This method models cataract surgery in vivo by removing lens fiber cells from adult mice and leaving behind the capsular bag with attached lens epithelial cells (LECs). The injury response is then assessed at various times post-surgery using molecular and morphological criteria.
Cataract surgery (CS) is an effective treatment for cataracts, a major cause of visual disability worldwide. However, CS leads to ocular inflammation, and in the long term, it can result in posterior capsular opacification (PCO) and/or lens dislocation driven by the post-surgical overgrowth of lens epithelial cells (LECs) and their conversion to myofibroblasts and/or aberrant fiber cells. However, the molecular mechanisms by which CS results in inflammation and PCO are still obscure because most in vitro models do not recapitulate the wound healing response of LECs seen in vivo, while traditional animal models of cataract surgery, such as rabbits, do not allow the genetic manipulation of gene expression to test mechanisms. Recently, our laboratory and others have successfully used genetically modified mice to study the molecular mechanisms that drive the induction of proinflammatory signaling and LEC epithelial to mesenchymal transition, leading to new insight into PCO pathogenesis. Here, we report the established protocol for modeling cataract surgery in mice, which allows for robust transcriptional profiling of the response of LECs to lens fiber cell removal via RNAseq, the evaluation of protein expression by semi-quantitative immunofluorescence, and the use of modern mouse genetics tools to test the function of genes that are hypothesized to participate in the pathogenesis of acute sequelae like inflammation as well as the later conversion of LECs to myofibroblasts and/or aberrant lens fiber cells.
The lens is a highly organized, transparent tissue that refracts light to produce a clearly focused image onto the retina1,2,3. This specialized organ is surrounded by an uninterrupted basement membrane (the capsule), which isolates the lens from other parts of the eye. The inner anterior surface of the capsule anchors a monolayer of lens epithelial cells (LECs), which then differentiate at the lens equator into lens fiber cells, which comprise the vast majority of the lens3. A cataract occurs when the lens loses its transparency due to factors like aging, genetic mutation, UV radiation, oxidative stress, and ocular trauma4. Cataract is a leading cause of blindness worldwide, particularly in countries with poor medical care5. However, this condition is now readily treated by surgical removal of opaque lens cells by phacoemulsification in which the central lens capsule and attached lens epithelium are removed, followed by the use of a vibrating probe to break the cellular mass of the lens into smaller fragments that can be suctioned out; leaving behind a capsular bag with some attached equatorial LECs1. Vision is then most commonly restored by post-surgical implantation of an artificial intraocular lens (IOL).
While cataract surgery (CS) is a highly effective and minimally invasive treatment for cataracts, a patient's post-surgical recovery can be acutely hampered by the development of ocular inflammation5,6,8. This inflammation can cause post-surgical pain, retinal edema leading to retinal detachment, as well as exacerbation of other inflammatory and fibrotic conditions like uveitis and glaucoma4,7,8,9. Ocular inflammation post-CS (PCS) is generally treated with anti-inflammatory eye drops which are plagued by poor patient compliance or dropless cataract surgery that can lead to an increased prevalence of floaters8,10,11. In the long term, the outcomes of CS can be compromised by the development of posterior capsular opacification (PCO)12. PCO occurs when residual LECs left behind after surgery undergo a wound healing response, proliferate, and migrate onto the posterior lens capsule at months to years PCS contributing to secondary visual obstruction13,14.
Understanding the molecular mechanisms by which CS results in such acute and chronic responses represents a major challenge in the field, as much of the literature investigating PCO relies on the induction of LEC conversion to myofibroblasts in culture via treatment with active transforming growth factor beta (TGFβ)12,15. Human capsular bag models created from cadaver lenses cultured in vitro better reflect the biology of the lens PCS as LECs are cultured on their native basement membrane but are difficult to manipulate mechanistically, do not recapitulate the intraocular environment, and are inherently variable due to lens to lens (or donor to donor) variability16,17. Rabbit1,18 and non-human primate19,20 in vivo models of cataract surgery overcome some of these issues but still are not easy to manipulate for mechanistic studies. Notably, a prior study of mammalian lens regeneration found significant lens regeneration within four weeks after lens fiber cells were removed from mice, leaving the lens epithelial cells and capsule behind, while no lens regrowth occurred when the entire lens was removed21,22. We subsequently streamlined this procedure and optimized it for the study of the acute response of LECs to lens fiber cell removal and their subsequent conversion to a mixed population of cells with myofibroblast and fiber cell properties.
Using this mouse model of cataract surgery, we have shown that lens epithelial cells drastically remodel their transcriptome by 6 hours PCS to produce numerous proinflammatory cytokines23, while they begin expressing fibrotic markers as early as 24 h PCS12,23, prior to the onset of canonical TGFβ signaling. Mechanistic experiments in knockout mice using this model revealed that cellular fibronectin expression by LECs is required for a sustained fibrotic response PCS, likely due to both its role in the assembly of the fibrotic ECM and cell signaling12. Other studies showed that αVβ8-integrin is required for the transition of LECs to myofibroblasts due to its function in TGFβ activation, and the potential of this approach to identify anti-PCO therapeutic leads was confirmed as a αVβ8-integrin antagonist also blocked LEC EMT15.
Here, we provide a detailed protocol describing how to remove the lens fiber cells from living mice (Figure 1and Figure 4) while leaving behind the lens capsule and LECs to model the LEC response to cataract surgery.
The following protocol was approved by the University of Delaware Institutional Animal Care and Use Committee under protocol #1039-2021-1. In compliance with the Association for Research in Vision and Ophthalmology (ARVO) for the Use of Animals in Ophthalmic and Vision Research24, all ocular survival surgeries can only be performed in one eye. See Table of Materials for details about reagents and instruments used in this protocol. Alternative approaches for extracting the lens fibers and suturing are shown in Figure 2 and Figure 3.
1. Animals
2. Preparation of solutions and surgical tools
3. Prior to anesthesia
4. Anesthesia
5. Making the incision
6. Separating the lens fibers from the capsular bag
7. Inflating the anterior chamber and suturing the corneal incision
8. Post-surgical care
NOTE: Topical antibiotic ointment is applied to the eye, and analgesic agents are administered. Mice do not usually show signs of distress after initial surgery. If needed, slow-release version of buprenorphine could be used.
9. Monitoring animal recovery
10. Euthanasia
11. Performing further analysis on harvested lens tissues
Based on the results from this surgical method, the Duncan Lab has used this mouse model of cataract surgery to construct an injury response time course of lens epithelial cells post-surgery (Figure 5). By 6 h after surgery, 5% of the lens epithelial transcriptome is differentially expressed (Figure 6A), including the upregulation of numerous immediate early response genes and proinflammatory cytokines23 (Figure 6B). By 24 h PCS, the LEC transcriptome further remodels, with 16% of its transcriptome differentially expressed compared to naïve LECs23. This includes robust upregulation of various genes involved in the inflammatory response, like CXCL1, S100a9, and CSF3/G-CSF, and the first detected upregulation of fibrotic markers such as alpha smooth muscle actin, tenascin C, and fibronectin31,32. Fibrotic marker proteins and the regulator of TGFβ activation, αVβ8 integrin is significantly upregulated by 48 h PCS, and the transcriptome further remodels31,32. Consistent with this, the marker of canonical TGFβ signaling, pSMAD3, is also first observed at 48 h and its levels robustly upregulate by 72 h (Figure 6C)31. Maximal LEC EMT and fiber cell marker upregulation is observed at 5 days PCS33.
Figure 1: Schematic of the in vivo surgical technique that mimics cataract surgery in a mouse model. Created with BioRender.com Please click here to view a larger version of this figure.
Figure 2: Alternative approaches for extracting the lens fiber mass from the lens capsule. The nuclear mass can be removed in 2 approaches either using the (A) needle/syringe apparatus or (B) micro forceps. Created with BioRender.com Please click here to view a larger version of this figure.
Figure 3: Alternative approaches for suturing the corneal incision. The number of stitches required to properly close the corneal opening depends on the size of the initial corneal incision. These approaches will close either (A) a small corneal incision using 1 square knot stitch or (B) a larger corneal incision using 2 square knot stitches, one square knot stitch in the upper and lower sections of the original incision. Created with BioRender.com Please click here to view a larger version of this figure.
Figure 4: Visualization of the critical steps to model cataract surgery in mice. Here we depict the following key steps of the protocol: (A) the typical size of a dilated mouse pupil when it is ready for surgical manipulation; (B) a corneal incision; (C) what the lens nucleus and fiber mass look like outside of the ocular cavity, as well as how to identify an intact capsular bag; (D,E) suturing of the corneal flaps; and lastly, (F) how the final square knot stitch should close the corneal gap. Please click here to view a larger version of this figure.
Figure 5: Injury response time course in lens epithelial cells PCS in a mouse model. This represents data collected from various published immunostaining and RNA sequencing experiments12,15,23,25,26,31. Staining was analyzed using embedded post-surgical and 'time zero' control mouse eyes. RNA sequencing was performed on mouse capsular bags post-surgery. Created with BioRender.com. Please click here to view a larger version of this figure.
Figure 6: Efficacy of cataract surgery model. (A) Volcano plot of genes differentially expressed in LECs at 6 h PCS. Blue dots represent negative fold change, red dots represent positive fold change, and yellow dot represents Il19. (B) Successful implementation of this technique has been used in studying the acute response of LECs following CS evident in this immunostaining of classic inflammatory markers like CXCL1 (red), S100a9 (red), and G-CSF (red), which upregulate as early as 3 h PCS with a more robust response observed at 6 h PCS; nuclei stained with Draq5 (blue). (C) Additionally, further efficacy of this surgical model is evident in this immunostaining of pSMAD3 (red) in LECs PCS, which correlates with F4/80-positive macrophage infiltration (green) in lens capsular bags. Detectable pSMAD3 is observed at 48 h PCS with peak pSMAD3-positive nuclei at 5 days PCS31. Merge: pSMAD3 (red); F4/80 (green); nuclei stained with Draq5 (blue). Scale bar: 100 µm. Abbreviations: e, remnant lens epithelial cells/lens cells; lc, lens capsule. Panel A has been reprinted with permission from Novo et al.23. Panel B and Panel C have been reprinted with permission from Jiang et al.31. Please click here to view a larger version of this figure.
This surgical technique requires advanced mouse handling and micro-surgical skills that take practice to develop. The most difficult to master is the placement of fine sutures in the cornea to close the wound. Suppose the experimenter is a total novice to suturing. In that case, it is recommended to first practice using string and cloth, then move to practice using large-diameter sutures on small fruits to master the needed hand movements to make surgical square knots. Then, the experimenter must practice making incisions with a scalpel and then suturing those cuts on fruits with decreasing sizes (like grapes and then move on to blueberries). In addition, the experimenter must practice the square knot stitches using materials of decreasing thickness (like yarn to string to strands of hair) until he/she masters the ability to make square knots using the thin 10-0 nylon suture needed for this procedure.
Here, we describe a technique that mimics the injury that LECs perceive following cataract surgery in order to elucidate the molecular mechanisms underlying the response of LECs to injury. Its main advantages are (1) The LEC injury response occurs in the environment of the living eye; (2) This mouse model allows the use of both genetic tools and pharmacological inhibitors to study mechanisms of the lens injury response and the resulting phenotypic transitions of remnant lens epithelial cells; (3) It allows for testing of potential therapeutics intended to alter the course of the LEC injury response in order to block PCO; and (4) The full progression of LEC responses to surgery (cell proliferation, migration, inflammation, lens fiber regeneration and epithelial-mesenchymal transition) all occur without the addition of exogenous agents such as in vitro activated TGFβ.
Its main disadvantages are that (1) the mouse eye is too small to perform an anterior capsulotomy as is typical in humans, so all LECs are present to enter injury responses, not just the equatorial epithelium; (2) the mouse eye is too small to implant a commercially produced intraocular lens although it has been reported by another research group that appropriately sized implants produced in the laboratory can be placed in the mouse capsular bag34 and used for the slow release of anti-PCO drugs; (3) the model is unable to study the often late onset of posterior capsular opacification as defined by the escape of lens cells from the barrier imposed by the square edge of the intraocular lens sealing to the lens capsule since square edge intraocular lenses are not implanted; and (4) it is a mouse model and may not fully recapitulate the response of human LECs to cataract surgery due to potential species differences in the injury response.
Thus, the mouse model of cataract surgery described here is best used for mechanistic studies of the acute LEC injury response. It can also be used to study the long-term regulation/survival of the cell populations found in Soemmerring's ring at extended times post cataract surgery which is presumably the cells that give rise to late onset PCO, which is currently the most common manifestation of PCO in humans13,18. These investigations have already revealed that αVβ8-integrin is a likely therapeutic target to prevent myofibroblast formation post cataract surgery, which would likely prevent fibrotic PCO15.
The authors have nothing to disclose.
This work was supported by the National Eye Institute (EY015279 and EY028597) and Delaware INBRE (P20 GM103446).
0.5% Erythryomycin opthalmic ointment USP | Baush Lomb | 24208-910-55 | |
1% Atropine sulfate ophthalmic solution | Amneal | 60219-1748-2 | |
1% Tropicamide opthalmic solution USP | Akorn | 17478-102-12 | |
10-0 Nylon suture | Ethicon | 7707G | |
2.5% Phenylephrine hydrochloride ophtahlmic solution USP | Akorn | 17478-200-12 | |
26 G 1/2 Needles – straight | BD PrecisionGlide | 5111 | |
27 G 45° bent dispensing tip 1" | Harfington | ||
Balanced saline solution | Phoenix | 57319-555-06 | |
Bupranorphine (0.1 mg/kg) | APP Pharmaceticals | 401730D | |
Chlorhexidine solution | |||
Needle holder 2-110 | Duckworth & Kent | 2-110-3E | |
Noyes scissors, straight | Fine Science Tools | 12060-02 | |
SMZ800 Nikon model microscope | Nikon | ||
Sterile disposable scalpel No.11 | Feather | 2975#11 | |
Tweezers #5 Dumont | Electron Microscopy Sciences | 72700-D | |
Xylazine/Ketamine/Acepromazine (35 mg/kg, 80 mg/kg, 0.5 mg/kg) solution | APP Pharmaceticals | 401730D |