Glaucoma is one of the most frequent causes of irreversible blindness worldwide [1
], and the number of glaucoma patients will double by 2040 as a result of the growing elderly population [2
]. Glaucoma is defined as a progressive loss of the innermost nerve cells of the retina, the retinal ganglion cells, with a simultaneous characteristic loss of the peripheral visual field [1
]. A major risk factor for the development of glaucoma is elevated intraocular pressure (IOP), and the sole treatment strategies currently available are IOP-lowering medical or surgical treatments [4
]. Although IOP-lowering treatment strategies often slow the rate of glaucoma progression, far too many go blind despite well-treated IOPs [4
]. In this context, a Swedish study has shown that 42% of diagnosed glaucoma patients lose sight in one eye, while 15% end up blind [4
The most common form of glaucoma in the western world is primary open-angle glaucoma (POAG). POAG can be subdivided into two clinical phenotypes depending on the IOP. Up to 50% of POAG patients have glaucomatous neurodegeneration despite an IOP within the normal range, denoted normal-tension glaucoma (NTG), while the rest have increased IOP, denoted high-tension glaucoma (HTG) [8
]. In addition to these clinical subgroups within POAG, there is a group of patients with elevated IOP, but no signs of glaucomatous neurodegeneration, patients with ocular hypertension (OHT). The apparent resistance towards increased IOP in OHT patients formed the basis of the present study.
It is recognized that glaucoma is a multifactorial condition with a number of competing risk factors [1
]. The susceptibility towards the different risk factors is most probably different between patients. Thus, IOP may be a major risk factor in some patients, whereas other risk factors may be more significant in other patients. A widely suggested IOP-independent risk factor is dysfunctional retinal autoregulation [11
]. Such dysfunctional retinal autoregulation will cause a fluctuating oxygen supply to the retina, thereby increasing the level of reactive oxygen species (ROS) [13
]. In accordance with this, elevated levels of oxidative stress have been observed in POAG patients [10
In the present study, the levels of oxidative stress, as well as antioxidants, were investigated in patients with glaucomatous neurodegeneration and compared to patients with OHT. Patients with NTG were selected based on a hypothetical assumption that these patients are more vulnerable towards other risk factors than IOP, compared to HTG patients, and thus, the most distant phenotype from OHT patients with high IOP but no signs of glaucomatous neurodegeneration. In addition to examining baseline levels of oxidative stress molecules and antioxidants, the response to altered oxygen supply in the test groups was investigated. We imagine that changes in oxidative stress may be less in OHT patients compared to NTG patients during a fluctuating oxygen supply. For this purpose, all tested people were exposed to two hours of hypoxia, followed by 30 min of reperfusion in a full monitor setup. Blood samples were taken before, during, and after hypoxia, and systemic levels of oxidative stress markers, as well as antioxidants, were measured by means of multiple assays. Since polyunsaturated fatty acids (PUFAs), including the
-3 fatty acid docosahexaenoic acid (DHA) and the
-6 fatty acid arachidonic acid, can act as antioxidants [20
], we further performed liquid chromatography with tandem mass spectrometry (LC-MS/MS)-based lipidomic analysis to investigate levels of lipid peroxidation and pro-homeostatic lipid mediators in plasma at baseline and in response to a fluctuating oxygen supply.
In summary, by utilizing a human experimental model, the oxidative–antioxidative balance was assessed in response to a fluctuating oxygen supply in patients with glaucomatous neurodegeneration compared to patients with OHT, a condition with an apparent resistance towards glaucoma despite high IOP. The presented results suggest that patients with NTG might suffer from oxidative stress due to an ineffective antioxidant defense compared to OHT patients. In conclusion, our results indicate that OHT patients’ apparent resistance to glaucomatous neurodegeneration may be due to a higher level of antioxidant capacity.
A link between oxidative stress and glaucomatous neurodegeneration has been suggested [10
]. In this study, we show increased baseline levels of TAC in patients with OHT. Moreover, we report an increased abundance of oxidation products of DHA and derivatives of arachidonic acid, i.e., lipid mediators with antioxidative functions in OHT patients. In our human experimental model, we find significant regulations of pro-homeostatic lipid mediators in response to a fluctuating oxygen supply in patients with NTG and controls, whereas there was no significant regulation in patients with OHT. Although increased levels of oxidative stress were not detected in any of the test groups, neither at baseline, nor in response to a fluctuating oxygen supply, our results suggest a relationship between increased antioxidant capacity and resistance towards glaucomatous neurodegeneration in patients with OHT. To our knowledge, this is the first study evaluating oxidative stress and antioxidative mediators in patients with NTG and OHT during oxygen stress.
Decreased TAC in serum from glaucoma patients compared with control subjects [15
] and cataract patients [42
] has been reported. Such findings indicate that glaucoma patients have a reduced capacity to cope with increasing oxidative stress during aging [49
]. One previous study has found contradicting results with increased levels of TAC in patients with NTG compared to controls [44
]. We could not confirm any differences in TAC between patients with NTG and controls. Our present finding of increased TAC in plasma from patients with OHT compared to both patients with NTG and controls indicates a critical role of antioxidant capacity in preventing glaucomatous neurodegeneration. In support of our results, a study by Lascaratos et al. has shown enhanced systemic mitochondrial efficiency in patients with OHT compared to patients with NTG and controls in isolated lymphocytes from peripheral blood samples [1
To explore specific antioxidants, we examined the activity of the extracellular antioxidant, SOD3. Our results did not identify any significant differences in SOD3 between the tested groups. Generally, there are discrepancies between studies on SOD levels in glaucoma patients compared to controls. Whereas some studies have reported decreased SOD in serum from glaucoma patients compared with controls [17
], other studies have found increased SOD in glaucoma patients compared to cataract patients in aqueous humor [45
] and red blood cells (RBCs) [3
]. To our knowledge, only one study has explored SOD levels in patients with OHT. In this study, patients with NTG had elevated levels of SOD in isolated lymphocytes compared to patients with OHT [1
]. The contradicting results may be explained by the different SOD isoenzymes of blood cells, aqueous humor, serum, and plasma. Extracellular SOD3 is more relevant in serum and plasma, while the cytosolic SOD1 and mitochondrial SOD2 are more relevant in blood cells. The expression and activity of SOD isoenzymes are known to be highly tissue- and compartment-specific, which should be taken into account when comparing observations. Based on the studies described above, it seems that changes during glaucomatous neurodegeneration are similar in aqueous humor and RBCs but different in serum. The lack of a significant increase in SOD3 of patients with NTG or an elevation during hypoxic exposure in the included groups may though also indicate an exhaustion of the antioxidant defense in plasma. It leaves us with an unanswered question about the involvement of SOD3 in the vulnerability towards glaucomatous neurodegeneration and in the antioxidant defense of OHT.
To further investigate the oxidative–antioxidant balance, we investigated lipid peroxidation. Previous studies have reported increased MDA in serum [15
] and RBCs [3
] in patients with glaucoma. Increased MDA in the aqueous humor has also been observed in glaucoma patients compared to patients with cataract [48
]. Results from the current study did not identify significant differences in free radicals or MDA levels between patients with NTG, patients with OHT, and controls. Overall, no direct evidence for increased levels of oxidative stress between the test groups was found. Thus, we could not confirm any of these previous findings regarding the glaucoma patient group. Even though we did not detect higher levels of free radicals or lipid peroxidation in either of the analyzed groups, further studies are needed to rule out these processes in the pathology of NTG and OHT. The increased levels of TAC in OHT patients continue to indicate that some processes of oxidative stress are actively counteracted by antioxidant defenses, and thereby also could be involved in the clinical phenotypes of OHT and NTG. Therefore, it remains to be elucidated whether levels of other enzymes or molecules of oxidative stress-related processes are altered in patients with OHT and patients with NTG in comparison to controls.
To examine the high antioxidant defense observed in patients with OHT, we chose to investigate pro-homeostatic lipid mediators derived from PUFAs [20
]. The biosynthesis of docosanoids from DHA leads to the formation of stable HDHAs, three of which were measured in the current study. Anti-inflammatory properties of all three have been described [23
]. Elevated levels of two of the measured HDHAs were identified in patients with OHT compared to patients with NTG and/or controls, indicating that these pro-homeostatic lipid mediators may contribute to higher antioxidant capacity levels in OHT patients. Arachidonic acid can yield HETEs, of which two were analyzed. Some discrepancies regarding the oxidant or antioxidant effect of these derivatives are found in the literature. A former study has identified neuroprotective effects of both HETEs [59
]. Some contradicting studies have suggested that 12-HETE cause mitochondrial dysfunction [60
] and that 15-HETE can induce ROS production [61
]. Increased levels of both HETEs were identified in the OHT group compared to controls and/or patients with NTG. In line with current results, we believe that these increased levels indicate components of the antioxidant defense in OHT patients. High levels of pro-homeostatic lipid mediators add to the understanding of how patients with OHT might withstand elevated IOP and glaucomatous neurodegeneration.
In the present study, we utilized our human experimental model in which strictly characterized clinical phenotypes, with and without glaucoma, underwent hypoxia followed by a reperfusion period. All groups regulated systemic vital parameters, HR, SAT, pO2, and pCO2, significantly during the hypoxia model, confirming that the model allowed us to compare both baseline levels and stress responses of the included groups. Levels of lipid mediators, 14-HDHA, 20-HDHA, and 15-HETE, were varying during exposure to fluctuating oxygen supplies in controls and/or patients with NTG, whereas levels remained stable in OHT patients. These results indicate that the naturally occurring high antioxidant defense present in patients with OHT could be enough to tolerate oxygen stress without regulating levels of lipid mediators.
Apart from pro-homeostatic lipid mediators, no other analyzed markers varied significantly with fluctuating oxygen levels. These results might indicate that alterations in the level of enzymes, lipids, and smaller molecules related to oxidative and antioxidative mechanisms in plasma are highly heterogeneous. A measurable oxidative stress response in plasma induced by hypoxia depends on both the duration and intensity of hypoxia [62
]. It is possible that some of these molecular changes occur more rapidly or might be a more long-term effect. In our previous study with the same human experimental model we showed changes in plasma levels of amino acids and lactate in response to hypoxia [32
]. Another study measured increased markers associated with oxidative stress during normobaric hypoxia with 12.9% oxygen for nine hours [63
]. Additional studies have evaluated the effect of hypobaric hypoxia for either four hours [64
], 72 h [65
], three days [66
], or 13 days [67
]. Three of these studies investigated oxidative stress and antioxidants in plasma and found a hypoxia-induced increase in oxidative stress [64
] or TAC [64
]. Two hours of normobaric hypoxic exposure with 10% oxygen was chosen as the upper limit of acute hypoxia concerning ethics and the comfort of participants. More studies are needed to confirm whether the analyzed oxidative and antioxidant markers are regulated during fluctuating oxygen supplies.
To increase the chance of identifying changes in systemic markers between the analyzed groups, we have carefully considered which clinical phenotypes we should include. Since our goal was to investigate risk factors other than IOP that may play a role in the development of glaucoma, we have chosen to include NTG patients who may be more vulnerable to risk factors other than elevated IOP. To investigate whether some factors, on the other hand, prevent vulnerability to elevated IOP, we have selected the clinical phenotype OHT, characterized by having elevated IOP with no evidence of glaucomatous neurodegeneration. Based on our results, it will be interesting in future studies to investigate whether HTG patients (patients with elevated IOP and glaucomatous neurodegeneration) are similar to OHT or NTG patients to further understand why some are more vulnerable to increased IOP than others.
The retina and aqueous humor, both potentially involved in the pathogenesis of glaucoma, are supplied with nutrients from plasma through the blood–retinal barrier and blood–aqueous barrier that control which and the quantity of substances transferred. It may be, that development of or protection against glaucomatous neurodegeneration is not only affected by systemic levels of free radicals and antioxidants but also by the permeability of the barriers which may be affected by pathological conditions. As the current study focused on a highly complex human experimental setup, it was unfortunately not possible to compare plasma levels of free radicals and antioxidants to intraocular levels. Such comparison would indeed be important in future studies, to investigate whether differences and possible pathological explanations for glaucoma relate only to plasma or can also be found in aqueous humor and ocular tissues as a result of transfer across the barriers.