Abstract
Purpose:
To examine the influence of short-term reduction in cerebrospinal fluid pressure (CSFP) as compared with short-term elevation in IOP on axonal transport.
Methods:
The study included 111 adult Sprague-Dawley rats. For 6 hours, IOP was unilaterally elevated to 40 mm Hg (IOP40-group; n = 27), IOP was unilaterally increased to a value of 25 mm Hg below the mean blood pressure (PP25-group; n = 27), or CSFP was reduced by continuous aspiration of cerebrospinal fluid (Low-CSFP-group; n = 27). A sham control group (with a trocar in cisterna magna without cerebrospinal fluid release) included 24 rats. The left eyes of the IOP40 study group and PP25 study group served as additional contralateral control group. Orthograde axonal transport was examined by intravitreally injected rhodamine-β-isothiocyanate; retrograde axoplasmic flow was assessed by fluorogold injected into the superior colliculi.
Results:
At 24 hours after baseline, rhodamine-β-isothiocyanate (RITC) staining intensity of the optic nerve was lower (P < 0.05) in the IOP40-group, PP25-group, and Low-CSFP-group than in the control groups. At 6 hours after the fluorogold injection, fluorogold fluorescence was significantly lower in the IOP40-group, the PP25-group, and the Low-CSFP-group than in the control groups. At 5 days after baseline, the fluorogold fluorescence no longer differed significantly between the IOP40-group or the Low-CSFP-group and the control groups. At 1 week after baseline, retinal ganglion cell density was markedly reduced only in the PP25-group.
Conclusions:
Both short-term lowering of CSFP and short-term rise in IOP were associated with a disturbance of both the orthograde and retrograde axonal transport. The findings support the notion of an association between abnormally low CSFP and optic nerve damage.
Glaucoma, a leading cause of irreversible vision loss, is characterized by loss of retinal ganglion cells and their axons. Recent clinical and anatomical investigations have suggested that a low orbital cerebrospinal fluid pressure (CSFP), parallel to an elevated IOP, may have a role in the pathogenesis of glaucomatous optic neuropathy.
1–9 The results obtained in these investigations were supported by the findings of a recent experimental study conducted on monkeys with an artificially reduced CSFP and which developed a damage of the optic nerve.
10 However, the mechanism underlying the optic nerve damage in these monkeys with an experimental and long-standing lowering of CSFP has not fully been elucidated yet.
An impairment of the orthograde axonal transport and retrograde axonal transport has been considered an important pathogenic feature in glaucomatous optic neuropathy.
11–17 Impediment of the axonal transport in the retinal ganglion cell axons compromises the viability of the retinal ganglion cells by reducing the supply of neurotrophic factors. The bottleneck in the visual afference also with respect to the orthograde and retrograde axoplasmic flow is the lamina cribrosa of the optic nerve head with its translamina pressure difference (TLPD) between the intraocular compartment (i.e., the IOP) and the compartment behind the lamina cribrosa (i.e., the retrolamina optic nerve tissue pressure and the orbital CSFP). Previous studies showed that an abnormal TLPD due to an experimentally elevated IOP was associated with an impediment of both, the orthograde and the retrograde, axoplasmic flow in animals. By including the CSFP as one of the two determinants of TLPD into the discussion on the pathogenesis of glaucomatous optic neuropathy, we examined in this study whether an abnormally low CSFP is associated with abnormalities of the axoplasmic flow, similar to the situation with an elevated IOP. We assessed the orthograde axoplasmic flow by examining the distribution rhodamine-β-isothiocyanate (RITC) within optic nerve axons after injection of RITC into the vitreous, and we assessed the retrograde axoplasmic flow by examining the distribution fluorogold within optic nerve axons after injection of fluorogold into the superior colliculi of rats with either an abnormally low CSFP or with an elevated IOP.
The study included 8-week-old adult male Sprague-Dawley rats with a body weight of 200 to 220 g at baseline before the experiments were started. The animals were kept in temperature-controlled rooms with a 12-hour light/dark cycle and were provided with standard food and water ad libitum. The study was approved and monitored by the Institutional Animal Care and Use Committee of the Capital Medical University of Beijing. All experiments were performed in accordance with the ARVO Statement for Use of Animals in Ophthalmic and Vision Research.
The animals were divided into groups. For the low-CSFP-group of animals, the rats were anesthetized by intraperitoneal chloral hydrate (400 mg/kg; Sigma-Aldrich Corp., St. Louis, MO, USA). The head of the animal was positioned on a stereotactic guide instrument and a midline scalp incision was made. The nuchal muscles were cut in the midline and stripped laterally to expose the transparent dura mater. The CSFP was measured using the BIOPAC Systems MP150 workstation (BIOPAC Systems Co., Gleeta, CA, USA). A cannulation hole, 0.8-mm caudal and 1.5-mm lateral to the bregma, was made with a dental drill. A 1.6F Pressure Catheter (4.8-mm tube length below the skull; Scisense, Inc., London, Ontario, Canada) was inserted vertically deep into the brain parenchyma. Using the BIOPAC Systems MP150 workstation, the CSFP measured in millimeters per mercury was continuously monitored. We then exposed the dura mater overlying the cerebellum with the dorsal surface of medulla oblongata. Using an operation microscope (Leica AG, Heerbrugg, Switzerland), a taper glass capillary tube was inserted into the cisterna magna. For the rats of the Low-CSFP-group, cerebrospinal fluid was aspirated every 15 minutes over a study period of 6 hours. For the rats of the sham control group, a trocar was positioned into the cisterna magna without removing the guide wire. The trocar was formed from polyethylene-50 tubing (Becton Dickinson, Franklin Lakes, NJ, USA) and was threaded with a 30-G stainless steel wire (Ted Pella, Inc., Redding, CA, USA).
The rats of the IOP40-group were anesthetized with an intraperitoneal injection of chloral hydrate (400 mg/kg) followed by baseline IOP measurements using a Tonopen XL (Mentor; Norwell, MA, USA). After inducing mydriasis with eye drops containing tropicamide and phenylephrine (Santen Pharmaceutical Co. Ltd., Osaka, Japan), the anterior chamber of the right eye was cannulated with a 30-G needle connected to a saline-filled bottle the height of which could be adapted. The IOP was adjusted to a level of 40 mm Hg by altering the height of the bottle for the whole study period of 6 hours.
In the rats of the reduced ocular perfusion study group (PP25-group), a cannula (PE-50; Becton Dickinson) connected to a pressure transducer was placed into the femoral artery to monitor the blood pressure. The blood pressure was measured every 15 minutes, and the IOP was maintained at a level of 25 mm Hg below the mean blood pressure in order to achieve an ocular perfusion pressure of 25 mm Hg for the whole study period of 6 hours. Ocular perfusion pressure was defined as the mean blood pressure minus the IOP. The retinal blood flow was checked by ophthalmoscopy and appeared similar to the retinal blood in eyes with normal IOP.
Only animals with unimpaired lenses and corneas after surgery were included into the study.
The anterograde axonal transport within the retinal ganglion cell axons was assessed using the tracer RITC dissolved in sterile PBS. At the same setting at which the IOP or the CSFP were modified, the pupil was dilated, and the retina and optic disc were observed through the operation microscope. A 28-G needle was introduced through the temporal sclera into the vitreous cavity, and 2 μL of freshly prepared 3% RITC was injected into the vitreous of each eye. At 24 hours after the intravitreal RITC injection, the rats were deeply anesthetized and transcardially perfused with 0.01 M PBS (pH 7.4), followed by 4% paraformaldehyde. The eyes were enucleated with the optic nerve kept as long as possible connected to the ocular globe. The eyes were postfixed in 4% paraformaldehyde for 1 hour and the anterior segments were removed. The dura was resected from the optic nerve which was then mounted in OCT (Sakura Finetek USA, Inc., Torrance, CA, USA) and longitudinally sectioned into 10-μm thick specimens using a cryotome set at −22°C. To avoid a tilting of the sections, we aligned the optic nerve parallel to the blade of the cryostat. In all rats, the optic nerves from both eyes were identically orientated prior to cryosectioning to allow a comparison between both optic nerves. Longitudinal sections were performed along the horizontal plane beginning in the superior portion of each optic nerve and proceeding to the inferior portion of the nerve. For the study of the axonal transport, only the central region of each optic nerve was assessed. All specimens were stored in the dark at −80°C immediately after sectioning. To show the intact optic nerve morphology, we counterstained the sections with III β-tubulin. Standard immunohistochemical procedures were performed with primary rabbit antibodies against III β-tubulin (TUJI, 1:250, rabbit; Cell Signaling Technology, Danvers, MA, USA), and FITC-conjugated anti-rabbit secondary antibody (1:500, goat; Cell Signaling Technology). They were examined after being mounted in fluoroshield (Sigma-Aldrich Corp.).
Digital images of the labeled tissues were taken with a confocal fluorescence imaging microscope (Leica TCS-SP5, DM6000-CFS; Leica AG) at ×10 magnification. Visualization of the RITC-labeled tissue was achieved by laser excitation at a wavelength of 561 nm of a green helium neon laser with emissions detected through a 575- to 616-nm band pass filter. Images of the eyes of the control groups and study groups were acquired at the same setting under the same conditions and using the same instruments to allow a comparison between both groups. A series of z-stacked images were captured of each slide, beginning in the inner limiting membrane and extending posteriorly into optic nerve tissue. Each z-stack consisted of a depth of optical sections collected at 2-μm increments along the z-plane, and the seven sequential images were averaged.
The methodology for quantifying orthograde axonal transport change was similar as performed in previous reports by Balaratnasingam and collegues.
18–20 Assessment of all images was performed on Image Pro Plus (Version 6.0; Media Cybernetics, Rockville, MD, USA). Six nerve bundles in each eye were analyzed. In a sample window of constant size, the average pixel intensity of RITC per 1 μm
2 was determined at specific points along each of the nerve bundles. Measurements were performed at a line connecting both ends of the retinal pigment epithelium–Bruch's membrane complex at the optic nerve head, immediately posterior to the inner limiting membrane, and at points along the optic nerve at distances of 50, 100, 150, 200, 250, 300, 350, 400, and 450 μm behind the retinal pigment epithelium–Bruch's membrane complex line. The measurements taken at all points posterior to the retinal pigment epithelium–Bruch's membrane complex line were expressed as percentage of the intensity measured at the location immediately posterior to the inner limiting membrane.
To determine observer reproducibility, two axonal transport images were quantified on three separate occasions, each at least 1 week apart, by the same masked observer who performed all the data analyses. Six nerve bundles from each axonal transport image were also quantified as described previously, with all points expressed as a percentage intensity of the location immediately posterior to the inner limiting membrane. The images that were used for reproducibility analysis of axonal transport data consisted of the normal- and high-IOP images from the same animal. A coefficient of variation (CV) was calculated for each axonal transport data.
To assess the retrograde axonal transport, the retinal ganglion cell axons were retrogradely labelled by injecting fluorogold into both superior colliculi at the same setting at which the IOP or the CSFP was elevated or reduced, respectively. The rats were deeply anesthetized with an intraperitoneal injection of 10% chloral hydrate in distilled water (4 mL/kg body weight) and positioned in a stereotactic apparatus. The skin over the cranium was incised and the scalp exposed. Holes approximately 2 mm in diameter were drilled into the skull on both sides of the midline, 6.2-mm posterior to the bregma and 1.5-mm lateral to the midline, using a dentist's drill (Dremel, Racine, WI, USA). Using a 10-μL Hamilton syringe (Reno, NV, USA) with a 28-gauge needle, a total volume of 4.0 μL (2.0 μL per hole) of the neurotracer dye hydroxystilbamidine (3% in saline, equivalent to FluoroGold; Biotium, Inc., Hayward, CA, USA) was injected into the superior colliculus on both sides. The volume injected into each hole was divided into two equal parts injected at two different levels, 4.0 and 4.2 mm below the pia mater. After each injection the needle was left in place for 3 minutes to avoid reflux of the solution. The needle was then slowly withdrawn, and the skin was resutured.
At 6 hours after baseline (IOP40-study-group: n = 6 animals; PP25-study group: n = 6; Low-CSFP-study-group: n = 6; control group: n = 12; sham control group: n = 6), at 1 day after baseline (IOP40-study-group: n = 6 animals; PP25-study group: n = 6; Low-CSFP-study-group: n = 6; control group: n = 12; sham control group: n = 6), and at 5 days after baseline (IOP40-study-group: n = 6 animals; PP25-study group: n = 6; Low-CSFP-study-group: n = 6; control group: n = 12; sham control group: n = 6), deeply anesthetized rats were transcardially perfused with 37°C 0.01 M PBS (pH 7.4), followed by 4% (wt/vol) paraformaldehyde. To explant the retina, the eye was cut open circularly behind the ciliary body to separate cornea and lens from the posterior portion of the eyeball. The retina was detached from the pigment epithelium and fully separated from the sclera by transection of the proximal optic nerve. The retinas were fixed in fresh 4% paraformaldehyde in PBS for 30 minutes and washed three times in PBS for 5 minutes each. The free-floating retinas were then flat mounted onto glass slides.
Retinal ganglion cells were counted in a masked manner at a magnification of ×200 through a fluorescence microscope (DM 400B; Leica, Wetzlar, Germany). Visualization of fluorogold-labeled ganglion cells was achieved by the following conditions: gain = 2.3×, saturation = 0.45, gamma = 2.76, exposure time = 510.6 ms for the animals killed at 6 hours after baseline; 212.2 ms for the animals killed at 1 day after baseline; 65.8 ms for the tissue obtained at 5 days after baseline. Each retina was divided into a superior, inferior, nasal, and a temporal quadrant; for each quadrant, three fields with a size of 800 × 600 μm were taken along the median line of each quadrant from the optic disc to the peripheral border of the retina at 1-mm intervals and were counted in a double-blind manner. The number of labeled cells was divided by the area of the region.
One week after baseline, 15 rats (IOP40-study-group: n = 3 animals; PP25-study group: n = 3; Low-CSFP-study-group: n = 3; control group: n = 6) were deeply anesthetized by an intraperitoneal injection of chloral hydrate and transcardially perfused with 0.01 M PBS (pH 7.4, 37°C), followed by perfusion with 4% paraformaldehyde. The globes and optic nerves were enucleated and completely immersed in 4% paraformaldehyde overnight at 4°C. The globes were dehydrated in a series of ascending ethanol concentrations, cleared in xylene, and embedded in paraffin. Tissue embedded paraffin blocks were sectioned with a thickness of 7 μm and mounted on glass slides. Sections were deparaffinized in xylene, rehydrated in descending series of ethanol concentrations, and rinsed three times in PBS. Endogenous peroxidase activity was quenched by immersion in 3% hydrogen peroxide for 10 minutes. Nonspecific binding was blocked with 10% normal goat serum for 15 minutes prior to incubation with primary rabbit antibodies against III β-tubulin (TUJI, 1:250, rabbit; Cell Signaling Technology) overnight at 4°C. Following washing out of the first antibody, sections were incubated with a biotinylated goat anti-rabbit secondary antibody (ZSGB-BIO, Beijing, China) at a concentration of 1:250. Slides were washed and incubated with the avidin-biotin-peroxidase complex (ZSGB-BIO) for another 15 minutes. Positive signals were visualized using diaminobenzidine tetrahydrochloride (DAB). The sections were stained with hematoxylin (SL90; Statlab, Lewisville, TX, USA), and dehydrated in a series of ascending ethanol concentrations. Finally, the retina sections were immersed in xylene and mounted in neutral balsam. A masked observer photographed ×40 fields (DM 4000B; Leica) from serial sections of treated and untreated eyes. Photographs were taken 1 to 2 mm from the optic nerve and in the periphery. Labeled RGCs were counted within the entire ×40 fields. Comparisons were made between untreated and treated eyes for both central and peripheral measurements.
The statistical analysis was performed using a commercially available software program (SPSS for Windows, version 21.0; IBM-SPSS, Chicago, IL, USA). All results were expressed as mean ± SD. The Kolmogorov-Smirnov test was applied to determine the normal distribution of data. Repeated-measures ANOVAs with post hoc tests were performed to assess the effects of pressure on the proportional change in RITC intensity along the optic nerve for the anterograde axonal transport assay, and to compare the relative retinal ganglion cell density between the high IOP groups and the control group, the low CSFP group and the sham group for the retrograde axonal transport assay. Normally distributed parameters were analyzed using ANOVA with Bonferroni-corrected post hoc tests. Nonparametric parameters were analyzed using ANOVA on ranks with the Tukey test employed for post hoc paired analysis. The Student t-test for two independent samples was applied to compare TUJI cell densities between experimental group and control group. A P value less than 0.05 was considered to be statistically significant.
The intensity of RITC staining of the optic nerve, measured at distances of 50-, 100-, 150-, 200-, 250-, and 450-μm posterior to the Bruch's membrane opening level, was significantly (P < 0.05) lower in the IOP40-group, the PP25-group, and the Low-CSFP-group than in the control group and sham-control group at 24 hours after baseline. The IOP40-group, the PP25-group and the Low-CSFP-group did not differ from each other nor did the control group and the sham-control group. At 6 hours after the fluorogold injection, fluorogold fluorescence in the retina was significantly lower in the IOP40-group, the PP25-group, and the Low-CSFP-group than in the control groups, with no significant differences between the IOP40-group and the Low-CSFP-group nor between the two control groups. The retinal fluorogold fluorescence was lowest in the PP25-group. At 24 hours after baseline, fluorogold fluorescence no longer differed significantly between the IOP40-group and the control group, and at 5 days after baseline, the Low-CSFP-group and the sham control group no longer differed, however fluorogold fluorescence was still significantly and markedly reduced in the PP25-group. Our experimental study on animals with either elevated IOP or with lowered CSFP, thus revealed that in both conditions, both the orthograde axoplasmic flow and the retrograde axonal transport in the retinal ganglion cell axons was impeded; and that the recovery of the retrograde axonal transport after the short-term CSFP-reduction was slower than the recovery of the retrograde axonal transport after the acute IOP elevation, although the translamina pressure difference was higher in the high-IOP group than in the low-CSFP group.
The results of our study confirm previous investigations on an impediment of the orthograde axoplasmic flow and retrograde axoplasmic flow in animals with an acute elevation of IOP. In several studies
15,21–23 radioactive leucine as a tracer was used, Balaratnasingam and colleagues
18–20 applied RITC as marker, and Abbott and coworkers
17 used cholera toxin B-subunit (CTB) as tracer and found that elevation of IOP interrupted the orthograde axonal transport in primates, pigs, rabbits, and rats. These studies revealed a partial reduction in the orthograde axoplasmic flow at moderate IOP levels, while a complete blockade of the axoplasmic flow was observed only at nonphysiologically high levels of IOP within a range of 25 mm Hg of the mean blood pressure (corresponding to an IOP of ∼75 mm Hg in humans). It agrees with the results of our study in which the impediment of the orthograde and retrograde axoplasmic flow was most marked in the PP25-group. The retrograde axoplasmic flow was traced by Minckler and coworkers
15 using horseradish peroxidase and by Pease and colleagues
13 applying brain-derived neurotrophic factor (
125I-BDNF) showing that elevation of IOP interrupted the retrograde axonal transport in primates and rats. In the study by Pease and coworkers,
13 an elevation of IOP to 50 mm Hg for 6 hours was associated with a reduction in the retrograde axoplasmic transport by 74%, and an IOP elevation to 75 mm Hg for 6 hours reduced the retrograde axonal transport by 83%.
13 Similar results were obtained in our study.
The findings of our study provide new information with respect to changes in the axonal transport in retinal ganglion cell axons of rats with an acute experimental reduction in CSFP. Recent clinical studies showed that some patients with normal (IOP)-pressure glaucoma had an abnormally low CSFP, as determined by lumbar puncture,
3,6,7 or as assessed by measuring the orbital cerebrospinal fluid space width as a surrogate for orbital CSFP,
9 or as indirectly assessed by using an algorithm for the estimation of the CSFP.
24–26 In addition, a recent experimental study on monkeys with an experimental chronic lowering of the CSFP by implanting a lumboperitoneal shunt revealed that the monkeys with low CSFP showed a loss of the retinal nerve fiber layer thickness and width of the neuroretinal rim. In that study it had remained unclear, whether the CSFP reduction associated damage to the optic nerve was glaucomatous or whether it was an unspecific optic nerve damage.
10 The results of our study showed a reduction in CSFP led to similar changes in axonal transport as did an elevation of IOP, which suggests that some aspects in the mechanism of optic nerve damage may be shared in animals with (short-term) high IOP and animals with (short-term) low CSFP.
At 5 days after baseline, the density of fluorogold-labelled retinal ganglion cells cell density no longer differed between the low-CSFP study group and the control group in our study. It raises the questions whether the experimental short-term reduction in CSFP only temporarily impeded the axoplasmic flow without leading to retinal ganglion cell loss. The results of our study can therefore not be taken as proof that low CSFP is associated with the pathogenesis of glaucomatous optic neuropathy. One may perhaps consider however, that repeated lowering or chronic lowering of CSFP might have led not only to a temporary reduction in the orthograde and retrograde axoplasmic flow but additionally to ganglion cell damage and loss. In that context, previous investigations may be of interest. A recent study by Joos and colleagues
27 demonstrated that an IOP elevation to 50 mm Hg for a duration of 8 hours did not cause permanent damage to the retinal ganglion cells, in contrast to repeated IOP elevations to 35 mm Hg daily for a duration of 1 hour. Other studies showed a deformation of the optic nerve head and peripapillary tissues,
28–30 retinal ganglion cell specific functional abnormalities,
31–34 and impaired axonal transport
15,18,19,21,22,35–38 in animals with brief periods (usually <8 hours) of an IOP elevation to sub-ischemic levels (≤50 mm Hg). These alterations however were completely reversible shortly after normalization of IOP.
15,22,30,32–34,37 One may also take into account that with a normal CSFP of 6 to 11 mm Hg in rats,
39,40 the eyes of the low CSFP group as compared with the eyes of the high IOP group experienced a markedly lower change in the translamina pressure difference. One may infer that more time may thus be needed in the low-CSFP model than in the high-IOP model to lead to optic nerve damage.
The recovery of retrograde transport from short-term CSFP reduction took more time than did the recovery of the retrograde transport from short-term IOP elevation to 40 mm Hg, although the high-IOP-study-group experienced a higher translamina pressure difference. It may suggest that additional other mechanism may be involved in the disruption of axonal transport in the low-CSFP model. Unlike all other cranial nerves, the optic nerve is surrounded by meninges and imbedded into cerebrospinal fluid throughout its entire extracerebral path. Anatomically, the subarachnoid space of the optic nerve resembles a cul-de-sac that has the potential for decreased turnover of cerebrospinal fluid.
41 The reduction in CSFP may potentially further reduce the turnover of the orbital cerebrospinal fluid, which, theoretically, may cause an accumulation of waste material and may eventually lead to a partial failure of the axonal transport. Another possible mechanism may be a meningoepithelial cells induced inflammatory response. The apices of the meningoepithelial cells that line the arachnoid layer face the subarachnoid space. These cells are known to be highly reactive to various stimuli including the change of CSFP, and proliferation of MECs was demonstrated in a postmortem study in glaucoma patients.
42 The activated meningoepithelial cells may produce some biologically active substances, such as prostaglandin-D-synthase (L-PGDS),
43 which may damage optic nerve function.
Potential limitations of the current study should be mentioned. First, due to marked differences in the anatomy of the optic nerve head between rats and humans with a markedly thinner lamina cribrosa in rats and a profound difference in the posture of the head and body, any findings obtained in a rat model cannot fully be transferred on the situations in humans. Second, the model of an experimental acute elevation of IOP does not reflect the situation of chronic open-angle glaucoma. Third, the model of an acute reduction in CSFP is not a correct surrogate for a chronic reduction or intermittently low CSFP in patients. Fourth, the reduction of CSFP lasted only 6 hours, and it remained unclear whether there will be an apoptotic cell loss in the retinal ganglion cell layer if we establish a rat model of longer-duration CSFP reduction or with repeated insults.
In conclusion, both short-term lowering of CSFP and short-term rise in IOP were associated with a disturbance of both, the orthograde and retrograde axonal transport. The results suggest that an experimental model with an acute reduction in CSFP as well as an experimental model with an acute rise in IOP may share similarities with respect to the optic nerve.