Ascorbate and other antioxidants in human tears : sources, biological variation and effect of collection methods and dietary supplementation

Abstract

Human tears contain water-soluble antioxidants and these may be important as a 'front-line' ocular defense against photo-induced oxidative damage. However, published data on tear antioxidants are scarce and conflicting. There is no clear picture on their amounts or sources, and neither biological variation nor response to dietary antioxidant supplementation has been investigated to date. The aims of this study, therefore, were to: 1. Evaluate a novel method ('FRASC') for total antioxidant capacity (as the Ferric Reducing (Antioxidant) Power (FRAP)) and ascorbate concentration as applied to human tears. 2. Determine the concentrations of the water-soluble antioxidants, ascorbate, urate, cysteine, glutathione and tyrosine, and the FRAP value of human reflex tears, using FRASC method and established HPLC and commercial enzymatic method. 3. Resolve differences in published data on tear antioxidant levels by investigating the effect of different tear collection methods (Schirmer strip and capillary tube). 4. Investigate intra-individual (biological) variation in tear ascorbate. 5. Determine the immediate source of ascorbate in tears. 6. Investigate baseline relationships between the ascorbate concentration and the FRAP values of tears and plasma. 7. Determine the effect of vitamin C supplementation on the ascorbate concentration and the FRAP values of tears and plasma. Linearity, precision and sensitivity results indicated that the FRASC method is suitable for analysis of ascorbate and total antioxidant capacity (as the FRAP value) of human tears, and ascorbate results agreed closely with those obtained using an HPLC method. Total antioxidant capacity, ascorbate and urate concentrations (mean+-SD) in reflex tears (collected by capillary tube) of 47 apparently healthy Chinese subjects were, respectively, 409.2+-162.2 uM, 23.6+-9.6 uM and 68.4+-46.1 uM. Cysteine, glutathione and tyrosine concentrations in reflex tear samples of 12 apparently healthy Chinese subjects were low, 3.7+-2.5 uM, 0.7+-0.2 uM, and 2.7+-1.3 uM, respectively. Tear antioxidants were significantly (P<0.05) and up to ten fold higher in reflex tears collected by the commonly used Schirmer strip method compared with tears collected concurrently by capillary tube from the other eye of the subject (n=12). Spuriously high antioxidant levels in tears collected using Schirmer strips are probably caused by contamination with intracellular constituents. With regard to intra-individual variation, no significant difference was seen for ascorbate in tears collected on two separate days within the same week. With regard to the immediate source of tear antioxidants, results show that ascorbate and urate are present in fresh reflex tears. Our results do not support the view that the cornea is the source of tear ascorbate, as has been suggested previously, but rather that the lacrimal gland is likely to transfer dietary ascorbate from the plasma into the tears. In a single-blinded, cross-over supplementation trial, tear ascorbate at baseline was around 33% that of plasma (n=21), with no significant correlation between plasma and tear ascorbate concentrations. After supplementation with 1 g per day vitamin C for 4 weeks, the ascorbate concentration increased significantly (P<0.001) in both fluids. However, the increase in tear ascorbate was small in absolute terms, averaging 5 uM, compared to an average increase of 38 uM in plasma ascorbate. No significant correlation was seen between the ascorbate response in plasma and in tears, and no post-supplementation change was seen in total antioxidant capacity in either tears or plasma. In summary, results of this study indicate that the FRASC method is suitable for measuring total antioxidant capacity and ascorbate in human tears, and that much of the previously published tear antioxidant data should be re-examined in the light of the clear effect of the Schirmer strip collection method. The capillary tube collection method is suggested as the method of choice for reflex tear collection for biochemical studies. Using this technique, this study has demonstrated that tears contain significant amounts of ascorbate and urate, but very low levels of cysteine, glutathione and tyrosine. This study also indicates that ascorbate is present in fresh lacrimal gland secretion, and that passage of ascorbate into the tears is probably not by simple diffusion. However, lacrimal gland uptake of ascorbate from plasma appears limited, and this may reflect a controlled or saturable process.