A Novel and Reversible Experimental Primate Ocular Hypertension Model: Blocking Schlemm’s Canal

Glaucoma is a progressive optic neuropathy characterized by the degeneration of retinal ganglion cells (RGCs) that can lead to irreversible blindness. It is a leading cause of vision loss worldwide and is estimated to affect 76 million people in 2020 [1]. Elevated intraocular pressure (IOP) is a major risk factor for glaucomatous optic nerve damage, and reducing IOP is currently the only effective therapy for glaucoma [2, 3]. However, glaucomatous optic neuropathy can develop or progress in glaucoma eyes with a normal IOP or controlled IOP [4, 5]. The exact mechanism of glaucoma damage is still unknown, and other factors are believed to be involved in the onset and progression of glaucomatous neuropathy, such as ocular pulse amplitude, corneal thickness, retinal vessel diameters, and choroidal blood flow [6‒8]. Current treatments are not adequate to prevent vision loss in 30–50% of glaucoma patients [9‒11]. Understanding the natural course and pathogenesis of this disease with a well-controlled IOP is urgently needed.

Studying animal models is important to reveal pathological mechanisms and develop more effective therapy for many eye diseases including glaucoma. A physiological basis for the maintenance of IOP is that aqueous humour exits the eye primarily through a trabecular meshwork (TM), enters Schlemm’s canal (SC) into an intrascleral venous plexus, and finally flows to aqueous and episcleral veins [12, 13]. Elevated IOP is derived primarily from an increased resistance to aqueous humour outflow through the trabecular meshwork, SC, and the juxtacanalicular connective tissue [14‒16].

Various methods have been used to increase the IOP in several species of animals. Cautery of episcleral vessels [17, 18], episcleral vein sclerosis with hypertonic saline [19], injection of microbeads into the anterior chamber [20], and laser scarification of the trabecular meshwork [21, 22] have been conducted in primates [23], rabbits [24], rats [25], and dogs [26]. The methods of inducible ocular hypertension can be categorized into three types of models: pretrabecular, trabecular, and post-trabecular. Intracameral injections of microbeads are the most used pretrabecular model [27, 28]. The trabecular model is represented by laser photocoagulation. SC is an important passageway of post-trabecular aqueous humour outflow, and a complex of the inner wall of SC and juxtacanalicular connective tissue is responsible for generating the major outflow resistance in the eye [29]. However, the IOP of these models is not very predictable and can only be maintained for a relatively short term. In the present study, we established a novel method to obtain sustained IOP elevation in primate eyes by inserting a microcatheter into SC and to reverse the elevation of IOP by removing the microcatheter.

This study complied with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. Approval has been given by the Animal Care and Ethics Committee at Wenzhou Medical University (Wenzhou, China).

All experimental procedures were monitored by the Institutional Animal Care and Use Committee of JOINN Laboratories in Suzhou, which is fully accredited by the Association for Assessment and Accreditation of Laboratory Animal Care International. All experiments were performed in accordance with standards published by the Institute of Laboratory Animal Resources, Commission on Life Sciences, National Research Council; National Academy Press, Washington, DC, 2010 and Public Law 99–198. Monkeys (Macaca fascicularis) were bred and housed individually in stainless steel cages and provided solid food and water ad libitum. The room temperature ranged from 18 to 21°C with 40–70% humidity and a 12-h light-dark cycle. Five cynomolgus monkeys (cynomolgus macaques; body weight, 5–7 kg; age, 4–6 years; 3 males and 2 females) were used in this study. The eyes of the animals showed no abnormalities as determined by a comprehensive ophthalmic examination, including slit-lamp microscopy (YZ5T, 66 VISION TECH Co., Ltd. Jiangsu, China), fundus photography (Kanghua APS-6M, Kanghua Inc., Chongqing, China), optical coherence tomography (Spectralis OCT; Heidelberg Engineering, GmbH, Heidelberg, Germany, software version 1.0.15.0), and rebound tonometer (TonoLab; Tiolat, Oy, Finland) under anaesthesia.

All measurements were obtained from the monkeys under anaesthesia by subcutaneous injection of 0.05 mg/kg atropine sulphate (50 mg/mL, Henan Runhong Pharmaceutical Co. Ltd, China) and intramuscular injection of 30 mg/kg tiletamine plus zolazepam (Zoletil 50, 50 mg/mL Virbac, Carros, France). Topical oxybuprocaine hydrochloride (Benoxil, 0.4%, Santen Pharmaceutical, Osaka, Japan) was used for all procedures involving contact with the cornea. Dilation of the pupils for fundus photography and optical coherence tomography was achieved with 0.5% tropicamide phenylephrine (Santen Pharmaceutical).

SC was blocked in four monkeys (monkeys A, B, C, and D), which comprised the operation group. One monkey (monkey E) was recruited for the sham operation. In each monkey of the operation group, a 200-μm-diameter microcatheter (iTrack-250A, iScience Interventional, Inc., Menlo Park, CA, USA) was inserted into the SC for 360°. The microcatheter has a 200-μm-diameter shaft with an atraumatic distal tip of 250 μm in diameter and incorporates an optical fibre to provide an illuminated beacon tip that assists in guidance. The protocol is a modified procedure based on canaloplasty [30]. The left eyes were used as an untreated control.

Before the microcatheter insertion, topical anaesthesia was applied by instilling topical oxybuprocaine hydrochloride (Benoxil, 0.4%) three times (with a 1-min interval). To begin the surgery, a fornix-based conjunctival incision was created to allow for a 4 × 4 mm superficial scleral flap (about one-third of the scleral thickness) followed by a 3 × 3 mm deeper inner scleral flap (Fig. 1a, b). The deep flap was fashioned to expose and unroof the SC (Fig. 1c). The microcatheter (iTrack-250A, iScience Interventional, Inc.) was then inserted and moved forward within the SC for the entire 360° (Fig. 1d–f). All the above procedures were performed in accordance with the canaloplasty procedure [30], with the exception of the microcatheter, which was cut off and kept in SC to block the trabeculocanalicular outflow system (Fig. 1g). The scleral flap was tightly sutured with 10-0 Vicryl sutures (Fig. 1h). The conjunctiva was then sutured to the limbus. The monkey of the sham operation underwent the same surgical procedure except that a microcatheter was not inserted into SC.

After the operations, the experimental eyes of all the monkeys were treated with topical levofloxacin (Cravit, 0.5%, Santen Pharmaceutical Co., Ltd) three times a day and topical tobramycin-dexamethasone (TobraDex®, Alcon, CUSI, s.a.) once a day for 1 week. To attempt a reversible ocular hypertension model, when the elevated IOP of the blocked eyes was stable, monkey A was randomly selected for removal of the microcatheter at 6 weeks (Fig. 1i).

IOP was monitored throughout the experiment in both eyes with the rebound tonometer (TonoLab; Tiolat) under topical anaesthesia with topical oxybuprocaine hydrochloride and general anaesthesia as mentioned above. The IOPs were recorded for 14 consecutive weeks, which started 1 day before and ended at 11 months after the microcatheter insertion. IOP (mm Hg) was determined bilaterally ten times and averaged for each time point between 9:00 a.m. and 10:00 a.m.

The animals were anaesthetized as described, and the pupils were dilated with 0.5% tropicamide phenylephrine. The fundus of both eyes was photographed using an ocular fundus camera (APS-6M, Kanghua Inc.). The inspection area needed to be covered with the full disc and its surrounding structure. Fundus photos of the optic nerve head were taken biweekly.

The peripapillary RNFL thicknesses were measured on both the treated and untreated eyes using the Spectralis OCT. The setting parameters for examining the peripapillary RNFL thickness in this study included a 30° × 20° scan angle consisting of 25 sections along a 12.0° (3.5 mm) circle diameter with high speed resolution modes. SD-OCT images of the optic nerve head were obtained biweekly.

At the end of the study, the animals were euthanized by over-anaesthesia. Eyes with attached optic nerves were removed and processed for histopathology, including fixation in 10% paraformaldehyde, dehydration in alcohol, paraffin embedding, and haematoxylin and eosin (HE) staining using standard techniques. The slice thickness was set at 3 μm, and five consecutive sections were selected from the blocked eyes and contralateral eyes in which the optic nerve head area could be found under a microscope [31, 32]. HE staining images were captured using ZEN 2012 software (x64 blue v.1.1.2.0; Carl Zeiss Microscopy GmbH) and ImageJ (National Institutes of Health, Bethesda, MD, USA) was used to perform quantitative analyses of RGC numbers. The number of cells in the RGC layer of each sample was counted in ten high powered fields (HPF, 400×) per eye. For the evaluation of the effect of the obstruction and the damage of the outflow system after the microcatheter insertion, the anterior segments of the blocked and non-blocked eyes were captured for microscopic analysis.

Statistical analyses were performed using standard statistical software (SPSS v.21.0 for Windows; SPSS, Chicago, IL, USA). Comparisons between the IOP of the eyes with and without SC blocking were based on the Mann-Whitney test. Wilcoxon signed-rank test was used to compare the baseline IOP of the right and left eyes. The one-sample T test was used to examine the changes of the IOP in monkey A before and after the microcatheter removal. A paired, independent sample t-test was used to compare the number of cells in the RGC layer between the right and left eyes. All tests were two-sided, and a p value of <0.05 was considered significant.

The surgical procedures performed on all the monkeys were successful and resulted in no severe inflammation or complications (Fig. 2). Four monkeys (A, B, C, and D) had their SC blocked in the right eye. One monkey (monkey E) underwent a sham operation in the right eye. Figure 3 shows representative samples of the anterior segment from the blocked and non-blocked eyes. Histological analysis shows that the SC of the microcatheter-inserted eyes was completely blocked. Additionally, severe damage in the aqueous outflow system was not observed.

The mean baseline IOP for all the monkeys was 11.8 ± 1.5 mm Hg in the right eyes and 12.7 ± 0.5 mm Hg in the left eyes (p = 0.084). After SC blocking, the IOP of the four blocked eyes (right eyes) was markedly elevated and followed by a reduction after reaching a peak IOP in 2–3 weeks; it remained stable at 1 month (Fig. 4). The peak IOP ranged from 37.0 to 60.0 mm Hg (mean ± SD, 48.8 ± 9.4 mm Hg). Three of the four (75%) blocked eyes had an IOP increase of more than 30% from baseline. After the sham operation, the mean IOP of the experimental eyes in monkey E was 13.7 ± 1.9 mm Hg, which was similar to that before the operation (12.5 ± 0.7 mm Hg).

At 6 weeks after SC blocking, the mean IOP in the experimental eye of monkey A was significantly higher than the baseline IOP (24.1 ± 13.1 mm Hg vs. 11.0 ± 2.8 mm Hg, p = 0.017, Fig. 4a). Then, the microcatheter in the SC of monkey A (randomly selected) was removed. One week after the microcatheter removal, the mean IOP of the experimental eye dropped back to baseline level (13.7 ± 0.6 mm Hg, Fig. 4a). Although the IOP of the experimental eye was moderately above that of the control eye at several time points, the mean IOP of the experimental eye was similar to that of the contralateral eye from 1 week to 11 months after the microcatheter was removed (16.1 ± 4.0 mm Hg vs. 15.3 ± 2.8 mm Hg, p = 0.170). The global RNFL thickness in the experimental eye of monkey A, which had the microcatheter removed after 6 weeks, decreased from 101.7 ± 1.6 μm at baseline to 86.3 ± 2.5 μm at 6 weeks after the SC blocking (p < 0.001, Fig. 5). After the microcatheter removal, the global RNFL thickness in the experimental eye of monkey A dropped continuously to 72.7 ± 0.6 μm at 11 months. The RNFL thicknesses of the temporal-superior (TS), temporal, temporal-inferior (TI), nasal-inferior, nasal, and nasal-superior peripapillary sectors in the experimental eye of monkey A at baseline were 136.0 ± 2.7 µm, 55.0 ± 1.7 µm, 141.7 ± 4.5 µm, 136.7 ± 4.9 µm, 80.0 ± 5.0 µm, and 130.3 ± 9.1 µm versus 93.0 ± 2.0 µm, 46.0 ± 1.7 µm, 94.3 ± 4.5 µm, 93.0 ± 1.0 µm, 57.3 ± 4.0 µm, and 92.0 ± 4.6 µm at 11 months after the microcatheter insertion, respectively. There was a continual decline in RNFL thickness throughout the duration of the study (Fig. 5c). Consisting with OCT scarn results, fundus photographs also revealed a mildly localized RNFL defect in its temporal-superior and TI regions in the blocked eye in monkey A (Fig. 6a, b).

The microcatheter inserted into the SC of the blocked eyes in monkeys B, C, and D was not removed. In these monkeys, the mean IOP of the blocked eyes was stable, ranging from 18.6 ± 8.5 mm Hg to 27.4 ± 11.2 mm Hg during the entire 11-month study period (Table 1). They also remained significantly higher than the IOP of the contralateral untreated eyes (22.7 ± 5.7 mm Hg vs. 15.5 ± 1.4 mm Hg, p < 0.001, Fig. 4). There was more than a 30% increase in the mean IOP elevation in 3 eyes (3/3,100%), 3 eyes (3/3,100%), and 2 eyes (2/3,66.7%) at 3 months, 6 months, and 11 months, respectively.

In monkeys B and C, the RNFL thickness in the blocked eyes remained relatively stable over 11 months (Fig. 7a). There was no significant reduction in the RNFL thickness of the sham-operated monkey eye and the control eyes of monkeys B, C, D, and E (Fig. 7b). In monkey D, the global average RNFL thickness of the blocked eye dropped from 91.3.7 ± 1.5 μm to 84.0 ± 2.7 μm at 11 months after the SC blocking. The TI RNFL thickness of the blocked eye dropped from 150.3 ± 2.5 μm to 134.0 ± 3.6 μm at 11 months after the SC blocking. The TS RNFL thickness of the blocked eye dropped from 136.3 ± 4.5 μm to 116.0 ± 4.4 μm at 11 months after the SC blocking. There was no significant reduction in the RNFL thickness of the temporal, nasal-inferior, nasal, and nasal-superior peripapillary sectors in the blocked eye of monkey D. Fundus photographs showed a modest localised RNFL defect in the temporal-superior region of the blocked eye in monkey D (Fig. 6b).

The number of HE-stained cells in the RGC layer of monkeys A, B, and C was 9.9 ± 2.03 cells/HPF, 17.6 ± 5.76 cells/HPF, and 17.6 ± 8.02 cells/HPF in the experimental eyes and 19.1 ± 4.51 cells/HPF, 29.7 ± 7.73 cells/HPF, and 28.0 ± 8.87 cells/HPF in the control eyes, respectively. The number of cells in the RGC layers was significantly decreased in the blocked eyes compared with the control (p < 0.001 for monkeys A and B, p = 0.013 for monkey C, Fig. 8). The number of cells in the RGC layer of the experimental eyes was 29.9 ± 4.84 cells/HPF in monkey D and 29.2 ± 6.70 cells/HPF in monkey E. There was no difference between the experimental eye and the control eyes of monkey D (29.9 ± 4.84 vs. 31.3 ± 5.48 cells/HPF, p = 0.552) and monkey E (29.2 ± 6.70 vs. 29.9 ± 3.54 cells/HPF, p = 0.775).

Non-human primates (NHPs) can mimic human subjects in response to experimental procedures and are widely used in animal experiments because of their psychological and physiological similarity to humans [33‒35]. In particular, the brains and eyes of NHPs more closely parallel human anatomy than that of any other animal. Therefore, their responses in basic and clinical research help to predict responses in humans [36]. SC is the natural trabeculocanalicular outflow system of primates, and the inner wall of SC is the one of the main sources of outflow resistance [29]. In the present study, we developed and characterized a novel primate ocular hypertension model by inserting a microcatheter into SC for 360°. The method was a modified approach based on canaloplasty, which has been recently used as a surgical procedure to lower IOP in primary open-angle glaucoma [37]. Blocking SC with a microcatheter successfully induced persistent elevation of IOP for about 11 months in primate eyes and without any severe complications, such as secondary neovascularization or ischaemia of the cornea or choroid. Different degrees of glaucomatous optic neuropathy were confirmed by fundus photographs, SD-OCT, and HE staining in the experimental eyes of all monkeys that underwent the SC-blocking operation.

A variety of procedures have been used to produce primate ocular hypertension models. Gasasterland and Kupfer introduced a sustained IOP elevation in primate eyes by treating the trabecular meshwork with an argon laser application [23]. Quigley and Hohman also produced chronic IOP elevations in monkey eyes by treating the trabecular meshwork with a laser, which resulted in the mean IOP to rise above 35 mm Hg from 2 weeks to 11 months [21]. Quigley and Addicks introduced another ocular hypertension primate model by an anterior chamber injection of autologous fixed red blood cells [38]. This method resulted in acute elevation of IOP (24–73 mm Hg) in a short period of time (2–42 days). Weber et al. [20] injected sterile latex microspheres into the anterior chamber of the primate eye to induce chronic elevation of IOP, and the mean IOP increased to 28.1 ± 0.7 mm Hg between 30 weeks and 36 months.

In the present study, the IOP of the blocked eyes was markedly elevated after the SC blocking procedure and was followed by a reduction after reaching a peak IOP in 2–3 weeks; the mean IOP was maintained at 18.6–27.4 mm Hg for 11 months. HE staining showed that the SC of the microcatheter-inserted eyes was completely blocked and that the microcatheter, which was inserted into the SC, did not cause severe damage to aqueous outflow system. When the microcatheter of the SC was removed, the mean IOP of the experimental eye decreased to near baseline level. The sustained IOP elevations of this model may be based on SC obstruction.

In monkey A, a continuous loss of RNFL thickness was observed even after the IOP returned to the baseline level due to removal of the microcatheter. The RNFL loss was significantly faster than that happened during the period when the IOP was elevated by blocking SC. The reason behind this phenomenon may be that a rapid (within 7 days) and significant (from 50 to 25 mm Hg) decrease of IOP might result in a reverse displacement of lamina cribrosa (LC), which could break the previous balance of microenvironment at the LC reached under the status of high IOP. The details of morphology changes of LC of the study animals were discussed in another paper.

In the blocked eyes of monkeys B, C, and D, a significant elevation in IOP was seen and maintained for up to 11 months post-surgery. The optic nerve injury been observed in these three experimental monkeys based on histological analysis (monkey B and C), OCT scans and fundus photographs (monkey D). However, in histological analysis of the blocked eyes in monkey D, no significant decreased in the number of cells in the RGC layer was detected. This may be due to that the RNFL defect of the blocked eye in monkey D took place in the TS region and were not reflected in the slice taken from the central axial of the optic nerve head area. The IOP pattern of monkey D is obviously different from monkeys B and C, with a greater fluctuation of IOP. This could be due to the following reasons: firstly, there might be variations exist at the structure of SC in the blocked eye of monkey D [39]. Secondly, monkey D might have a stronger uveoscleral outflow pathway [40].

There are several advantages with this model. Firstly, we induced the ocular hypertension model using a single operation to block the SC with a microcatheter. Compared to previously reported primate glaucoma models, the success rate of IOP elevation after a single operation is relatively high. The ocular hypertension animal models induced by laser photocoagulation and microsphere injection require multiple interventions to reach and maintain a desirable IOP elevation [41‒43]. Repeated interventions may increase the risk of complications. Secondly, this technique does not introduce any foreign material into the anterior chamber of the experimental eyes with the exception of the microcatheter. This model provides a clear optical path for the evaluation of damage to the retina and optic disc; monitor the disease process of the experimental eye throughout the study, which is a major difficulty for ocular hypertension models based on the intraocular injection of red blood cells or microspheres [20, 38]. Thirdly, the elevation of IOP in this model is reversible. The SC of the experimental eye maintains the function of aqueous humour outflow after removing the inserted microcatheter from the SC (Fig. 3). It will be useful to understand the mechanism of secondary damage to the optic nerve after the IOP returns to normal.

Limitations of this study must also be acknowledged. Firstly, the IOP of the blocked eyes was markedly elevated – up to 48.8 ± 9.4 mm Hg (37.0–60.0 mm Hg) – for 1 month after the SC blocking procedure. The majority of the blood supplied to the ONH comes from the posterior ciliary arteries, and an acute elevated IOP would result in acute obliteration or temporary hypoperfusion of the small ciliary arteries, inducing ischaemic inflammatory injury with consequent RGC death [35]. Acute optic nerve ischaemia might be another factor contributing to the optic nerve injury. The procedure involves opening the conjunctiva and making a scleral flap, then blocking the SC tube. Postoperative local inflammatory healing reaction of the scleral flap and conjunctiva might be the major reason for the acute elevated IOP. Nonsteroidal anti-inflammatory drugs are widely used in ophthalmic surgery to reduce postoperative inflammation [44]. In follow-up studies, we shall attempt to prevent the acute rise in IOP by using nonsteroidal anti-inflammatory drugs. Secondly, this model was built based on the SC that similar to the structures of human. Due to the major differences in the size or structure between rodents and human/NHP of SC, this model is not yet suitable for building in rodents. Future studies are required to further investigate and refine this technique.

We present a novel method to induce experimental ocular hypertension in primate eyes. This model achieved moderate IOP elevation for a long duration by a single insertion of a microcatheter into the SC of NHP. Further studies with a longer follow-up and greater sample size are needed to confirm our findings.

The authors would like to acknowledge JOINN Laboratories in Suzhou for providing their site and facilities and Elsevier (webshop.elsevier.com) for English language editing.

This study protocol was reviewed and approved by Animal Care and Ethics Committee at Wenzhou Medical University, approval number ACU17-020. All animals were treated in accordance with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research.

The authors have no conflicts of interest to declare.

This study was supported by Grant 81770919 from the National Natural Science Foundation of China and Grant LY17H120004 from the Natural Science Foundation of Zhejiang Province.

Yuanbo Liang conceived and designed experiments. Yuanbo Liang, Jing Xu, Yan Tao, Cheng Hu, and Xinmin Sun performed the experiments. Junhong Jiang and Cong Zhangcontributed to data analysis. Shaodan Zhang and Cong Ye interpreted results. Junhong Jiang and Jing Xu wrote and modified the paper. All authors read and approved the final manuscript.

All data generated or analysed during this study are included in this article. Further enquiries can be directed to the corresponding author.

Junhong Jiang and Jing Xu contributed equally to this work and are considered joint first authors.