Age-Related Changes in Rat Corneal Epithelial Nerve Density

Abstract

purpose. To determine the effect of aging on corneal epithelial nerve density in an animal model.

methods. Corneal whole mounts from rats aged 6, 12, 18, and 24 months were stained immunohistochemically with antisera against the pan-neuronal marker neurotubulin. Epithelial nerve terminals and subbasal nerves in standardized 1-mm² central and peripheral zones from each cornea were drawn using a drawing tube attached to a light microscope. Images were scanned, and nerve densities were calculated as the percentage of each 1-mm² area occupied by nerves. The diameters of subbasal nerves in 6- and 24-month old animals were measured. Subbasal nerve vortices were analyzed qualitatively with reference to location, morphologic appearance, and directionality.

results. Epithelial nerve terminal density decreased by approximately 50% between 6 and 24 months. The rate of decline was roughly linear and similar in both central and peripheral cornea. In contrast, subbasal nerve density increased by more than 50% between 6 and 24 months in both central and peripheral cornea. The mean diameter of corneal subbasal nerves decreased approximately 30% (0.384 μm vs. 0.271 μm) between 6 and 24 months. The morphologic appearance and directionality of the subbasal nerve vortex demonstrated considerable interanimal variability and did not correlate with age.

conclusions. Rat corneal nerve terminal density decreases, but corneal subbasal nerve density increases, as a function of age. The age-related loss of nerve terminal density seen in the rat cornea is in keeping with the decreased corneal sensitivity reported in elderly humans and may contribute to the pathogenesis of dry eye disease in aged persons.

The corneal epithelium, containing an estimated 7000 sensory nerve terminals per square millimeter of epithelium, is the most sensitive and densely innervated surface tissue in the human body. ¹ Corneal nerves promote the maintenance of the ocular surface by activating brain stem circuits that stimulate reflex tear production ² and by releasing trophic substances that support physiologic renewal of the corneal epithelium. ^{2 3}  

It has been reported that corneal sensitivity in human subjects decreases with increasing age and that this decrease becomes especially pronounced after the fifth decade of life. ^{4 5 6 7} Although multiple factors may contribute to age-dependent decreases in corneal sensitivity, the most likely is a progressive, age-dependent loss of corneal nerves. Despite the general acceptance of this hypothesis, morphologic data on this subject are extremely limited. Examination of human corneas by in vivo confocal microscopy (IVCM) have variously demonstrated either no change in subbasal nerve density in human corneas aged 15 to 79 ⁸ or a decrease in subbasal nerve density between young and old corneas. ^{9 10 11}  

Nerve endings represent the parts of the nerve fiber responsible for transducing sensory stimuli into nerve signals; thus, nerve terminal density is directly proportional to corneal sensitivity. ¹² Although IVCM can successfully image corneal subbasal nerves, it cannot reliably image intraepithelial nerve endings ⁸ ; thus, at the present time, the effect of aging on corneal nerve terminal density is largely conjectural. 

The purpose of the present study was to test the hypothesis that corneal nerve terminal and subbasal nerve densities decrease as a function of age. The hypothesis was tested quantitatively in a rat animal model by analyzing immunohistochemically stained nerves in corneal whole mounts at different ages. 

Methods

Immunohistochemistry

Fischer-344 rats aged 6, 12, 18, and 24 months (n = 3 per group) were purchased from Harlan Sprague-Dawley (Indianapolis, IN) from a colony subsidized by the National Institute of Aging. Animals were treated in accordance with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. Each animal was humanely killed with an overdose of sodium pentobarbital. Before enucleation, the superior pole of each cornea was marked at the corneoscleral limbus with an indelible pen. Whole globes were immersion-fixed for 15 minutes at room temperature (RT) in 4% paraformaldehyde–0.2% picric acid in 0.1 M phosphate-buffered saline (PBS). The anterior segment was then removed with a razor blade, and the cornea and approximately 1 mm of attached scleral rim were isolated and returned to the fixative solution for an additional 45 minutes. The corneas were then stored in 0.1 M PBS with 30% sucrose for 24 to 48 hours. 

Each cornea was mounted briefly on a dome-shaped post that approximated the radius of curvature of the rat cornea to facilitate subsequent cutting with a single-edge razor blade into standardized inferonasal, inferotemporal, superonasal, and superotemporal quadrants. The quadrants were then permeabilized by overnight incubation at 37°C in 0.1% EDTA (Sigma-Aldrich, Inc., St. Louis, MO) and 0.01% hyaluronidase (type IV-S, product no. H4272; Sigma-Aldrich) in 0.1 M PBS, pH 5.3. ¹³  

Tissues were rinsed for 90 minutes in PBS with 0.3% Triton X-100 (PBS-TX) and were blocked at RT for 2 hours in PBS-TX containing 1% bovine serum albumin (BSA; Sigma-Aldrich). Inferotemporal and inferonasal quadrants were incubated overnight at room temperature in a mouse monoclonal antibody against neuronal class III β-tubulin (TuJ1, 1:500; Covance Research Products, Berkeley, CA). The neurotubulin antibody recognizes a cytoskeletal component expressed in all peripheral axons, regardless of phenotype, and was used here as a “pan-neuronal” marker for corneal nerves. Superotemporal and superonasal quadrants were not examined in the present study but will be processed separately in a future investigation of age-related changes in corneal peptidergic innervation. 

The following morning, the quadrants were rinsed for 90 minutes in PBS-TX, followed by a 2-hour incubation at RT in secondary antibody (horse, anti-mouse IgG, rat-absorbed, 1:200; Vector Laboratories, Burlingame, CA). After a 90-minute rinse in PBS-TX, the tissues were incubated for 2 hours at RT in avidin-biotin-horseradish peroxidase complex (ABC reagent; Vector Laboratories), rinsed for 45 minutes in PBS, incubated for 8 minutes at RT in 0.1% diaminobenzidine (Sigma-Aldrich), and finally washed three times in PBS and twice in distilled water. The quadrants were then mounted onto subbed slides, air dried overnight, dehydrated in graded alcohols, cleared in xylene, and coverslipped with mounting medium (Permount; Biomeda, Foster City, CA). 

Data Analyses

Epithelial innervation densities were determined in 1-mm² central and peripheral zones from each left inferotemporal quadrant. The central zone was defined as a 1-mm² area whose apical border spanned exactly from edge to edge across the apical region of the quadrant (Fig. 1) . The peripheral zone was defined as a 1-mm² area whose peripheral border approximated the entry points of the most peripheral epithelial leashes. 

Corneal nerve densities in the 1-mm² zones were determined by careful drawing of the epithelial nerve fibers with the use of a drawing tube attached to an Olympus (Tokyo, Japan) light microscope equipped with a 40× objective (total final magnification, 312.5×). Two separate drawings were made from each 1-mm² zone: one showed only the subbasal fibers, and the other showed intraepithelial nerve terminals that derived from the subbasal fibers. All subbasal fibers and nerve terminals were normalized to the same diameter by drawing with a 0.5-mm diameter mechanical lead pencil. 

Completed drawings were scanned on a flatbed scanner connected to a personal computer. Images were then cropped to an area exactly 1 mm², and the nerves were highlighted using the thresholding tool of a public domain Java image processing program (Image J). Nerve density was calculated as the percentage of each 1-mm² area occupied by nerves. Differences in central and peripheral corneal nerve densities among groups were compared by analysis of variance (ANOVA). 

The number of subbasal nerves present from side to side and the average distance between them was determined for some of the central zones by placing a straight-edge inside the 1-mm² box perpendicular to the long axes of the subbasal nerves. The numbers of subbasal nerves that intersected the line were then counted at one apical and one peripheral location in each box. 

Subbasal nerve fiber diameters were also measured in the central corneal zones of two 6-month-old and two 24-month-old animals. High-magnification images of the central subbasal nerve plexus were captured with a digital camera mounted on a Leica (Wetzlar, Germany) microscope and exported to Image J. A calibrated line was applied perpendicularly to the long axes of the nerves, and the diameters of 100 consecutive subbasal fibers per central zone were measured using the Image J analyze-set scale utility. 

Finally, the location, morphologic appearance, and directionality (clockwise or counterclockwise) of the periapical subbasal nerve vortex ¹⁴ were assessed qualitatively in three to four corneas per age group. All subbasal fibers in an area 0.75 mm in diameter centered on the subbasal nerve vortex were drawn using a drawing tube attached to an Olympus light microscope and were converted to high-resolution black-and-white images by tracing with a rapidograph technical pen set. 

Results

Architecture of Rat Corneal Epithelial Innervation

The immunohistochemical protocol for corneal whole mounts stained completely all corneal subbasal nerves and their associated branches, from the points of nerve entry at the basal lamina to the most distal nerve terminals. In the account that follows, a brief review of the salient anatomic features of the rat corneal epithelial innervation will be presented first, followed by a description of the qualitative and quantitative changes that occurred with aging. 

Subbasal Nerves.

Corneal subbasal nerves originated from nerve fibers located in the subepithelial or limbal plexus. After penetrating the basal lamina, the subbasal fibers turned acutely in the horizontal plane and coursed between the basal lamina and basal epithelial layer in tightly clustered groups known as epithelial leashes ^{12 15 16 17} (Figs. 2a 2c) . The number of subbasal nerves composing each leash usually averaged between 3 and 8. Subbasal fibers often coursed for distances of 1 mm or longer, and they communicated with their neighbors by short, interconnecting links. Subbasal fibers in the peripheral cornea were oriented in a predominantly radial direction. In contrast, fibers in the intermediate and central cornea assumed gentle curvilinear trajectories and swept preferentially toward an area located approximately 0.2 to 0.4 mm inferonasal to the corneal apex. In most corneas, the subbasal axons in the vicinity of the apex converged in a gentle or an occasionally exaggerated spiral configuration resembling a whorl or a vortex. 

Intraepithelial Nerve Terminals.

The corneal epithelium overlying each subbasal fiber contained varying numbers of nerve terminals that originated from that subbasal fiber (Figs. 2b 2d) . Many nerve terminals branched once or twice before ending, and their distal tips typically displayed slightly bulbous varicosities. 

Age-Related Changes in Epithelial Nerve Density

Epithelial nerve terminal density decreased significantly as a function of age (Figs. 3 4 5) . The decline in terminal density was roughly linear between 6 and 18 months and then tapered to a more gradual rate of decline from 18 to 24 months. The magnitude of decline between 6 and 24 months of age was substantial and averaged 51% in the central cornea and 47.9% in the peripheral cornea. The rate of decline was similar in the central and peripheral corneal zones. 

In contrast to the decrease in nerve terminal density, subbasal nerve density increased as a function of age (Figs. 4 5 6 7) . The density of subbasal fibers increased slightly, but not significantly, between 6 and 18 months and then increased significantly between 18 and 24 months. Between 6 and 24 months of age, mean subbasal nerve density increased 63% in the central cornea and 55% in the peripheral cornea. The rate of increase did not differ between central and peripheral corneal epithelial zones. 

Mean numbers of central zone subbasal nerve fibers intersected by a 1-mm line drawn perpendicular to the long axes of the nerves were 101.7 ± 7.5, 124.8 ± 7.6, 136.5 ± 10.2, and 174.7 ± 13.0 at 6, 12, 18, and 24 months, respectively. Average distances between subbasal nerve fibers in the central zone were 9.8, 8.0, 7.3, and 5.7 μm at 6 to 24 months, respectively. Mean diameters of central cornea subbasal nerves decreased by almost 30% from 6 to 24 months (0.384 μm ± 0.097 μm and 0.271 ± 0.096 μm at 6 and 24 months, respectively; Fig. 8 ). 

The morphologic appearance of the subbasal nerve vortex did not correlate with age (Fig. 9) . Subbasal nerve density in the area of the vortex was higher in older animals than in younger animals, consistent with age-related changes in subbasal nerve density observed elsewhere in these corneas. Otherwise, the direction of spiral rotation, degree of curvature of participating subbasal nerve fibers, and location of vortex center relative to corneal apex varied considerably and at random among corneas in the same age group. Indeed, pronounced differences in vortex magnitude and directionality were often seen between left and right eyes of the same animal. 

Discussion

Methodologic Considerations

The tubulin immunohistochemical staining method used here provided optimal demonstrations of rat corneal nerve architecture, morphology, and density. The use of corneal whole mount preparations permitted visualization in three dimensions of the entire epithelial nerve network, from points of subbasal nerve origination at the basal lamina to the most superficial terminal branches, and provided a unique opportunity to assess independently (by changing the plane of focus) age-related changes in subbasal and nerve terminal densities. The use of whole mount preparations also permitted acquisition of nerve density data from large, 1-mm² areas of corneal epithelia and eliminated variability associated with smaller sample areas. Finally, the use of whole mount preparations permitted the simultaneous evaluation of nerve density changes in central and peripheral corneal areas. 

The immunohistochemical staining procedure in this study used a neurotubulin “pan-neuronal” marker that stains all corneal nerves regardless of origin or phenotype. Sympathetic innervation of the rat cornea is extremely sparse and is restricted largely to the corneoscleral limbus ^{18 19} ; thus, the changes in nerve density seen here almost certainly reflected age-dependent changes in corneal sensory innervation. 

By comparing the median lifespan data of male F344 rats (29.6 months ²⁰ ) and male inhabitants of the United States in the year 2006 (75 years ²¹ ), it is estimated that 1 “rat year” equals approximately 30.4 “human years.” Assuming for the purposes of this discussion a linear relationship, a 6-month-old F344 rat is the approximate lifespan equivalent of a 15-year-old human, and a 24-month-old rat is the lifespan equivalent of a 61-year-old human. 

Age-Related Changes in Corneal Nerve Terminal Density and Sensitivity

Results of the present study in an animal model provide unequivocal anatomic evidence of a significant decrease in corneal nerve terminal density as a function of age. The decrease was progressive, nearly linear, of a magnitude that approximated 50% over an 18-month period, and occurred at similar rates in both the central and the peripheral cornea. The data further show that age-related loss of corneal nerve terminal density begins early in life and is not a phenomenon restricted to advanced age. For example, a significant decrease in nerve terminal density occurred in young adult rats between 6 and 12 months of age, or roughly 15 to 30 years in human years. 

The anatomic data generated here may have implications for understanding age-related loss of corneal sensitivity in human subjects. Age-related loss of functionally intact nerve endings from the ocular surface is probably the single most important reason corneal sensitivity decreases in elderly patients; however, other factors, such as altered synaptic processing of corneal sensory information in the central nervous system, ⁸ age-related decreases in level of patient alertness, ⁶ and decreased nerve responsiveness resulting from altered tear film composition, ⁷ may also contribute to this phenomenon. 

Decreased corneal sensitivity in elderly patients is well documented clinically; however, the approximate age at which this decrease first occurs and whether the decrease proceeds in a linear or a nonlinear fashion is unclear. Boberg-Ans ⁴ and Millodot ⁵ measured the sensitivity of the human cornea with a mechanical aesthesiometer and concluded that sensitivity remained relatively unchanged until age 40 and then decreased significantly in a nonlinear (parabolic) fashion to about half this level by age 65 to 70. The area of cornea tested by Boberg-Ans ⁴ was not specified, whereas Millodot ⁵ tested a point near the limbus in the 6 o’clock position. Roszkowska et al. ²² reported that mechanical sensitivity of the central and peripheral human cornea declined at different rates, with peripheral sensitivity decreasing gradually throughout life and central corneal sensitivity remaining stable until age 60 and then decreasing sharply thereafter. Murphy et al. ⁶ used a noncontact corneal aesthesiometer to stimulate central corneal thermoreceptors and concluded that central corneal sensitivity declines in a linear fashion beginning at approximately age 20 and decreases by almost half between 20 and 50 years of age. The results of the present study agree most closely with the findings of Murphy et al. ⁶ because both studies reveal a steady, nearly linear rate of decline in sensitivity (human) or nerve terminals (rat) that begins early in life. 

Alterations in Subbasal Nerve Density

An unexpected finding of the present study was that subbasal nerve density increases significantly in rat corneas between 6 and 24 months of age. In contrast, the effect of aging on subbasal nerve density in human corneas, as determined by IVCM, is uncertain. The results of one clinical study show no correlation between patient age (15–79 years) and subbasal nerve density in the central cornea. ⁸ In contrast, the results of other IVCM studies show a slight decrease in central subbasal nerve density in older versus younger patients. ^{9 10 11} The reasons for the disparate clinical findings are unknown but may include differences in instrument resolution, light intensity setting, patient sample size, location and size of the corneal areas examined, and methods used to calculate nerve density from the confocal images. 

The mechanism by which subbasal nerve density increases in older rats could not be determined in the present study; however, two hypotheses are proposed. First, age-related loss of nerve terminals may provide a stimulus for new subbasal fibers to sprout from the subepithelial nerve plexus. Second, as the cornea ages, some large-diameter subbasal nerves may divide into multiple small-diameter subbasal nerves. Each daughter fiber, regardless of diameter or number of individual axons, would continue to be recorded in the drawing tube illustrations as a single 0.5-mm diameter “nerve fiber,” yielding an increase in the number of subbasal nerve fibers per unit area. 

In the present study, average subbasal nerve diameter decreased significantly between 6 and 24 months; however, in the absence of electron microscopic data, we were unable to determine whether this occurred through splitting of larger subbasal fibers into multiple smaller ones, axon degeneration, age-related reductions in axon diameter, or combinations of these. The results of this study differ from those reported for human corneas examined by IVCM; in the latter studies, subbasal nerve diameters did not change ⁹ or they increased ¹⁰ as a function of age. 

It should be emphasized that age-related changes in subbasal nerve density, whether assessed by immunohistochemistry or IVCM, must be interpreted cautiously because of resolution limitations inherent to these techniques. Subbasal nerves in rats and mice usually contain one to four and occasionally as many as 12 individual axons, ^{15 23} whereas subbasal nerves in human corneas can contain as many as 20 to 40 axons. ^{24 25 26} Thus, the true effect of aging on subbasal axon density remains unclear pending electron microscopic investigation. 

Human subbasal nerves (but not nerve terminals) are readily imaged by IVCM, and the density and morphologic appearance of the subbasal nerves are often assessed to determine the integrity of the corneal innervation in normal and diseased eyes. ^{11 27} The results of our study in an animal model reveal an increase in subbasal nerve density, but a decrease in nerve terminal density, with advancing age. These observations suggest that age-associated remodeling of the corneal epithelial innervation is a complex process and that clinicians should exercise caution in using changes in subbasal nerve density as the only criterion for evaluating the status of corneal innervation. 

The incidence of dry eye disease increases greatly in aged persons. ^{11 28 29} Age-related loss of functional sensory nerve terminals, such as shown in the present study in a rat model, may play a role in the pathogenesis of this disease by disrupting the brain stem circuit that produces reflex lacrimal gland secretion. 

Subbasal Nerve Vortex

The whorl-like patterns of subbasal nerves seen here in the periapical inferonasal region of the rat cornea confirm and extend previous observations of spirallike assemblages of subbasal nerves in similar locations in murine and human corneas. ^{14 30} The mechanisms that govern the formation of these whorl-like assemblages remain uncertain; however, it has been postulated that basal epithelial cells and subbasal nerves may migrate centripetally in tandem ³¹ and that one or the other may provide a substrate that directs and patterns the migration of the other. Chemotrophic guidance, electromagnetic cues, and population pressures have been suggested as possible mechanisms for the whorl-like patterns of basal epithelial cells observed near the mouse corneal apex. ^{32 33} Results of the present study have documented for the first time the presence of subbasal nerve vortices in the rat cornea, but they provide no evidence of age-related changes. 

Summary

In conclusion, the results of this study have demonstrated in a rat model a significant decrease in corneal nerve terminal density and a concurrent significant increase in subbasal nerve density as a function of age. The decrease in nerve terminal density begins early in life and occurs at similar rates in the central and peripheral cornea. It is tempting to infer from these data that similar age-related changes occur in human corneas; if true, this may explain the diminished sensitivity and increased incidence of dry eye seen in older age groups. 

 Submitted for publication October 15, 2007; revised November 8, 2007; accepted January 22, 2008.

 Disclosure: L. Dvorscak, None; C.F. Marfurt, 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 illustration of a typical rat corneal quadrant showing the locations of the standardized central and peripheral zones analyzed in this study.

Figure 2.

 

Subbasal nerves (a, c) and their terminal branches (b, d) from a representative 6-month-old animal. Through-focus series through the central (a, b) and peripheral (c, d) corneal zones of the same corneal quadrant.

Figure 3.

 

Effect of aging on nerve terminal density in the central (•) and peripheral (▪) corneal epithelium. *P < 0.05; **P < 0.01.

Figure 4.

 

Line drawings of corneal nerve terminals and subbasal nerves in the peripheral 1-mm² corneal zones of representative 6- and 24-month-old animals.

Figure 5.

 

Line drawings of corneal nerve terminals and subbasal nerves in the central 1-mm² corneal zones of representative 6- and 24-month-old animals.

Figure 6.

 

Effect of aging on subbasal nerve density in the central (•) and peripheral (▪) corneal epithelium. **P < 0.01.

Figure 7.

 

Densely packed subbasal nerves in the central corneal zone of an 18-month-old animal. Calibration bar, 100 μm.

Figure 8.

 

Histogram comparing the distribution of subbasal nerve fiber diameters in 6- and 24-month-old animals. Note the shift toward smaller nerve diameters in the older age group.

Figure 9.

 

Line drawings of subbasal nerve vortices. The vortices illustrated here represent the most prominent examples from the right eyes in each age group. Most vortices observed in this study were ambiguous or considerably less well developed than were those shown here. Note that two of the vortices (12 and 18 months) spiral clockwise, whereas the other two (6 and 24 months) spiral counterclockwise. Circle diameter, 0.75 mm.

The authors thank Sylvia Deek (Valparaiso University) for technical assistance and the National Institute of Aging for permission to purchase animals from their subsidized colonies.