Lenticular Oxygen Toxicity

Abstract

purpose. To investigate the possible toxic effect of oxygen on lenses in an organ culture.

methods. Bovine lenses were exposed to four different combinations of ambient pressure and oxygen concentration in an organ culture throughout a 7-day period. Lens transparency, histology, enzymatic activities, and photomicrographs were compared in study and control groups.

results. No differences were observed between study and control lenses in all measured parameters in a group subjected to a single exposure of 100% oxygen under increased (i.e., hyperbaric) ambient conditions and a group exposed repeatedly to high ambient pressure and normal oxygen partial pressure. Decreased lenticular transparency and enzymatic activities along with structural changes were observed in lenses exposed repeatedly to 100% oxygen concentration under both normal and increased ambient pressures. The observed changes were oxygen-load–dependent: the higher the oxygen partial pressure and the longer the time of exposure, the more severe the changes observed. Optical and structural changes in the lens occurred in a centripetal orientation: the greater the oxygen load, the more central the damage.

conclusions. High oxygen load has a toxic effect on bovine lenses in organ culture. These effects appear to be cumulative: the higher the oxygen partial pressure and the greater the number of exposures, the more severe the changes observed in the lenses. Changes marking toxicity follow the route of oxygen diffusion into the lens, from the periphery to the center. Cautious interpretation of the results may indicate a role of oxygen (and/or its derivatives) in human cataract formation.

The paradox of aerobic life (the “oxygen paradox”) is that higher eukaryotic organisms cannot survive without oxygen, yet oxygen is partially toxic. This dark side of oxygen relates directly to the fact that atomic oxygen is a free radical and molecular oxygen is a (free) biradical. ¹ Although concerted tetravalent reduction of oxygen by the mitochondrial electron transport chain is a relatively safe process, the univalent reduction of oxygen generates reactive intermediates. The superoxide anion radical hydrogen peroxide and the extremely reactive hydroxyl radical, which are common products of life in an aerobic environment, appear to be responsible for oxygen toxicity. To survive in such an unfriendly oxygen environment, all known aerobic organisms synthesize a series of antioxidant enzymes that intercept and inactivate reactive oxygen intermediates. Antioxidant enzymes and compounds are not completely effective in preventing oxidative damage. Human cells contain a series of damage removal and repair enzymes that deal with the oxidative damage. Oxidative stress levels may vary from time to time. Organisms can adapt to such fluctuating stresses by upregulating the synthesis of antioxidant enzymes and damage removal and repair enzymes. ² Despite the antioxidant and repair mechanisms just described, oxidative damage remains an inescapable outcome of aerobic existence. In recent years, oxidative stress has been implicated in a wide variety of degenerative processes, diseases, and syndromes, as well as in a wide variety of age-related disorders, possibly including factors underlying the aging process itself. ^{3 4} Some of these oxidation-linked disorders can be initiated or exacerbated by numerous environmental pro-oxidants and/or pro-oxidant drugs and foods. 

Cataracts are a major cause of loss of lens transparency in the aging population. Oxidative stress has been proposed to play a role in senile cataract formation. ^{5 6 7} It is known that molecular oxygen (O₂) can oxidize −SH groups of proteins, resulting in intramolecular disulfide bonds or intermolecular aggregates. Although it has been argued that little if any O₂ ever reaches the innermost regions of the lens, ⁸ an O₂ tension of 20 mm Hg has been measured in the anterior cortex of the lens, ⁹ and the tension of O₂ at the surface of the posterior lens has been reported to be 15 to 20 mm Hg. ¹⁰  

Hyperbaric oxygen (HBO) therapy is indicated in the treatment of several human medical disorders, including decompression sickness, carbon monoxide poisoning, air embolism, anaerobic infections, and ischemic vascular and diabetic lesions. ¹¹ During the course of multiple HBO treatments, some patients report visual difficulties resulting from lenticular myopia. ¹¹ The precise reasons for this are not well understood. In some individuals, refractive error may not completely revert to its pretreatment level. HBO treatment has also been reported to induce lens opacification in human subjects and in animals. In a study in which patients were treated with HBO for 150 to 850 sessions, 7 of the 15 patients with initially clear lens nuclei showed development of a nuclear cataract that decreased visual acuity. In seven of the remaining eight subjects, increased nuclear light-scattering was observed. ¹² In a study in which mice were treated with HBO, one half of the surviving animals had nuclear cataracts within 8 months. ¹³  

Studies with HBO in tissue culture can be used to mimic the in vivo situation and elucidate mechanisms involved in the HBO-induced loss of lens transparency. Previous reports on treatment of cultured lenses with HBO demonstrated significant oxidation of the reduced form of glutathione, ¹⁴ formation of oxidized proteins, and modification of certain susceptible enzymes, with effects being initiated primarily in the nuclear region of the lens. ^{15 16 17} These observations may support oxidative stress as a possible factor in senile cataract formation. Because hyperbaric conditions induce very high oxygen partial pressure, exposing lenses in tissue culture to HBO permits the study of lenticular oxidative damage. The results of such an investigation may hint at possible mechanisms of human senile cataract formation. 

The purpose of the present study was to investigate the character, magnitude, and relevant variables influencing the optical damage caused by hyperbaric and normobaric oxygen (NBO) to bovine lenses in organ culture conditions. 

Methods

Experimental Groups

Altogether, 260 bovine lenses were included in the present study. The lenses were divided into four experimental groups: (1) HBO single-exposure group: 25 lenses treated by a single exposure to HBO for 120 minutes of 100% oxygen in a pressure chamber at 2.5 atmospheres absolute (ATA); (2) HBO repeated-exposure group: 35 lenses treated daily by HBO for 4 days, each with 120 minutes of 100% oxygen in a pressure chamber at 2.5 ATA; (3) NBO group: 35 lenses treated daily for 4 days by 100% oxygen at 1 ATA for 120 minutes; and (4) hyperbaric normoxic oxygen (HNO) group: 35 lenses treated daily for 4 days by HNO (160 mm Hg oxygen pressure achieved through 8.4% oxygen concentration at 2.5 ATA) for 120 minutes. The conditions of 2.5 ATA 100% oxygen for 2 hours mimicked the conditions applied to human patients therapeutically. 

Lenses were incubated at 35°C. During the pressure exposure study, lenses were removed from the incubator for approximately 155 minutes and placed in the pressure chamber. The pressure chamber was kept at room temperature of 30°C, with temperature variations not exceeding 2°C. The control group for each experiment consisted of 130 lenses, extracted from the fellow eyes of the study animals. During exposure of the study lenses to the experimental conditions described, control group lenses were exposed to normal room air (1 ATA, 160 mm Hg oxygen pressure) for approximately 155 minutes. All lenses were returned to 35°C incubation at the end of exposure. Lens optical quality was assessed throughout the 7 days of culture. Catalase analysis was performed on day 7. 

Organ Culture System

Lenses were carefully excised from eyes obtained from 1-year-old male calves, 2 to 4 hours after enucleation. Eyes were enucleated after the calves had been killed in an abattoir. One lens of each calf was used for experimental oxygen exposure, and the other served as the control. Each lens was placed in a specially designed culture chamber consisting of two compartments connected by a round hole with a diameter that was 1 mm smaller than the diameter of the lens. The lens was located between the two compartments, leaving a clear space filled with culture medium both below and above the lens. Both lens surfaces were bathed in 24 mL of culture medium consisting of M199 with Earl’s balanced salt solution, 8% fetal calf serum, and antibiotics (penicillin 100 U/mL and streptomycin 0.1 mg/mL). The medium was changed daily. The lenses were incubated at 35°C after they were placed in organ culture for preincubation. Twenty-four hours later, lens optical measurements were taken, and only the lenses with good optical quality were used for the experiment. 

Hyperbaric Chamber

The specially designed lens culture chambers were exposed to different gas mixtures in a sealed pressure chamber. In the group exposed to hyperbaric conditions, pressure was raised over a 20-minute period to 2.5 ATA. Each exposure was 120 minutes long. In the groups exposed to normobaric conditions, pressure in the chamber was kept at 1 ATA. During the exposures, the pressure chamber was kept at room temperature (30°C) with temperature variations not exceeding 2°C. Oxygen saturation inside the pressure chamber was monitored and held constant throughout the exposure session. At the end of the hyperbaric exposures, the pressure was lowered to ambient pressure over a 15-minute period. 

Optical Quality Monitoring

Lens optical quality was monitored daily throughout the culture period. Lens optical measurements were determined by an automated scanning laser system that recorded both relative transmission and focal length across the lens. The laser scanner consisted of a low-powered helium-neon laser mounted on a computer-driven x−y table with two video cameras and a video frame digitizer. The laser was programmed to scan across the lens in the axial direction in steps of 0.5 mm, while the video cameras transmitted the image of the refracted beam to the video digitizer. A custom software program determined the focal length and relative intensity of each refracted beam from the digitizer image. The optical center was first determined for each lens by finding the position of minimum refraction for both the x and y directions, and then the program determined the focal lengths for 11 beam positions at equal step sizes on each side of the center. The system measured relative changes in transmission of the lens by measuring the excitation state of a 512 × 512-pixel television screen matrix for each exiting (refracted) laser beam. However, scattered measurements attempted in the past had been difficult to interpret in comparison with focal measurements, ¹⁸ and therefore this study concentrated on focal length results. Focal length variability (FLV) represented the variation in the focal lengths of the 22 beams passing through the lens during each scan and was calculated as the standard error of the mean of the 22 focal lengths. 

The beam examined for FLV included 22 rays, from peripheral to central. In an attempt to look into a possibly different effect of experimental conditions on peripheral versus central rays, we examined the changes in each of these groups of rays separately. The central rays were the three laser rays that passed through the middle of the incubated lens. The peripheral rays were the six laser rays that passed most peripherally (three ray on each side of the lens; Fig. 1 ). 

Preparation of Lens Samples for Enzyme Analysis

The lens epithelium was dissected under a binocular stereomicroscope. Cuts were made along the equators, and the lens capsule and adherent epithelium were removed from the entire lens. Care was taken that no fiber tissue remained attached to the epithelium. The tissue was immersed immediately in a 200-μL volume of 50 mM phosphate buffer (pH 7.0). All further work was performed at 0°C to 4°C. The tissue was sonicated in a ultrasonic disintegrator (MSE 150; Misonix Inc., Farmingdale, NY) at 50 W for 10 seconds. This procedure was performed twice and was followed by centrifugation at 14,000g for 10 minutes. Enzyme activities of the supernatant were measured. Catalase activity was measured according to the method of Beers and Sizer, ¹⁹ by spectrophotometric recording of the cleavage of H₂O₂ at 240 nm. The reaction mixture contained 0.023 M H₂O₂ in 0.05 M phosphate buffer (pH 7.0). One unit of enzyme activity was defined as 1 μM of H₂O₂ cleaved per minute at 37°C. 

As H₂O₂ was added externally, the amount was far in excess of that of glutathione. For this reason glutathione peroxide activity (for which glutathione is an obligatory substrate) was assumed to be negligible compared to catalase (whose only substrate is H₂O₂). 

Histochemical analysis of magnesium-activated ATPase was performed according to Sheeham and Hrapchak ²⁰ on five lenses of the HBO exposure group and their control fellow eye lenses. Quantitative measurements were made with an image-analysis system (analySIS docu 3.0; Soft Imaging System, Münster, Germany). 

Lenticular Damage Photography

Photographs of five lenses from each group and their fellow eye lenses were taken daily using an inverted microscope (Axiovert 135; Carl Zeiss Meditech, Jena, Germany). Photomicrographs of the lenses’ periphery and their centers were taken at ×20 and ×40. 

Histology Preparations

Five lenses from each experimental group and their fellow eye lenses (a total of 40 lenses) were fixed in 10% formalin and embedded in paraffin. Paraffin sections (5-μm-thick) were subjected to hematoxylin-eosin (H&E) staining and histochemical localization of adenosine triphosphatase (ATPase) activity. ²⁰  

Statistical Analysis

All results were analyzed using Student’s paired t-test. A change was defined as significant if the difference between control and treated groups reached P < 0.05. 

Results

On gross examination, all study lenses appeared transparent to the unaided human eye. Optical quality during the 168 hours of culture is shown in Figures 2 3 4 5 . FLV represented the variation in the focal lengths of the 22 beams that passed through the lens during each scan and was calculated as the standard error of the mean of the 22 focal lengths. There was almost no change in FLV in the control lenses during the culture period. The lenses exposed to a single HBO application did not show an increase in FLV throughout the 168 hours of the incubation period (Fig. 2) . 

The lenses exposed repeatedly to HBO showed an increase in FLV after 144 hours in culture (Fig. 3A) . FLV, which represented the loss of sharpness of focus, increased up to twofold after 144 hours of culture and up to threefold after 168 hours. These changes did not return to control values during the experimental period. 

The mean FLV of the central rays in the HBO-exposed lenses is shown in Figure 3B . There was a 2-fold increase in FLV after 144 hours and a 2.5-fold increase after 168 hours. The mean FLV of the peripheral rays that passed through the HBO-exposed lenses is shown in Figure 3C . There was a fourfold increase in FLV after 96 hours of incubation, which remained unchanged until the end of the experimental period. 

The lenses exposed to NBO showed an increase in FLV after 120 hours in culture (Fig. 4A) . There was a maximal 1.5-fold increase in FLV that occurred after 168 hours in culture. The mean FLV of the central rays in the NBO-exposed lenses is shown in Figure 4B . There was almost no change in FLV in the central rays. The mean FLV of the peripheral rays that passed through the NBO-exposed lenses is shown in Figure 4C . There was a fourfold increase in FLV after 120 hours of incubation, which remained unchanged until the end of the experimental period. 

The lenses exposed to HNO showed an increase in FLV after 48 hours in culture (Fig. 5A) . These changes returned to control levels 48 hours later. The FLV was maintained at control levels until the end of the experimental period. The FLV of the central rays that passed through the hyperbaric air-exposed lenses is shown in Figure 5B . There was almost no change in FLV in these central rays. The FLV of the peripheral rays that passed through the hyperbaric air exposed lenses is shown in Figure 5C . There was almost no change of the FLV in these peripheral rays. 

Catalase activity was evaluated on the seventh day. There was a reduction in enzyme activities in the lenses exposed to HBO repeatedly. There was no significant reduction in enzyme activities in all other treatment groups in comparison with the control group (Fig. 6) . 

Histochemistry

ATPase activity at the center of the lens on the seventh day of culture is shown for the control lens in Figure 7A and for the fellow eye’s lens exposed to HBO in Figure 7B . Diminished ATPase activity in the center of the lens exposed to HBO was demonstrated. In standard microscopic fields the control lenses had a mean of 521,008 μm² (18%) ATPase activity in comparison with a mean of 194,639 μm² (5.6%) ATPase activity in HBO-exposed lenses. The difference in ATPase activity between the study and control groups was statistically significant (P < 0.01). 

Photographs

Photomicrographs of lenticular damage were recorded throughout the incubation period. The photographs showed clear undamaged lenses in the control eyes (Fig. 8A) . Lenses that were exposed to NBO showed peripheral subepithelial and cortical bubbles, with a clear central zone (Fig. 8B) . Lenses that were repeatedly exposed to HBO showed peripheral and central subepithelial and cortical bubbles (Figs. 9A 9B) . Lenses that were exposed to HNO showed subepithelial bubbles that appeared on day 3, but vanished on day 5. These bubbles were located in the intermediate zone of the lenses, and the central and peripheral zones remained relatively clear (Figs. 10A 10B) . The bubbles are most prominent along the suture lines. This might indicate that the sutured area is more vulnerable to oxidative damage. 

Histology

Histologic examination showed that all the lenses in the control groups maintained their regular dense fibrillar pattern, and their epithelium remained intact after 7 days in culture (Fig. 11) . All lenses exposed to HBO demonstrated a loose fibrillar pattern resembling shutter slits, in the lenticular center and peripheral zones (Figs. 12A 12B) . All lenses exposed to NBO maintained a regular dense fibrillar pattern in the central zones (Fig. 13A) , but showed the “shutter slit” appearance in their periphery after 7 days in culture (Fig. 13B) . 

Discussion

In the present study, we investigated the damage caused to bovine lenses in tissue culture by externally applied oxygen under different ambient conditions. Measurements of lenticular damage were conducted through a specially designed, highly sensitive system. This system has been described and used to determine ultraviolet radiation–induced lenticular damage in organ culture. ¹⁸ In this system, damage is measured through the change in focal distance of 22 laser rays passing through the lens at predetermined points. The two main advantages of this system are its high accuracy (and reproducibility) and sensitivity to relatively small changes—the system can detect change in focal distance of the damaged lens long before any opacity can be seen on gross observation of the lens. In fact, in all lenses reported to have sustained optical damage, as detected by the increase in focal lens variability, in the present report, we were unable to see any change in lens clarity by careful unaided visual examination. The accuracy and sensitivity of the measuring system enabled us to detect relatively small changes of lens optical quality in comparison with the ability to detect such changes by simple observation of the lenses. 

Under physiological conditions, oxygen reaches the lens through diffusion from both the aqueous humor and the vitreous. Oxygen reaches the aqueous humor from atmospheric air through the cornea or from the ciliary processes and then reaches the vitreous through the retinal and choroidal circulation. Under standard conditions, oxygen concentration in air is 21%, and breathing air pressure is 1 ATA. Oxygen perfusion distance is directly related to its partial pressure: the higher the oxygen partial pressure, the greater its diffusion distance. Tenfold elevation of oxygen partial pressure increases its diffusion distance threefold. ²¹ The partial pressure of a gas in a gas mixture is the gas concentration multiplied by the total pressure of the mixture (Boyle’s law). Obviously, one can change the partial pressure of a gas in a gas mixture either by changing its concentration in the mixture or by altering the total pressure of the mixture, or both. Increasing the partial pressure of oxygen from normal atmospheric conditions in the medium in which lenses are kept in tissue culture increases oxygen diffusion distance at a rate proportional to the increase in the partial pressure. The increase allows for higher quantities of oxygen to penetrate deeper into the lens and to achieve higher oxygen content compared with the amount and depth under normal atmospheric conditions. 

In the present study we increased the partial pressure of oxygen in the study groups by increasing either oxygen concentration (up to 100%) or gas mixture total pressure (up to 2.5 ATA), or both, which caused the intact lenses to face a higher oxygen load. Reports have shown that oxygen partial pressure in blood increases fivefold during 100% NBO inhalation ²² and 12.5-fold in blood and aqueous humor during hyperbaric (2.5 ATA) 100% oxygen breathing. ²³ In our present investigational setup, the difference between study and control lenses was the exposure to high oxygen partial pressure in the study group. In the present study, the matched eye from the same animal served as the control to the study eye in each treatment group. For this reason, the observed changes in lenticular clarity between the study and control groups can be attributed to the effects of change in partial pressure of oxygen on the lenses under investigation. 

Oxidative damage was not observed in lenses exposed to a single 100% HBO session. Damage was clearly demonstrated in lenses that were repeatedly exposed to 100% HBO. This observation suggests that there is a threshold of oxygen exposure that must be crossed to create demonstrable damage to lenses. Because each exposure was of similar oxygen partial pressure and length of time, one can conclude that the effect of exposure to oxygen is a cumulative process. This conclusion is supported by the fact that the oxidative damage became more severe as the lens experienced repeated exposure to oxygen (Figs. 3A 4A) . 

When faced with repeated oxidative load under both hyperbaric and normobaric conditions, damage to lenses was demonstrated on days 5 and 6 of incubation (Figs. 3A 4A) . When all rays crossing the lens were observed, exposure to HBO seemed to cause lenticular damage, as demonstrated in the distribution of the lenses’ focal length (Fig. 3A) . When the central and peripheral rays were studied separately, it appeared that all the damage in the normobaric group was caused by the effect on peripheral rays (Fig. 4C) , whereas the central rays were unimpaired (Fig. 4B) . In the hyperbaric exposure group (Fig. 3C) , peripheral rays provided an effect similar to those of the normobaric group (Fig. 4C) , but unlike in the normobaric group, central rays were also damaged (Fig. 3B) . The damage to central rays in the hyperbaric group can account, at least in part, for the more severe damage observed in the summation of all rays in the hyperbaric group compared with the normobaric one (Fig. 3A 4A) . 

The above observations are in agreement with the diffusion distance of oxygen, in which under hyperbaric conditions oxygen partial pressure is much higher, enabling it to diffuse a longer distance and reach the inner parts of the lens. By reaching the center of the lens, oxygen can cause oxidative damage demonstrable for central rays. Under normobaric conditions, the lower partial pressure of oxygen makes its diffusion distance smaller, not allowing high oxygen concentration in the center of the lens and leaving the central rays unimpaired. This observation is of special interest because, to the best of our knowledge, no previous distinction between peripheral and central lenticular damage has been made. Furthermore, previous reports on both clinical and experimental cataracts induced by hyperbaric conditions were based on the presence of nuclear (i.e., central) opacities. ^{12 13} It appears from our results that these previously reported changes in lens transparency could have been the result of relatively advanced oxidative damage (less advanced damage would affect the periphery and be clinically undetectable). We speculate that the relative insensitivity of the methods used to detect lens opacification could not allow for earlier detection of lesser damage. This is probably the reason that damage was not observed after four treatment sessions but that as many as 19 HBO sessions were needed to detect lenticular damage, ²³ whereas we were able to detect it after only four HBO treatments. Another possible explanation for more treatments needed before observation of damage may be related to the use of a live guinea pig model, which may have viable antioxidant systems lacking in the isolated bovine lens of our model. 

We investigated the independent role of hyperbaric conditions on lens damage by placing lenses under HNO with oxygen levels kept at standard “normal” partial pressure. The optical quality of lenses with all rays, as well as with the peripheral and central rays alone, showed no damage of the sort demonstrated for the oxygen-exposed lenses. However, an interesting finding was observed during the early phase of the study: after 72 hours of exposure to HNO lenses showed mild damage that disappeared later. This damage was not demonstrated on the peripheral and central rays that were examined separately, but only for all rays considered jointly. This was probably due to the small magnitude of damage that needed the summation of many rays to become apparent. This may also imply that the cortical region of the lens is more sensitive to HBO. These results suggest that the hyperbaric conditions themselves are not the reason for lenticular damage, but that high oxygen partial pressure is essential for the observed damage to take place. 

During days 2 and 3 in the repeated-exposure groups, the following was observed: (1) The study lenses were worse than the control lenses in the HNO group. The difference was statistically significant (Fig. 5A) . (2) No difference was observed between study and control lenses in the NBO group (Fig. 4A) . (3) The study group appeared better than the control group in the HBO single- and repeated-exposure groups. This difference did not reach statistical significance (Figs. 2 3A) . These observations suggest that hyperbaric stress causes early reversible general optical damage to lenses that is later repaired by natural mechanisms. The presence of oxygen at higher concentrations in the medium at this early stage prevents hyperbaric-stress–induced damage. If this is true, then oxygen, similar to its effect on general human survival, has a biphasic contradictory effect on lens survival. It helps the lens to survive and cope with external stresses at physiological conditions, but it may be toxic when allowed to reach overly high concentrations. Whether or not one accepts our explanations and speculations regarding our results, it would appear that there is no argument about the fact that oxygen causes damage to lenses, that this damage progresses from the periphery of the lens to its center (in accordance with oxygen diffusion) and that the process is cumulative—the larger the stress, the more severe the damage. 

A reduction in catalase activity on day 7 in the HBO exposure group was observed. O₂ and the hydroxyl free radical OH^· are involved in the damage caused to the catalase enzyme. ^{24 25} The reduction in catalase activity enables a rise in H₂O₂ and oxygenation of −SH groups in proteins. The lenses exposed to HBO showed greater damage to their catalase, probably because the amount of O₂ and OH⁻ was higher in their environment. NBO conditions do not seem to be sufficient to damage lenticular catalase. It appears that catalase has a role in protecting the lens from oxidative damage. The reduction in catalase activity caused by repeated exposure to high oxygen concentrations may explain (at least in part) the observed damage to lenses lacking part of this protective activity. ATPase activity was clearly lower in lenses exposed to HBO when compared with control lenses. Increased concentration of oxygen does not explain the reduction in ATPase activities by means of chemical equilibrium, as oxygen is not a direct reactant in the reaction catalyzed by this enzyme. Therefore, the reduction in enzyme activities must be caused either by toxicity caused by oxygen (or its free radicals) or by its being used for energy needs of antioxidant defense mechanisms of the lens. 

In both photomicrographs and light microscopy, structural changes were observed in the study group lenses compared with control lenses. Although we do not fully understand the mechanisms responsible for these changes, we can speculate that they were the outcome of the oxygen-induced tissue damage. Even without comprehensive understanding of the processes leading to the lenticular damage, it appears that in both examination techniques the damage was both oxygen dependent (the heavier the oxygen load faced by the lens, the more damage observed) and centripetal. (The lenses were affected first in the periphery, then in the center. For damage to reach the center of the lens, a higher oxygen load was needed. In lower oxygen loads only the periphery was damaged.) 

These observations are in agreement with our findings regarding the effects of oxygen on focal lens variability in the present report. We believe that the focal lens variability was the outcome of structural damage. The reduction in catalase and ATPase activities may indicate that the structural changes resulted from oxygen toxicity to lenticular enzymes. The consistency of the findings in different measuring systems supports the existence and nature of the observed lenticular oxygen-induced damage. 

One must be cautious when trying to extrapolate results of nonphysiological animal studies to processes occurring under physiological conditions in the human body. Still, the results of the present study, especially their consistency among all different study groups and five different measuring systems (optical quality, photomicrography, histology, catalase activity, and ATPase activity), justify suggesting the possible role of oxygen toxicity in human cataract formation. 

 Presented at the annual meeting of the Association for Research in Vision and Ophthalmology, Fort Lauderdale, Florida, May 2002.

 Submitted for publication February 5, 2003; revised March 5, 2003; accepted March 7, 2003.

 Disclosure: S. Schaal, None; I. Beiran, None; I. Rubinstein, None; B. Miller, None; A. Dovrat, None

 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.

 

Figure 1.

 

Schematic representation of distribution of rays in the examined beam. Longitudinal view (left): P, peripheral rays; I, intermediate rays; C, central rays. Vertical view (right): A, peripheral rays; B, intermediate rays; C, central rays. FLV for central rays was calculated for the three most central rays, and FLV for peripheral rays was calculated for the six most peripheral rays.

Figure 2.

 

FLV in the HBO single-exposure group. Single HBO exposure did not affect lens optical quality. No significant difference in FLV between study and control group lenses was observed throughout the study period.

Figure 3.

 

FLV in the repeated HBO-exposure group. Repeated exposures to HBO caused significant damage to lens optical quality. Damage was observed in both the center and periphery of the lens. (A) All rays: no difference in FLV was observed between study and control groups up to 120 hours in culture. FLV was 2-fold and 3-fold higher at 144 and 168 hours, respectively, in the study group compared with control. (B) Central rays: no difference was observed between study and control lenses up to 120 hours in culture. FLV was 2-fold higher after 144 hours and 2.5-fold higher after 168 hours in the study group compared with control. (C) Peripheral rays: no difference was observed between study and control lenses up to 72 hours in culture. FLV was 4-fold higher at 96, 120, 144, and 168 hours in the study group compared with control.

Figure 4.

 

FLV in the repeated NBO-exposure group. Repeated NBO exposure caused minor damage to the optical quality of the whole lens, no damage to the center of the lens, and significant damage to the periphery of the lens. (A) All rays: no difference in FLV was observed between study and control lenses up to 144 hours in culture. After 168 hours, FLV was 1.5-fold higher in the study group compared with control. (B) Central rays: no difference in FLV between the study and control groups was observed throughout the study period. (C) Peripheral rays: no difference in FLV was observed between the study and control lenses up to 96 hours in culture. FLV was 4-fold higher at 120, 144, and 168 hours in the study group compared with control.

Figure 5.

 

FLV in the repeated HNO-exposure group. HNO caused transient early minor damage to whole lens optical quality but no damage was observed in central or peripheral rays examined separately. The damage subsequently resolved. (A) All rays: FLV was increased in the study group compared with control at 48 and 96 hours (the difference was statistically significant at 96 hours). At 24 hours and from 96 hours up to the end of the study period, no difference in FLV was observed between the study and control groups. (B) Central rays: no difference in FLV was observed between central rays of the study and control lenses throughout the incubation period. (C) Peripheral rays: no difference in FLV was observed between peripheral rays of the study and control lenses throughout the incubation period.

Figure 6.

 

Catalase activity on day 7 in the study and control groups. Repeated HBO exposure (B) caused a decrease in catalase activity on day 7 compared with control. Single HBO (A), repeated NBO (C), and repeated HNO (D) exposures had no effect on catalase activity on day 7.

Figure 7.

 

Histochemical localization of ATPase activities (indicated by black clumping) in (A) control group lenses and (B) repeated HBO-exposure group lenses on day 7. The area of black clumps used as an indicator for ATPase activity showed a significant (P < 0.01) difference between the study (repeated HBO exposure) and control lenses. Magnification, ×10.

Figure 8.

 

Photomicrograph of (A) control lens on day 7 and (B) a repeated NBO-exposure lens on day 7. Note the location of the lens suture in the lower third of (B); the upper part shows the periphery of the lens. NBO exposure caused damage to the lens periphery, but not to its center. The uniform appearance of the lens material indicates an unharmed lens in the control group. In the study lens, subepithelial and cortical bubbles can be seen, whereas the center of the lens has a uniform healthy appearance with no bubbles. Magnification, ×20.

Figure 9.

 

Photomicrograph of a repeated HBO-exposure lens on day 7. Repeated HBO exposure caused damage to both the periphery and the center of the lens. Bubbles were spread throughout the lens (periphery and center). Magnification: (A) ×20; (B) ×40.

Figure 10.

 

Photomicrograph of a repeated HNO-exposure lens on (A) day 3 and (B) day 5 (×20). Repeated HNO exposure caused peripheral damage on day 3. Subepithelial bubbles with an intact center can be seen. Lens damage disappeared on day 5. Intact uniform lens material can be seen in both the center and the periphery. Magnification, ×20.

Figure 11.

 

Histologic section of control lens on day 7, showing undamaged lens, a regular dense fibrillar pattern, and intact epithelium. H&E; magnification, ×10.

Figure 12.

 

Histologic section of a repeated HBO-exposure lens on day 7. (A) Center and (B) periphery. Repeated HBO exposure had a negative effect on the histology of both the peripheral and central areas of the lens. Note the loose fibrillar pattern of both the lens center and periphery, indicating damage to both areas. H&E; magnification: (A) ×10; (B) ×40.

Figure 13.

 

Histologic section of a repeated NBO-exposure lens on day 7. (A) Central and (B) peripheral zones. Repeated NBO exposure caused damage to the periphery, but not to the center of the lens. (A) A regular dense fibrillar pattern in the center of the lens indicates undamaged lens tissue (B). A loose fibrillar pattern in the periphery of lens indicates damaged lens tissue. H&E; magnification, ×10.

The authors thank Yair Weiss, Samuel Fruchter, Yuval Argov, and Eyal Tadmor for assistance with the study animals, Alvira Bohomosov and Pessia Shentzer for assistance with the enzymatic and histologic studies, and Ruth Singer for editorial assistance.