**3. Results**

Table 1 lists the demographic data of patients with a single risk factor, those with double risk factors, and the control group (with cataract). The mean age of single-risk-factor patients was 72.30 ± 10.14 years, for double-risk-factor patients was 69.38 ± 9.87 years, and for cataract control individuals was 74.00 ± 5.72 years. All patients had cataracts as revealed by a slit lamp examination. The mean protein concentrations were 0.36 ± 0.21 μg/μ<sup>L</sup> in the single-risk-factor group, 0.34 ± 0.11μg/μ<sup>L</sup> in the double-risk-factor group, and 0.22 ± 0.06 μg/μ<sup>L</sup> in the cataract control group. There were statistical differences among total protein contents in these three groups (*p* = 0.049) but no statistical differences in age among these groups (*p* = 0.390). In total, 136 proteins were successfully identified by LC-ESI MS/MS in single-risk-factor, double-risk-factor, and cataract control AH samples (Table 2, Figure 2).

**Table 2.** List of aqueous humor (AH) proteins identified by LC-ESI-MS/MS.



**Table 2.** *Cont.*

Comparing the single-risk group to the cataract control group, 125 proteins were found, which included 42 proteins that were present at higher expression levels and 83 proteins that were present at lower expression levels in the single-risk group. In the doublerisk group, as compared to the cataract control group, 124 proteins were disclosed, among which 39 proteins had higher expression levels and 85 proteins had lower expression levels in the double-risk group. To understand the biological meaning of the changes of protein expression observed in different risk factor groups, differentially expressed proteins were analyzed for "molecular functions", "biological processes", and "cellular components" by GO annotations. Our results demonstrated that differentially expressed proteins in the three groups had different molecular functions, biological processes, and cellular components (Figure 3). The major biological processes of these proteins were biological regulation, including immune responses, metabolic processes, and responses to stimuli of the AH (Figure 3A). The major molecular functions of AH proteins enriched among single-risk and

double-risk patients were antigen binding and enzyme inhibitor activity (Figure 3B). As per cellular component terms of the GO, most significant AH proteins were categorized as extracellular region proteins (Figure 3C). Then, we used Ingenuity Pathway Analysis (IPA, Qiagen) to show canonical pathways that are potentially involved in the pathogenesis of cataracts under the risks of diabetes and smoking. Table 3 lists pathways associated with AH proteins from single-risk patients, double-risk patents, and the cataract controls.

**Figure 2.** Label-free Nanoflow UHPLC-MS/MS analytical workflow for the proteomic analysis of human aqueous humor. Samples were digested using trypsin and were analyzed using an LTQ-Orbitrap DiscoveryTM hybrid mass spectrometer (Thermo Electron). Proteins were identified and quantified using the SEQUEST algorithm followed by analysis using Xcalibur 2.0 SR1 (Thermo Electron). The intersection of each area represents the number of significant expression (*p* < 0.05) proteins between each groups. Only one protein was significantly deferentially expressed in each group.

**Figure 3.** *Cont.*

**Figure 3.** Gene ontology (GO) analysis of differentially expressed proteins of the aqueous humor (AH) in the cataract control, single-risk, and double-risk groups. We compared identified AH proteins from the three groups: (**A**) biological processes; (**B**) molecular functions; (**C**) cellular components.

The top canonical pathways, including LXR/RXR activation, FXR/RXR activation, and acute-phase response signaling, demonstrated significant associations with AH proteins. Statistical analysis was performed on these 136 proteins. In total, 47 proteins exhibited statistically significant changes in content in the group with a single risk factor compared to the cataract control group (Table 4).


**Table 3.** Pathway analysis of aqueous humor (AH) proteins using IPA tools.

Single risk, patients with the DM or smoking risk factor; double risk, patients with both the DM and smoking risk factors; control, cataract patients with neither of these cataract risk factors.


**Table 4.** List of selected potential biomarker candidates.


**Table 4.** *Cont.*

Single risk, patients with the DM or smoking risk factor; double risk, patients with both the DM and smoking risk factors; control, cataract patients with neither of these cataract risk factors; Spc, spectral count.

> In a comparison of the double-risk-factor group with the cataract control group, 40 proteins were statistically significantly (*p* < 0.05) expressed (Table 4). Among the 51 proteins that were significantly changed, 10 proteins were increased in the single- or doublerisk groups, including 26S proteasome non-ATPase regulatory subunit 1, alpha-2-HSglycoprotein, apolipoprotein A-I, apolipoprotein A-II, apolipoprotein A-IV, apolipoprotein E, opticin, potassium voltage-gated channel subfamily S member 2, complement C4-A, and complement C4-B. Another 41 proteins exhibited decreased expression in the singleor double-risk groups compared to cataract controls (Table 4). In particular, alpha-2-HSglycoprotein was the only one that presented a significant change among all three of the groups (cataract control vs. single: *p* = 0.00338; cataract control vs. double: *p* = 0.00062; single vs. double: *p* = 0.03309), which demonstrated an increasing trend with increase in risk (Figure 4).

**Figure 4.** Proteomics analysis revealed significant concentration changes in the alpha-2-HSglycoprotein (SpC, spectral count) among the three groups. Single risk, patients with the diabetes mellitus (DM) or smoking risk factor; double risk, patients with both the DM and smoking risk factors; control, cataract patients with neither of these cataract risk factors.

Furthermore, we performed an ELISA analysis to determine the concentration of alpha-2-HS-glycoprotein. Compared to the cataract control group, the average concentration of alpha-2-HS-glycoprotein was significantly higher in single-risk-factor group (0.43 μg/mL) patients (0.16 μg/mL) (*p* = 0.002) (Figure 5).

**Figure 5.** ELISA analysis of significant concentration (μg/mL) changes of the alpha-2-HS-glycoprotein

between risk factor and cataract control groups. However, there was no significant concentration change between the single- and double-risk-factor groups. Single risk, patients with the diabetes mellitus (DM) or smoking risk factor; double risk, patients with both the DM and smoking risk factors; control, cataract patients with neither of these cataract risk factors.

Furthermore, the average concentration significantly increased in double-risk-factor group (0.43 μg/mL) patients compared to the cataract control group (0.16 μg/mL) (*p* < 0.001) (Figure 5). The ELISA analysis revealed significant concentration changes between the risk factor and cataract control groups. However, there was no significant concentration change between the single- and double-risk-factor groups. A subgroup analysis was performed to confirm that DM and smoking risk factors significantly influenced the ELISA concentration compared to the cataract control group (Figure 6).

**Figure 6.** (**A**) ELISA analysis of significant concentration (μg/mL) changes in the alpha-2-HSglycoprotein between the diabetes mellitus (DM) groups and cataract control group. (**B**) ELISA analysis of significant concentration (μg/mL) changes in the alpha-2-HS-glycoprotein between the smoking groups and cataract control group. DM group (*n* = 13): DM single-risk patients (*n* = 5) + double-risk patients (*n* = 8); Non-DM group (*n* = 14): smoking single-risk patients (*n* = 5) + cataract control group (*n* = 9); Smokers (*n* = 13): smoking single-risk patients (*n* = 5) + double-risk patients (*n* = 8); Non-smokers (*n* = 14): DM single-risk patients (*n* = 5) + cataract control group (*n* = 9).

In our study, we analyzed the aqueous protein contents of the AH samples of singlerisk and double-risk patients and a control group (with cataract) using label-free n-UPLC-MS/MS quantitation. We reported that in cataract patients with different risk profiles, 51 AH proteins were significantly changed compared to cataract controls. The alpha-2- HS-glycoprotein was significantly differently expressed between risk groups and cataract controls and could be a potential aqueous protein marker for detecting smoking and DM cataract risk factors. The increased levels of total protein concentrations were reported in the AH, which provides a possible marker to monitor the AH of cataract risk exposure. Note that additional studies exploring the roles of this protein in the development or the pathogenesis molecular pathway of cataracts would be beneficial. To our knowledge, this is the first study to analyze how cataract risk factors influenced the AH in the development of cataract disease. We reported that only one protein had significantly changed, which was the alpha-2-HS-glycoprotein; its expression increased in the presence of risk factors. Alpha-2-HS-glycoprotein, known as fetuin-A, was reported to be a systemic inhibitor of precipitation of basic calcium phosphate, thereby preventing unwanted calcification [15] and influencing the mineral phase of bone [16]. The alpha-2-HS-glycoprotein is synthesized in the liver, electively concentrated in the bone matrix, and secreted in plasma. The dysfunction of the gene represented by this entry is associated with alopeciamental retardation syndrome [17]. There was previous evidence demonstrating that the alpha-2-HS-glycoprotein was present in the rabbit AH following two different cataract surgery incision procedures [18]; furthermore, there were significant decreases in the AH of 5-year-old buphthalmic rabbits [19] but not in the 2-year-old group, demonstrating that alpha-2-HS-glycoprotein alters with pathologic changes in DM, anterior lens capsule, and the angular meshwork. In humans, it was shown to be an inhibitor of transforming growth factor (TGF)-β2 [20], a protein that shows increased expression in the trabecular meshwork (TM) in open-angle glaucoma causing extracellular matrix (ECM) deposition in the human TM [21]. The alpha-2-HS-glycoprotein inhibits bone morphogenetic proteins that are changed in the TM in open-angle glaucoma [22]. This evidence suggests the potential interactions of the alpha-2-HS-glycoprotein with multiple proteins that are important in open-angle glaucoma. However, there is scarce evidence demonstrating a relationship between the alpha-2-HS-glycoprotein and cataract disease in human beings to date. Interestingly, the serum levels of alpha-2-HS-glycoprotein, called fetuin-A, are known to be highly associated with DM in humans. Initially, it was discussed in the context of preventing glucose toxicity in early 2002 [23,24]. Then, in the past two decades, the alpha-2-HS-glycoprotein was linked to insulin resistance, obesity, and cardiovascular diseases [25–31]. Guo et al. and Roshanzamir et al. revealed evidence using meta-analyses that higher serum alpha-2-HS-glycoprotein levels are associated with increased risk of type 2 DM [32,33]. All these previous studies reported the correlation of alpha-2-HS-glycoprotein levels in urine [34] or serum [35] with diabetes. Yuksel et al. performed a serum and AH alpha-2-HS-glycoprotein (fetuin-A) level comparison in pseudoexfoliation syndrome (PEXS) patients [36]. They found significantly increased alpha-2-HS-glycoprotein levels in the AH of patients with PEXS, but no correlation between the AH and serum levels of alpha-2-HS-glycoprotein between the groups. They suggested that the increase in alpha-2-HS-glycoprotein levels in the AH was due to disruption of the blood–aqueous barrier because of the hypoperfusion and anterior chamber hypoxia in PEXS. Thus, until now there was scarce evidence to prove that the serum level of alpha-2-HS-glycoprotein was associated with that in AH. However, our results are the first to report that human aqueous levels of the alpha-2-HS-glycoprotein are associated with diabetes risk factors for cataract formation. The ELISA confirmation of aqueous alpha-2-HS-glycoprotein levels confirmed these results. In certain diabetic patients, we provide a novel way of thinking about changes in alpha-2-HS-glycoprotein levels in the circulation and in the aqueous fluid. We sugges<sup>t</sup> that the alpha-2-HS-glycoprotein could be an aqueous-specific marker of cataract risk, which is highly associated with diabetes. The alpha-2-HS-glycoprotein is known as an immune-reactive protein that was determined to be smoking- and age-associated

with the development of head and neck cancers. The consistent association of chronic smoking shows an immune reactivity status that changes the serum levels of alpha-2-HSglycoprotein in head and neck cancer patients [37]. Marechal et al. demonstrated a negative correlation between serum fetuin-A levels and a history of smoking, in which fetuin-A levels were determined by a common haplotype of the *AHSG* gene, low plasma cholesterol, and a history of smoking in renal transplant recipients [38]. They considered that it might reflect consequences of tobacco smoking on liver function, physical activity, or weight loss, which increased aortic calcification and risk of cardiovascular events in renal transplant recipients. These previous studies support our result that the alpha-2-HS-glycoprotein may be associated with the smoking habit. We considered that the alpha-2-HS-glycoprotein could be an aqueous-specific marker of cataract risks that is highly associated with smoking. However, multiple limitations of this study should be reported. First, only eight to ten samples in each group were investigated, and future large-scale studies could help confirm our results. The small sample numbers may be attributed to ELISA, which could not validate the proportional results of aqueous alpha-2-HS-glycoprotein levels in the three groups. Second, only a small amount of AH could be obtained because of anatomical features, which limited our ability to conduct subsequent validation assays. Third, the development of multiplex immunoassays can be improved. Finally, we can only provide the results of proteomic and ELISA data correlated with smoking and DM risk factors. The exact pathway by which the alpha-2-HS-glycoprotein is involved in cataract pathogenesis remains unclear. More future investigations of molecular pathways are required to discuss how and why the proteomics data varied with smoking and DM, and finally to supply better knowledge of cataracts for the whole of humanity. More studies are also required to analyze the alpha-2-HS-glycoprotein levels in AH of non-diabetic cataract patients, along with further serum and AH comparison analyses of cataract patients with diabetes. In conclusion, our results are from a pioneering exploration of the protein profile for the risk factors involved in cataracts. Cataracts form because of a complicated pathological process involving several proteins that participate in immune reactions and metabolic processes that were identified in AH using a proteomics analysis. The alpha-2-HS-glycoprotein, called fetuin-a, could be a potential aqueous biomarker associated with DM and smoking, which are cataract risk factors. Additional studies are required to complete the analysis and to understand the functions of these cataract-specific proteins, which could provide significant information for the diagnosis, clinical treatment, and prognosis of cataracts.

**Author Contributions:** Conceptualization, W.-C.C. and C.-W.C.; methodology, C.-C.L. and C.-W.C.; software, C.-C.L. and C.-H.L.; validation, W.-C.C., C.-W.C. and S.-H.C.; formal analysis, C.-H.L.; investigation, W.-C.C.; writing—original draft preparation, W.-C.C.; writing—review and editing, C.-W.C.; visualization, C.-C.L.; supervision, C.-W.C.; funding acquisition, W.-C.C. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Institutional Review Board Statement:** The study was conducted according to the guidelines of the Declaration of Helsinki, and approved by the Institutional Review Board of Taoyuan General Hospital, Ministry of Health and Welfare (TYGH109009 and 30 April 2020) (Taoyuan, Taiwan).

**Informed Consent Statement:** Informed consent was obtained from all subjects involved in the study. Written informed consent has been obtained from the patients to publish this paper.

**Data Availability Statement:** Not applicable.

**Conflicts of Interest:** The authors declare no conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results.
