black than White patients (468.7 ± 105.7 Trolox units vs. 559.7 ± 144.8 Trolox units; P = 0.027), and AsA was also lower in Black patients (1.61 ± 0.48 mM vs. 1.97 ± 0.70 mM; P = 0.042). However, these were nominal (unadjusted) subgroup findings and did not remain statistically significant after adjustment using a two-way ANOVA model including Race, Severity, and the Race × Severity interaction and Šídák multiple comparisons testing, indicating that the apparent severe-stage differences were not robust after multiplicity adjustment across severity strata for either TRAP or AsA (Fig. 2). TRAP correlated with AsA in White patients (Pearson r = 0.63; P = 0.000), but not in Black patients (r = 0.29; P = 0.079) (Supplementary Fig. S1). In multivariable regression analyses, no covariates attained statistical significance (results not shown).

Figure 2.

Race differences in AH antioxidant markers overall and by POAG severity. (A1) TRAP and (B1) AsA measured in AH from Black (red) and White (blue) POAG patients across all severity stages. Each dot represents one eye/patient. Overall Black vs. White comparisons were performed using a two-tailed Mann–Whitney U test (overall distributions were non-normal by Shapiro–Wilk). (A2) TRAP and (B2) AsA stratified by POAG severity (mild, moderate, or severe). Within each severity stratum (approximately normal by Shapiro–Wilk), p_raw values were calculated using two-tailed unpaired t tests with Welch's correction (Black vs. White within each severity level). p_adj values were obtained from a two-way ANOVA model including Race, Severity, and the Race × Severity interaction, followed by Šídák multiple-comparisons testing for Black vs. White within each severity level. In (A2/B2), p_raw values shown in red indicate P < 0.05. Data are presented as mean ± SD.

Ocular Tissue Study: qPCR Quantification of Antioxidant Enzymes in Paired TM and LE Specimens

A total of 42 paired tissue specimens from 42 patients with POAG undergoing combined POAG (goniotomy) and cataract surgery were included in this analysis. Demographic characteristics are summarized in Table 3 (Black: n = 21; White: n = 21). The two racial groups were similar in age (69.1 ± 9.0 vs. 72.8 ± 4.3 years; P = 0.106), sex distribution (P = 1.000), and visual field severity distribution (P = 0.249). TM and LE samples were obtained from the same surgical eye, and only high-quality RNA (RNA integrity number of ≥8.0) from well-identified, minimally contaminated TM specimens (Fig. 1) and corresponding LE samples were included. Expression of NRF2, CAT, and SOD2 was quantified by qPCR, normalized to GAPDH, and analyzed using within-eye paired comparisons and race-stratified analyses.

Among Black patients, normalized NRF2 and CAT expression was significantly lower in TM than in the paired LE (both P < 0.0001) (Fig. 3), indicating a pronounced within-eye TM-LE difference in these antioxidant-related transcripts. In contrast, White patients showed no significant TM–LE differences for NRF2 or CAT (P = 0.995 and P = 0.912, respectively) (Fig. 3). For SOD2, TM expression showed a borderline decrease relative to LE in Black patients (P = 0.053) and no difference in White patients (P = 0.349).

Figure 3.

Paired comparison of antioxidant gene expression between LE and TM grouped by race. Normalized qPCR expression levels (fold change) of NRF2 (top), CAT (middle), and SOD2 (bottom) are shown for Black (left) and White (right) POAG patients. For each subject, matched LE and TM samples were obtained from the same surgical eye and analyzed as a paired set; paired values are connected by a line (each line and dot color corresponds with one patient, linking that patient's TM and LE values). P values shown in each panel were calculated using a two-tailed paired t test comparing LE vs. TM within subjects.

To determine whether the within-eye tissue contrast differed by race, we calculated a within-eye difference for each gene as Δ(TM – LE) = (expression fold change in TM) − (expression fold change in LE) and compared Δ(TM – LE) values between races (Fig. 4). This within-eye contrast focuses on tissue-relative differences and reduces interindividual variability compared with TM-only or LE-only comparisons. The Δ(TM – LE) was significantly more negative in Black than White patients for NRF2 (p_raw = 0.0001; p_adj = 0.0004) and CAT (p_raw = 0.0024; p_adj = 0.0048), indicating a larger within-eye TM–LE differential gene expression (TM expression of these two antioxidant genes lower than matched LE expression) in Black eyes. In contrast, the Δ(TM – LE) for SOD2 did not differ by race (p_raw = 0.3532; p_adj = 0.3506). The p_adj values reflect Šídák correction across NRF2, CAT, and SOD2 (p_raw from the Welch-corrected unpaired t test).

Figure 4.

Racial differences in within-eye TM–LE antioxidant marker expression (Δ[TM − LE]). For each surgical eye, paired TM and LE samples were collected and a within-eye difference was calculated for each marker as Δ(TM – LE) = (expression fold change in TM) − (expression fold change in LE). Negative Δ(TM – LE) values indicate lower expression in TM than in matched LE, whereas values near 0 indicate similar expression in the two tissues. Violin plots show the distribution of Δ(TM – LE) values for Black and White patients for NRF2, CAT, and SOD2; white points (open circles) represent individual eyes, and the horizontal dashed line at 0 denotes no TM–LE difference. For each marker, p_raw values were obtained from two-tailed unpaired t tests with Welch's correction comparing Δ(TM–LE) between Black and White eyes. p_adj values reflect Šídák-adjusted P values accounting for comparisons across the three markers (NRF2, CAT, and SOD2).

Discussion

This study provides a tissue-focused assessment of antioxidant capacity in POAG patients by measuring (1) TRAP and AsA in AH and (2) NRF2, CAT, and SOD2 in TM and LE samples from the same eye. Oxidative stress and mitochondrial dysfunction are widely implicated in POAG, yet human tissue-level evidence—particularly in racially stratified cohorts—remains limited.^6,9–12,29 By integrating biomarkers in AH with tissue-specific qPCR data, we provide a tissue-specific view of oxidative stress and antioxidant regulation within the ocular anterior segment.

A major strength of this study is intraoperative TM collection by excisional goniotomy, yielding fresh TM with minimal adjacent-tissue contamination and RNA suitable for qPCR (Fig. 1).^27,30,31 Pairing TM with LE from the same eye reduces interindividual variability and enables within-eye contrasts that may better resolve localized patterns than AH alone.

In AH, TRAP and AsA showed limited race-associated differences after adjustment. This finding is consistent with AH functioning as a composite compartment influenced by multiple ocular tissues and systemic/treatment factors, which can dilute localized signals and increase variability in modest cohorts.^28,32–34 Prior work, including our studies, supports medication- and oxygen-related influences on AH antioxidant measures.^28,35 Together, these considerations suggest that tissue-based readouts may be more sensitive than AH biomarkers for detecting subtle, site-specific differences.¹²

In contrast, paired tissue analyses revealed compartmentalization of antioxidant-related gene expression. In Black eyes, NRF2 and CAT showed a greater TM–LE separation (Fig. 3). Because these are relative qPCR measures, they reflect within-eye differences between paired tissues and do not establish absolute mRNA content changes in the TM or LE, nor do they directly quantify NRF2 pathway activity. However, they are consistent with tissue-specific regulatory demands and a TM-centered pattern in a subset of patients.^12,21,36,37 We emphasized Δ(TM – LE) because it reduces between-subject noise compared with TM-only or LE-only comparisons (Fig. 4). The Δ(TM – LE) was computed as the difference in normalized expression fold change between TM and its eye-matched LE. More negative Δ(TM – LE) values for NRF2 and CAT indicate a larger within-eye TM–LE gap (TM lower relative to matched LE) in Black eyes. These findings support tissue compartmentalization that may be obscured in AH, although causality cannot be inferred from this cross-sectional design.^6,12,29,37 The TM pattern contrasted with relatively preserved LE expression, and may be compatible with epidemiological observations that Black individuals, despite a higher glaucoma risk, do not necessarily exhibit a greater incidence or progression of age-related cataract.³⁸ This connection remains indirect and should be interpreted cautiously.

The TM is a pressure-regulating, contractile-like tissue with substantial mitochondrial demand, raising the possibility that its susceptibility to oxidative stress may share conceptual parallels with other load-bearing, energy-intensive cell types. In this context, well-documented Black–White disparities in cardiovascular outcomes provide a relevant population-level example; however, we emphasize that this comparison is intended only as a conceptual framework and does not imply a shared mechanism.³⁹

Because POAG is strongly linked to oxidative stress,^6,12 our data are consistent with selective vulnerability of TM rather than generalized ocular oxidative stress. These patterns are consistent with the biological role of the NRF2 signaling pathway (Fig. 5). Under oxidative stress, NRF2 dissociates from Keap1 and translocates to the nucleus, where it activates genes encoding key antioxidant enzymes, including CAT and SOD2.^36,37,40 Defining tissue-specific antioxidant regulation across racial groups may inform targeted therapeutic strategies.

Figure 5.

Regulation of NRF2 signaling pathway in response to oxidative stress. Under basal conditions, NRF2 is inactivated by its inhibitor, Keap1 (Kelch-like ECH-associated protein 1), leading to its ubiquitination and subsequent degradation via the proteasome. On the right, upon exposure to oxidative stress, reactive oxygen and nitrogen species such as hydroxyl radical (OH·), nitric oxide (NO), superoxide anion (O₂⁻), hydrogen peroxide (H₂O₂), singlet oxygen (¹O₂), and peroxyl radical (ROO·) trigger the dissociation of NRF2 from Keap1. This allows NRF2 to translocate into the nucleus, where it binds to the antioxidant response element (ARE) and activates the expression of antioxidative enzymes such as CAT, glutathione peroxidase (GPx), glutathione S-transferase (GST), and superoxide dismutase (SOD). These enzymes help mitigate oxidative damage, providing cellular protection. Red indicates the gene expressions evaluated in this study.

Study limitations include modest paired-tissue sample sizes and limited TM RNA yield precluding protein-level validation. The use of self-reported binary race categories may not reflect genetic ancestry. Prior studies^41–43 have indicated that self-reports may be unreliable and correlate poorly with true racial origins, because many individuals in the United States have a heterogeneous genetic background. Future studies incorporating ancestry-informative markers and/or mitochondrial haplogroups could refine stratification.^16,41–45 Glaucoma severity based on visual field status is a practical proxy for TM dysfunction, because outflow facility was unavailable, and preoperative IOP was influenced by multidrug therapy.²⁷

Conclusions

Our findings highlight the importance of tissue-specific analysis in understanding racial disparities in POAG. AH antioxidants (TRAP and AsA) showed limited race-associated differences, whereas paired TM–LE data revealed compartmentalized antioxidant-related regulation within the same eye. Expanded, racially diverse cohorts with protein/activity validation using refined ancestry/treatment covariates are needed to confirm these findings.