d-β-Hydroxybutyrate Protects against Corneal Epithelial Disorders in a Rat Dry Eye Model with Jogging Board

Abstract

purpose. The purpose of this study was to establish a rat dry eye model of corneal epithelial disorders by inducing improper tear dynamics and change in blink frequency. The protective effect of d-β-hydroxybutyrate (HBA) on the corneal epithelia was also investigated.

methods. A series of treatments were performed under continuous exposure to low-humidity airflow. Rats were placed on a jogging board (JB) made of a plastic pipe for 7.5 h/d, and, for 16.5 hours, they were placed in individual cages without JB treatment. The resultant changes in tear dynamics and corneal epithelial structure were then analyzed. Five days after the rats were exposed to the treatment, eyes that showed corneal fluorescein staining were examined, to investigate the effect of HBA, by administration of eye drops containing 80 mM HBA four times daily during JB treatment for 5 days.

results. Significant reductions in blink frequency, Schirmer score, and tear clearance were recorded during JB treatment in eyes that showed persistent punctate staining of almost one half of the corneal surface. The application of HBA-containing eye drops significantly reduced the punctate staining compared with the initial or phosphate-buffered saline–treated eyes.

conclusions. This rat dry eye model, established by repeated JB treatment in desiccating conditions, induced abnormal tear dynamics and superficial punctate keratopathy similar to that in humans. These findings suggest the potential clinical application of HBA in corneal surface epithelial disorders in patients with moderate to mild dry eye.

Dry eye represents various abnormal states involving the quality and/or quantity of the tear film and integrity of ocular surface cells. ^{1 2} Because dry eye consists of various clinical subtypes, its etiology is multifactorial. Much effort has been made to develop animal models of dry eye that mirror the etiologic factors of each clinical subtype, for the basic investigation of the pathophysiology of the disease. To induce abnormal changes in tear dynamics in experimental animals, various treatment methods have been attempted, such as pharmacologic blockade of cholinergic muscarinic receptors, ^{3 4} surgical excision of the lacrimal glands ^{5 6} or mechanical prevention of blinking. ⁷ However, these techniques are disadvantageous, because it is difficult to exclude the complex influence of surgical insult or the adverse effects of pharmacologic agents. Tear dynamics are maintained by a complicated arrangement of the blink frequency and tear production, drainage, and evaporation from the ocular surface. ⁸ Therefore, in addition to applying minimally invasive procedures, the key concept of developing an animal model of dry eye involves inducing the ocular surface disorder by deranging the balance of each factor essential in the maintenance of proper tear dynamics. This strategy would further the understanding of multiple pathophysiologies and the development of new treatments against dry eye. 

d-β-Hydroxybutyrate (HBA) is a ketone body produced by hepatocytes and astrocytes through the degradation of long-chain fatty acids. ^{9 10} It exists abundantly in human plasma and peripheral tissues, in which levels are maintained below 0.1 mM in the normal state. ¹⁰ The role of HBA as an alternative energy source during glucose starvation or hypoxia in the brain has been well investigated. ^{11 12 13} Recently, in vivo and in vitro evidence showed that HBA prevents neuronal damage after challenge from neurotoxins that inhibit dopaminergic neuron activity. ^{14 15} These findings suggest that HBA plays a therapeutic role in preventing neurodegenerative conditions such as Parkinson’s disease. In ocular surface disorders, we showed that topically applied HBA ameliorates the appearance of acute-phase corneal epithelial erosion through suppression of apoptosis in a tear fluid depletion–induced rat dry eye model. ¹⁶ The potency of 80 mM HBA was the same as for 20% serum, in which the efficacy of clinical application has been well proven. ^{17 18 19} These findings suggest that there is potential for the use of HBA in ophthalmic formulations for curing ocular surface epithelial disorders in patients with dry eye. However, the effect of HBA on mild to moderate types of dry eye has not been investigated. 

The purpose of this study was to establish a rat dry eye model of superficial punctate keratopathy (SPK), a hallmark of corneal surface disorders of mild to moderate types of dry eye, and assessed the effect of HBA. For development of corneal surface disorders, we induced dysfunctions in tear dynamics by a novel treatment method, which was inspired by the evidence that visual tasking is accompanied by a change in tear dynamics, abnormal blink frequency, and symptoms of dry eye. ^{20 21 22 23 24}  

Materials and Methods

Animals

Female 8-week-old Sprague-Dawley rats (Tokyo Laboratory Animal Science, Tokyo, Japan) were used for this study. They were quarantined and acclimatized before the experiments for 1 week under standard conditions (SC) as follows: room temperature 23 ± 2°C, relative humidity of 60% ± 10%, alternating 12 hour light–dark cycle (8 AM to 8 PM), and water and food available ad libitum. All procedures were performed according to the ARVO statement for the Use of Animals in Ophthalmic and Vision Research. 

Establishment of a Rat Dry Eye Model

Study Design.

For the investigation of changes in tear dynamics, we compared the blink frequency (n = 4), tear fluorescein clearance (n = 8), and Schirmer score (n = 16) within two groups: rats placed in desiccated conditions (DC) for the entire time with daily jogging board (JB) treatment (DC+JB) and those placed in DC for the entire time without JB treatment (DC). For the investigation of pathologic changes, we compared corneal epithelial fluorescein staining (n = 16), histopathological changes (n = 8–10), and barrier function (n = 10) within three groups: DC+JB, DC alone, and SC with 7.5 h/d JB treatment (SC+JB). 

Dry Eye Treatment.

A schematic representation of the dry eye treatment used is shown in Figure 1 . 

Desiccated Conditions.

A series of treatments was performed in DC, with room temperature of 23 ± 2°C, relative humidity 25% ± 2%, and constant air flow at 2 to 4 m/s. 

JB Treatment.

Each rat was stationary, placed for 7.5 h/d between 9 AM and 5 PM on a JB. The JB was positioned outside of the cage. The seat of the JB was made of plastic piping 30 mm in diameter × 50 mm in length and suspended 60 cm above the bottom and 30 cm below the top frame by a wire. To prevent the rat from slipping from the JB, the seat was covered with metallic mesh. The interval between neighboring JB seats was kept at more than 30 cm. Along with the JB treatment, the rats were exposed to constant air flow at 2 to 4 m/s toward the face, produced by an 18-cm diameter electric fan (Morita Denko, Osaka, Japan) that was fixed 25 cm horizontally from the JB seat. Although the rats rarely left the JB, we checked on the condition of each rat at least once every hour during the JB treatment (Fig. 1A) . 

After 4 hours on the JB, the rats were returned to their cages for 30 minutes for food and water and again placed on JB for 3.5 hours. For the remaining 16 hours, they were individually placed in cages with water and food ad libitum (Fig. 1 B) . This series of treatments was repeated for up to 10 days. 

Assessment of Changes in Tear Dynamics

In the DC+JB group, each measurement was performed immediately before, after, and at the end of the JB treatment, to determine the blink rate and tear fluorescein clearance, and the Schirmer test was performed before and at the end of the treatment. In the DC group, each measurement was performed at the same time, corresponding to before and at the end of the JB treatment for blink rate and tear fluorescein clearance and at the end for the Schirmer test. Nontreatment measurements were taken under SC. 

Measurement of the Spontaneous Blink Rate

The cephalic region of each rat was continuously monitored for 30 minutes from a horizontal angle with a digital video recorder, and the blink rate was counted by analysis of the video images. 

Tear Fluorescein Clearance

Tear clearance was evaluated by measuring of the residual ratio of fluorescein in a tear. Two microliters of 1% fluorescein was instilled into each eye and after 2 hours, stained tear fluid was collected by a cotton thread. The thread was placed on the eyelid margin for 15 seconds, and tear samples were extracted with 100 μL of phosphate-buffered saline (PBS). The collected samples were then placed in a 96-well plate and read at an excitation wavelength of 485 nm and an emission wavelength of 530 nm (CytoFluor 4000; PerSeptive Biosystems, Framingham, MA). The retained fluorescein was expressed as a percentage of the instilled concentration. 

Schirmer Test

We used a modified Schirmer test on the rats’ eyes to measure tear fluid secretion under topical anesthesia induced with a 0.4% oxybuprocaine hydrochloride solution (Santen Pharmaceutical, Osaka, Japan). After 3 minutes of anesthesia, a phenol red thread (Zone-Quick; Menicon, Nagoya, Japan) was placed on the temporal side of the lower eyelid margin for 1 minute. The length of the moistened area from the edge was then measured to within 1 mm. 

Assessment of Pathologic Changes

Corneal Fluorescein Staining.

Changes in the corneal surface were determined by applying a fluorescein solution under a blue-free barrier filter, ²⁵ and corneal staining of the area was graded. The total area of punctate staining was denoted as grade 0 when there was no punctate staining, grade 0.5 when less than one sixteenth was stained, grade 1 when less than one eighth was stained, grade 2 when one fourth was stained, grade 3 when greater than one half was stained, and grade 4 when the entire area was stained. Representative corneas for each staining grade are shown in Figure 2A . For preliminarily confirmation of the validity of the grading method, the correlation between the fluorescein grade and stained area was analyzed in 50 randomly selected eyes (Fig. 2B) . There was a significant correlation (R = 0.914; P < 0.0001; Spearman rank correlation coefficient) between our grading criteria and the fluorescein stained area of the cornea, suggesting that our criteria permitted us to assess objectively the level of punctate staining. 

Corneal Epithelial Barrier Function.

Rats were anesthetized with pentobarbital sodium, and then 5 μL of 0.5% fluorescein sodium solution was instilled into the conjunctival sac. The eyes were kept closed with surgical tape for 10 minutes, and then the excess fluorescein was washed out with saline. The eyes were then held closed for an additional 20 minutes. The fluorescein intensity of the central cornea was measured with a slit lamp fluorophotometer (FL-500; Kowa, Tokyo, Japan) which was modified for rats. The fluorescence intensity was measured eight times, and the background mean fluorescence level was subtracted and averaged. The fluorescein penetrance was then expressed in terms of photon counts per millisecond. 

Histopathologic Examination.

The rats were killed with an overdose of a mixture of ketamine and xylazine, and their eyeballs were removed and fixed in 10% formalin. After dehydration, the corneal specimens were embedded in paraffin, cross-sectioned, and stained with hematoxylin and eosin. They were subsequently examined with a light microscope. 

Microscopic Morphometry.

Morphometric analysis was performed on corneal sections by a modified method as previously described. ²⁶ The cornea was cut at 4 μm along the equator and stained with hematoxylin and eosin, and five images, each 100 μm in length and equally spaced over the length of the cornea section, were digitally photographed. The 5-μm interval dots were superimposed on the image of the cornea using image-analysis software, and the number of dots was counted on the corneal epithelia, in the superficial layer including the wing cell layer and in the basal cell layer. Finally, the average thickness (in micrometers) was calculated. 

Scanning Electron Microscopy.

Corneas were fixed with 2.5% glutaraldehyde in 0.1 M phosphate buffer (pH 7.4) and postfixed with 1% osmium tetroxide in 0.1 M phosphate buffer. After dehydration in a graded series of ethanol, the specimens were freeze dried with t-butyl alcohol, sputter-coated with platinum, and examined by scanning electron microscopy (X-650; Hitachi, Tokyo, Japan). 

Assessment of the Effect of HBA by Topical Eye Drop Application

Sodium HBA was synthesized by Ophtecs Co. (Osaka, Japan) and its purity was >99% when tested by HPLC analysis. For the ophthalmic solution, 80 mM of HBA (wt/vol) was formulated in PBS, and the osmolarity was adjusted with NaCl from 290 to 300 mOsM. We selected the absolute concentration of HBA on the basis of the results of our previous in vivo study. ¹⁶  

A series of assessments were performed during our dry eye treatment schedule. Five days after the rats were exposed to dry eye treatment, eyes that showed fluorescein staining higher than grade 3 were used in the examination. Eyes with SPK were randomly selected for HBA eye drops or PBS as a control. Five microliters of eye drops were then given every 2 hours for 8 hours, during which the rats were placed on a JB. After starting the 5 days of application, we used corneal fluorescein staining (n = 9 to 11), morphometric analysis of the epithelia (n = 10 to 11), blink frequency analysis (n = 4), tear fluorescein clearance analysis (n = 16), and Schirmer scoring (n = 10) to evaluate the effect of HBA. 

Statistical Analysis

For the fluorescein staining score, data were analyzed by the Wilcoxon signed ranks test or the Mann-Whitney test comparison of two groups, and the Steel test was used for multiple comparisons. Other data were analyzed by the Student’s t-test for comparison of two groups and the Dunnett test for multiple comparisons. Differences were accepted as being statistically significant at P < 0.05. 

Results

Before starting the experiments, the rat were quarantined and acclimatized in SC for 1 week. 

Tear Dynamics Changes

Blink Rate.

Figure 3Ashows the variation in blink frequency in the rats. From days 1 to 10, similar patterns of variation in the blink frequency were repeated in the DC+JB and DC group. In the DC+JB group, before placement on the JB, measurements were performed in the cage in rats maintained in DC, and the blink rates were higher by 1.5-fold compared with the nontreatment rates measured under SC. Immediately after placement on the JB, a dramatic reduction in the blink rate was observed and was sustained until the end of the JB treatment. Significant blink rate reductions were observed in the DC group at all post-JB times compared with before placement on the JB on each treatment day (Fig. 3A , left). In the DC group, a slight increase in frequency was observed compared to the nontreatment values measured under SC (Fig. 3A , right). 

Tear Fluorescein Clearance.

Figure 3Bshows the changes in tear fluorescein clearance. From days 1 to 10, similar patterns of variation in tear fluorescein clearance were repeated in the DC+JB group. Before JB treatment, the amount of retained fluorescein was decreased 0.75-fold compared with the nontreatment value, and during JB treatment, a significant increase in retained fluorescing was detected immediately after and at the end of treatment. Significant increases were observed compared with before JB treatment on day 5 and at all times corresponding to the times for the DC group (Fig. 3B , left). In the DC group, a 0.75-fold decrease of retained fluorescein was detected compared with the nontreatment levels during the experimental period (Fig. 3B , right). 

Schirmer Test.

Figure 3Cshows the changes in the Schirmer score for the DC+JB (left) and DC (right) groups. We did not apply the nontreatment values measured under SC, because the difference was nonsignificant when the Schirmer scores were compared in different humidities. For the DC+JB group, a reduction in the Schirmer score was observed at the end of JB treatment compared with before treatment and in the DC group from days 1 to 10. A significant reduction was also observed compared with the DC group on day 5 and before JB treatment on day 10. 

These observations suggest that no adaptation had arisen to affect the changes in tear dynamics during JB treatment throughout the experimental periods. 

Pathologic Features

Corneal Fluorescein Staining.

On day 1 for the DC+JB group, significant increases in the corneal staining score were observed at the end of JB treatment compared with before JB treatment, and on days 5 and 10, the score increased further before and at the end of JB treatment (Fig. 4A) . In contrast, for the DC or SC+JB groups, no apparent changes in the score were observed throughout the experimental period. The average score for the DC+JB group was significantly increased compared with that for the DC and SC+JB groups at all time points throughout the experiment. Representative patterns for the DC+JB group are shown in Figures 4B and 4C . The characteristic alteration of the corneal fluorescein staining pattern for the DC+JB group closely resembled that for human SPK (Fig. 4C) . 

Histopathological Examination.

Representative patterns of histopathologic changes in the corneal epithelia are shown in Figure 5 . In the DC+JB group, no apparent change was observed on day 1 compared with the untreated cornea (Fig. 5B) . However, thinning of the corneal epithelia was apparent on days 5 and 10. Furthermore, the basal cells showed abnormalities in size and arrangement (Figs. 5C 5D) . In contrast, no apparent changes were noted in corneal sections of the DC group and SC+JB groups up to day 10 (Figs. 5E 5F) . 

In the DC+JB group, significant increases in the corneal stain scoring, as well as thinning of the corneal epithelia, were sustained for at least up to 30 days of treatment, and recovery was observed within 3 days when the rats were again kept under normal conditions after the 10 days of DC+JB treatment (data not shown). These results suggest that the dry eye treatment was necessary to produce sustained changes in the corneal epithelia. 

Microscopic Morphometry.

In the DC+JB group, microscopic examination showed significant reductions (nonteatment, 40 μm; DC+JB, 29 μm) in the epithelial thickness compared with the untreated cornea in the DC and SC+JB groups on day 10 (Fig. 6A) . The thickness of the superficial layer, including the wing cell layer, and that of the basal cell layer were significantly reduced compared with the untreated cornea in the DC+JB group (Fig. 6B) . 

Corneal Epithelial Barrier Function.

To confirm the relationship between the morphologic change in the epithelia and the barrier function, we measured the fluorescein permeability of the corneal epithelium on day 10. A significant increase in fluorescein permeability was observed only in the DC+JB group compared with the initial value (Fig. 7) . The permeability in the DC+JB group was 1.5-fold higher than that in the DC or SC+JB groups. 

SEM Analysis of Corneal Epithelia

Figure 8shows SEM images of the corneal epithelia on day 10. In the untreated cornea, two types of superficial cells were detected: dark and bright cells (Fig. 8) . The dark cells had a flat surface, and the bright cells were covered by high-density microvilli and microplicae (Fig. 8B) . In the DC+JB group, a large area was covered with the dark cells (Fig. 8C) . Finally, in the DC and SC+JB groups, no changes were noted compared with the untreated cornea (data not shown). 

Effect of HBA

Effect on the Corneal Epithelia.

The use of HBA dramatically reversed the fluorescein staining score after 5 days of treatment, and significant decreases were observed compared with the initial (P < 0.01) and PBS treatment (P < 0.05; Fig. 9 ) scores. In the PBS group, slight decreases in the fluorescein staining scores were observed, although the differences were not significant compared with the initial scores (Fig. 9A) . Representative patterns of SPK restored by HBA are shown in Figures 9Band 9C . 

Consistent with the reduced fluorescein staining, a significant recovery of the corneal epithelial thickness was observed compared with the initial and PBS treatment thicknesses (P < 0.01). The thickness with HBA treatment were almost equal to those in untreated corneas (Fig. 9D) . 

Effect on Tear Dynamics.

No significant changes were noted in the blink frequency, tear fluorescein clearance, or Schirmer score between the HBA- and PBS-treated groups, suggesting that HBA does not affect tear dynamics (Table 1) . 

Discussion

In this study, for the first time, a rat model of moderate dry eye was established by a novel treatment method: persistent strain by JB treatment in combination with exposure to an evaporative environment, which induces disordered tear dynamics and abnormal blink frequency. Recent progress in understanding the pathophysiology of dry eye has demonstrated that the pathogenic mechanism is not limited to dysfunction of the lacrimal apparatus but also involves external factors such as a dehydrating environment ^{27 28 29 30} and visual tasking, ^{23 24 31 32} which induce changes in tear dynamics and blink frequency. Thus, our treatment procedure could be applicable to the study of multiple pathophysiologies and lead to the development of new treatments for dry eye. 

Blinking is necessary to provide a controlled environment that maintains the integrity of the ocular surface by proper formation of a tear film and support of the lacrimal pump system. ^{33 34} Various psychological factors and/or ocular conditions such as mental tension, eye irritation, ocular fatigue, or performance of visual tasks modify the frequency of blinking. ^{35 36} When rats were placed on the JB in DC, a reduction in the blink frequency was observed during repeated treatment. Vision and eye movement play a primary role in maintaining postural equilibrium, and the fixation of vision promotes postural stability. ^{37 38} During the time that the rats were settled on the JB, postural equilibrium was constantly disturbed, making continuous visual fixation necessary to promote postural stability. The visual performance necessary for visual display terminal use, driving a car, or reading is associated with a reduction in blinking. ^{23 32 39} Therefore, the close relationship between postural equilibrium and visual performance may play a role in the mechanism responsible for the reduction in the blink frequency observed during JB treatment. 

In addition to the reduction in the blink frequency, decreases in the Schirmer score and tear clearance were observed during JB treatment. We suspect that the reduction in tear production arose from the restrained response of the autonomous nervous system, which was elicited from persistent strain due to the adjustment of postural equilibrium. As a result of insufficient drainage with prolonged exposure of the ocular surface to an evaporative environment due to the decreased blink frequency and reduced tear production, tear clearance may have been significantly reduced. To clarify the mechanism responsible for the changes in tear dynamics during the dry eye treatment, physiologically visual and praxiologically based investigations are needed. 

During our dry eye treatment, each factor used to determine the tear dynamics tear production, tear drainage, and evaporation from the ocular surface exhibited a declining trend for the healthy ocular surface. Thus, our results suggest that the simultaneous aggravation of these factors may be a critical risk factor in the pathogenesis of dry eye. 

In animals and humans, there have been few investigation made into the relationship between the appearance of SPK and structural changes in the corneal epithelia. Our results showed that chronic SPK in the rat is accompanied by thinning and abnormal arrangement in the superficial layer including wing cells and in the basal columnar cell layer, poorly developed microvilli on surface cells and a reduction in barrier function. This indicates that the function and structure of the epithelia changed to an abnormal state and that this state was sustained during the dry eye treatment. The mature corneal epithelial cell layer is maintained by a balance between shedding superficial cells, cell proliferation and differentiation of the basal layer, and centripetal migration of cells from the limbus. ⁴⁰ Taken together, our findings suggest that not only changes in surface cell integrity but also improper differentiation of the corneal epithelial cell layer are involved in the pathogenesis of SPK. The homeostasis of ocular surface cells is supported by proper interaction between nutrients, growth factors, cytokines, and retinoids presented in tear fluid. ^{1 41 42 43 44} For our dry eye condition, due to abnormal tear dynamics, prolonged imbalances in the composition, and the concentration of these factors may have been induced in a precorneal tear film that lead to the disturbance of corneal epithelial differentiation. 

Our present study showed that topically applied HBA restores chronic SPK and thinning of the corneal epithelial cell layer. Previously, we also showed a similar effect of HBA on acute corneal epithelial degeneration due to thinning of the cell layer accompanied by extensive exfoliation in our rat dry eye model. ¹⁶ In neuronal tissues, HBA decreases the rate of cell death in Alzheimer’s and Parkinson’s disease models ^{14 15 45} and plays a role in preserving neuronal integrity during development. ⁴⁶ Together the findings show that HBA does not affect the tear dynamics and may directly reverse epithelial cell degeneration, leading to the normalization of corneal epithelial differentiation under dry eye conditions. Recent investigations suggest that the ability of HBA to protect against neurotoxins, which cause dopaminergic neurodegeneration deficits reminiscent of Parkinson’s disease, is related to enhancing of the energy status ^{15 45} or suppression of free radical production, ¹⁴ by preserving the neuronal mitochondrial respiratory chain function. Although in dry eye little is known about the involvement of mitochondrial dysfunctions in corneal epitheliopathy, it is possible that this function plays a role in the protective effect of HBA. 

In conclusion, we established a rat dry eye model of tear dynamic dysfunction that mirrors etiologic features in humans. The results of this study also suggest the potential usefulness of HBA for the clinical treatment of ocular surface epithelial disorders in patients with chronic symptoms of dry eye. 

 Submitted for publication November 18, 2004; revised March 10, 2005; accepted March 21, 2005.

 Disclosure: S. Nakamura, Ophtecs Corp. (E); M. Shibuya, Ophtecs Corp. (E); H. Nakashima, Ophtecs Corp. (E); T. Imagawa, None; M. Uehara, None; K. Tsubota, 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.

 

Representations of the rat dry eye treatment used in this study. Schematic of treatment on the jogging board (A), and a daily treatment schedule (B).

Figure 2.

 

The grading scale used for the observed fluorescein patterns was 0, no punctate staining, to 4, staining over the entire area of the cornea (A). The correlation between our grading criteria and the fluorescein-stained area on the cornea was calculated by the Spearman rank correlation coefficient test (R = 0.914; P < 0.0001) (B). Digital images of the punctate stained area were quantified with image analysis software (Image for Windows; Scion Corp., Frederick, MD).

Figure 3.

 

The change in tear dynamics during dry eye treatment. DC+JB group (right), DC group (left), blink frequency (A), tear fluorescein clearance (B), Schirmer test (C). Significant reduction in blink frequency, Schirmer score and tear clearance were revealed during JB treatment. The data represents the mean ± SE of 4 blink frequencies, 8 tear fluorescein clearance levels, and 16 Schirmer scores. *P < 0.05, **P < 0.01 versus the DC group by t-test; #P < 0.05, ##P < 0.01 versus before JB treatment on the corresponding day by t-test for the Schirmer score or the Dunnett test for blink frequency and tear fluorescein clearance.

Figure 4.

 

Changes in the fluorescein score during each treatment (A). (B) Typical initial pattern and (C) in the same eye after 5 days of treatment with DC+JB. The average score in the DC+JB group was significantly increased compared with the DC and SC+JB groups at all time points. The data represent the mean ± SE of results in 16 eyes. **P < 0.01 versus the DC and SC+JB group; #P < 0.05 versus before treatment on day 1 by the Steel test.

Figure 5.

 

Histopathological change in the corneal epithelial during dry eye treatment. Nontreatment (A); DC+JB group on days 1 (B), 5 (C), and 10 (D); and DC (E) and SC+JB (F) groups on day 10. Thinning of the corneal epithelial cell layer accompanied by arrangement of the basal cell layer appeared on days 5 and 10 in the DC+JB group. Bar, 100 μm.

Figure 6.

 

Microscopic morphometry of the corneal epithelia after 10 days of dry eye treatment. Change in the corneal epithelial thickness of each treatment group (A), changes in the corneal epithelial superficial layer with wing cells and in the basal columnar cell layer in the DC+JB group (B). A significant reduction in the epithelial layer was observed in the DC+JB group compared with the untreated cornea. The data represent the mean ± SE for 8 to 10 corneas. **P < 0.01 versus DC and SC+JB by the Dunnett test; ##P < 0.01 versus nontreatment by the Dunnett test (A) or the t-test (B).

Figure 7.

 

Changes in the corneal epithelial barrier after 10 days of dry eye treatment. A significant increase in fluorescein permeability was observed in the DC+JB group compared with the initial value. The data represent the mean ± SE of 10 eyes. *P < 0.05 versus the initial value in the DC+JB group, by t-test.

Figure 8.

 

SEM analysis of the corneal epithelial surface after 10 days of dry eye treatment. Nontreated cornea (A), a higher magnification of nontreated cornea. The dark cells had a flat surface, and the bright cells were covered by high-density microvilli and microplicae (B), DC+JB group. A large area was covered with darkish surface cells (C). Bar: (A, C) 100 μm; (B) 1 μm.

Figure 9.

 

The effects of topically applied HBA on the corneal epithelia; rats that showed corneal staining up to grade 3 were used. Drops were applied at 2-hour intervals for 8 hours during JB treatment, and assessments were performed 5 days after the beginning of treatment. (A) Effect on the corneal fluorescein staining score. (B) Typical initial pattern. (C) Pattern in the same eye after 5 days of treatment with 80 mM HBA. (D) Effect on the recovery of corneal epithelial thickness. Significant recovery of corneal fluorescein staining and the epithelial thickness was observed compared with the initial and PBS treatment levels. The data represent the mean ± SE of 10 to 11 eyes. **P < 0.01 versus PBS by the Dunnett test for epithelial thickness; *P < 0.05 versus PBS by the Mann-Whitney test for the fluorescein staining score; ##P < 0.01 versus the initial value by the Wilcoxon signed ranks test for the fluorescein staining score or by the Dunnett test for epithelial thickness.

Table 1.

 

Changes in Tear Dynamics after Dry Eye Treatment with HBA or PBS

	Eyes (n)	PBS	d-β-Hydroxybutyrate
Blink frequency (times/min)	4	5.04 ± 0.58	5.41 ± 1.43
Fluorescein retained (% of instilled concentration)	16	0.037 ± 0.008	0.034 ± 0.009
Schirmer score (mm/min)	10	11.8 ± 0.79	11.8 ± 1.02

 Each measurement was taken at the end of the JB treatment. The data represent the mean ± SE. No significant changes were noted between the HBA- and PBS-treated groups.

The authors thank Toyoaki Yoneda (Ophtecs Corp.) for his generous support.