Assessment of Zebrafish Lens Nucleus Localization and Sutural Integrity

Summary

These protocols were developed to analyze cortical lens morphology, structural integrity of the zebrafish lens sutures in fixed and live lenses and to measure the position of the zebrafish lens nucleus along the anterior-posterior axis.

Abstract

The zebrafish is uniquely suited to genetic manipulation and in vivo imaging, making it an increasingly popular model for reverse genetic studies and for generation of transgenics for in vivo imaging. These unique capabilities make the zebrafish an ideal platform to study ocular lens development and physiology. Our recent findings that an Aquaporin-0, Aqp0a, is required for stability of the anterior lens suture, as well as for the shift of the lens nucleus to the lens center with age led us to develop tools especially suited to analyzing the properties of zebrafish lenses. Here we outline detailed methods for lens dissection that can be applied to both larval and adult lenses, to prepare them for histological analysis, immunohistochemistry and imaging. We focus on analysis of lens suture integrity and cortical cell morphology and compare data generated from dissected lenses with data obtained from in vivo imaging of lens morphology made possible by a novel transgenic zebrafish line with a genetically encoded fluorescent marker. Analysis of dissected lenses perpendicular to their optical axis allows quantification of the relative position of the lens nucleus along the anterior-posterior axis. Movement of the lens nucleus from an initial anterior position to the center is required for normal lens optics in adult zebrafish. Thus, a quantitative measure of lens nuclear position directly correlates with its optical properties.

Introduction

The zebrafish is an excellent model for studying lens development and physiology due to the anatomical similarities to mammalian lenses, relative ease of genetic and pharmacological manipulation, speed of embryonic eye development, small size and transparency at early stages allowing for in vivo imaging. The methods described here were developed to analyze zebrafish lens morphology at embryonic and adult stages with a focus on sutural integrity, cortical membrane morphology in vitro and in vivo, and location of lens nuclear position along the anterior-posterior axis ex vivo. These methods serve as a starting point for functional studies of lens development and physiology, as well as reverse genetic screens for lens phenotypes in zebrafish.

Imaging zebrafish lens morphology

Aquaporin 0 (AQP0) is the most abundant lens membrane protein and is critical for both, lens development and clarity, due to multiple essential functions in mammals. Zebrafish have two Aqp0s (Aqp0a and Aqp0b) and we have developed methods to analyze loss of their functions in both embryonic and adult lenses. Our studies reveal that aqp0a^-/- mutants develop anterior polar opacity due to instability of the anterior suture, and aqp0a/b double mutants develop nuclear opacity¹. AQP0 has been shown to play roles in water transport², adhesion^3,4, cytoskeletal anchoring⁵ and generation of the refractive index gradient⁶, but these studies have largely been performed in vitro. The zebrafish provides a unique opportunity to study how loss of function, or perturbed function of Aqp0a or Aqp0b would affect morphology and function in a living lens. To assess lens cell morphology and sutural integrity during development, we modified existing in vitro immunohistochemical methods⁷ for use in embryonic and adult lenses, and generated transgenics to monitor this process in vivo.

Immunohistochemical analysis of plasma membrane structure and sutural integrity was performed in whole fixed embryos and adult lenses. Zebrafish lenses are extremely small (lens diameter is ~100 µm in embryos and up to 1 mm in adults) compared with their mammalian counterparts and have point sutures⁸, which are infrequently captured in cryosections. Thus, whole lenses are essential for analyzing sutural integrity. For in vivo analysis of anterior suture formation, and imaging of precise lens cell architecture, we generated transgenics expressing mApple specifically labeling lens membranes.

Advantages of live imaging of lens membrane transgenics include: 1) avoiding fixation artifacts, 2) studying dynamic morphological changes in time-lapse movies, and 3) enabling longitudinal studies in which earlier events can be correlated with later phenotypes. Pigmentation of the iris normally prevents clear imaging of the lens periphery. Addition of 1-phenyl 2-thiourea (PTU) before the primordia-5 (prim-5) stage⁹ prevents melanogenesis and eye pigmentation up to around 4 days postfertilization (dpf). However, after 4 dpf, the lens periphery is obscured in vivo, particularly at older stages. Furthermore, the density of the lens itself obscures imaging of its posterior pole. Therefore, to study morphology of the lens periphery, or the posterior suture, after 4 dpf, lenses need to be excised and fixed.

Transgenic zebrafish lines have been used to analyze embryonic lens membrane structure in vivo¹⁰. The Q01 transgenic expresses a cyan fluorescent protein fused to a membrane targeting sequence, Gap43, driven by the EF1α promoter and a hexamer of the DF4 pax6 enhancer element ubiquitously in lens fiber cells¹¹. Q01 does have extra-lenticular expression, including amacrine cells in the retina, which adds background signal if the primary interest is the lens. We developed a novel transgenic line that expresses a membrane-tethered mApple specifically in the lens, with the aim of avoiding any extra-lenticular signal.

Lens nucleus localization

We discovered that the lens nucleus moves from an initial anterior location in larval zebrafish to a central location in the anterior-posterior axis in adult lenses. This shift in the position of the high refractive index lens nucleus is thought to be a requirement for normal development of zebrafish lens optics¹. Our methods allow quantification of lens nuclear centralization in relation to the lens radius. Using this method, we showed that Aqp0a is required for lens nuclear centralization¹, and this simple method can be applied to other studies of the development and physiology of the lens and its optical properties in the zebrafish model.

Protocol

The animal protocols used in this study adhere to the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research and have been approved by the Institutional Animal Care and Use Committee (IACUC) of University of California, Irvine.

1. Zebrafish Husbandry and Euthanasia

1. Raise and maintain zebrafish (AB strain) under standard laboratory conditions¹². Raise embryos in embryo medium (EM)¹². Add 0.003% PTU to EM from 20-24 h postfertilization (before the prim-5 stage) to prevent pigment formation in embryos¹² used for imaging at embryonic stages⁹. Raise larvae from 6-30 days postfertilization (dpf) on a diet of live rotifers until 14 dpf, and live artemia after 14 dpf¹².
   CAUTION: PTU is very toxic
2. Anesthetize fish using tricaine until non-responsive to touch.
   1. Prepare tricaine stock by combining 400 mg of 3-amino benzoic acidethylester, with 2.1 mL of 1 M Tris (pH 9), in 100 mL of ddH₂O. Adjust to pH 7.0 and store at -20 °C.
      CAUTION: Tricaine is toxic.
   2. Dilute 4.2 mL of tricaine stock in 100 mL of tank water to anesthetize larvae/adults or in EM to anesthetize embryos (final concentration of tricaine at 0.0165% w/v).

2. Fixation of Embryos and Larvae

NOTE: Immunohistochemical protocols were adapted from previously published materials⁷.

1. Fix dechorionated embryos or larvae up to 2 weeks postfertilization in 4% (v/v) paraformaldehyde (PFA) in phosphate buffered saline (PBS) overnight at 4 °C on a rocker.
   CAUTION: PFA is combustible and is carcinogenic.
2. Wash embryos three times for 10 min in PBS, and permeabilize in PBS with 10% Triton and 1% DMSO (PBS-T) overnight at 4 °C.
   CAUTION: DMSO is toxic, is harmful by ingestion or skin absorption, carries hazardous materials through skin, and is combustible. Triton is toxic, corrosive and is hazardous to aquatic environments.

3. Dissection of Larval and Adult Zebrafish Lenses

1. Anesthetize fish from 6 dpf larvae to adulthood with tricaine until non-responsive to touch but still showing a strong heartbeat. Measure fish standard length as per Schilling (2002) for staging¹³.
2. Immediately excise eyes using micro-dissection scissors and place into a dissection dish in PBS (Figure 2 shown for adult eyes).
   1. Make a custom 35 mm dish filled with silicone for lens dissections. Once silicone has set, excise a divot around 2-3 mm in diameter and 0.5 mm deep for immobilizing adult eyes/lenses during dissection.
3. Dissection of adult zebrafish lenses
   1. Place an adult eye into the dish divot posterior side up filled with PBS. Immobilize the eye by inserting forceps at <45° angle through the optic disc. Be careful not to nick or compress the lens. Make two or three radial incisions through retina and sclera from the optic disc to the ciliary zone with dissection scissors.
   2. Peel back the retina and sclera like flower petals and invert the eye, cornea side up. Immobilize the lens indirectly via manipulation of the sclera and cornea with the flat side of the scissors, while pulling away the retina and attached tissues with forceps. Carefully trim excess tissue from the lens.
      NOTE: It is vital to dissect carefully to obtain consistently healthy lenses.
4. Dissection of larval zebrafish lenses
   1. Place a larval eye posterior side up onto the flat part of the silicone dish filled with PBS and use a sharpened tungsten needle to make radial cuts through the retina and sclera while immobilizing the eye with another tungsten needle or forceps.
      NOTE: Be careful not to damage the lens.
      1. Sharpen the tip of a 2 cm length of 0.1 mm tungsten wire electrolytically by suspending the wire tip into 10% (w/v) NaOH and applying a low voltage alternating-current¹⁴. Secure the needle into a Pasteur pipette by melting the glass end using a Bunsen burner.
         NOTE: Alternative fine dissection tools can also be used.
         CAUTION: NaOH is corrosive.
   2. Gently scoop out the lens from the dissociated eye with a blunt side of the needle, and carefully pull away attached tissue.

4. Fixation of Dissected Lenses

1. Immediately fix dissected lenses in 1.5% (v/v) PFA in PBS for 24 h at room temperature (RT). Wash lenses three times in PBS for 10 min each.
2. Permeabilize lenses for whole mount analysis or cryoprotect for cryosectioning (see section 8).
   1. Permeabilize lenses in PBS-T overnight at 4 °C.

5. Lens Immunohistochemistry

1. Label membranes of fixed embryo/larval or adult lenses by incubation in Phalloidin- Alexa Fluor 546 (1:200) and cell nuclei with DAPI (1:1,000 of 5 mg/mL stock) in PBS-T overnight at 4 °C.
   CAUTION: DAPI is a skin and eye irritant.
2. Wash three times with PBS-T for 10 min each. Clear tissue by incubation in 30%, 50% and 70% glycerol in PBS-T (v/v) for at least 1 h at RT each, or overnight at 4 °C.
3. Mount fish in 70% glycerol in PBS-T (v/v) onto glass bottom 35 mm microwell dishes with eyes as flat and straight against the cover slip as possible. For younger fish, a slight anterior-lateral tilt may help to orient the lens suture to be parallel to the coverslip.

6. Analysis of Zebrafish Anterior Lens Sutures Using a Transgenic Line In Vivo

1. Generate Tg(βB1cry:mAppleCAAX) lines using the Tol2 kit¹⁵. The Tol2 transposable element system¹⁵ enables stable integration of the construct where mApple is driven by 300 bp of the human βB1-crystallin promoter¹⁶ tethered to the membrane by the CAAX sequence resulting in expression specifically in lens fiber cell membranes.
   NOTE: This construct (ID:122451) is available for purchase.
   1. On the morning of injection, prepare the injection mixture and keep on ice. To reach a final volume of 10 µL of containing RNase- and DNase-free H₂O, add: 1.5 µL of huβB1cry:mAppleCAAX DNA construct at 50 ng/µL - [0.375-0.75 pg/final injection], 1.5 µL of Tol2 transposase mRNA at 300 ng/µL (Tol2 kit)¹⁵ [15-30 pg/final injection], and 1 µL of phenol red indicator at 1% w/v [1 x10^-5% (w/v)/final injection].
   2. Inject 50-100 pL of injection mixture into 1-cell stage embryos¹².
2. Measure transgenesis efficiency in F0 injected embryos by measuring frequency of embryos with mApple positive lenses at 3 dpf.
   NOTE: Expect to have >80% transgenesis efficiency using this method. F0 mosaics allow detailed analysis of membrane morphologies of individual cells. Varying levels of mosaicism allows one to image the morphologies of single or small groups of cells, while more global lens expression allows analysis of multicellular structures like lens sutures in vivo.
3. Outcross F0 mosaic fish to generate stable lines, which label all fiber cell membranes. F0 founders with the transgene integrated into the germline will generate offspring with stable integrations that can be maintained as transgenic lines.

7. Analysis of Zebrafish Anterior Lens Sutures Using a Transgenic Line In Vivo

1. Anesthetize 3 dpf or adult mosaic or stable transgenic fish and mount in 1% low melt agarose (LMA) with tricaine with the eye flat against the cover slip of glass bottom microwell dishes.
   1. Dissolve 1 g of LMA in 100 mL of EM (without methylene blue) to make 1% LMA. Store long term as 15 mL stocks at 4 °C. Microwave to dissolve stock and store at 42 °C until each tube is used.
2. Cover fish up to 6 dpf with EM with tricaine, or with tank water with tricaine for adults when LMA is set.
   1. Mix 5 mL of EM without methylene blue or tank water with 250 µL of tricaine stock and 7 µL of PTU if imaging PTU treated fish.

8. Lens Cryosectioning and Immunohistochemistry

1. Cryoprotect fixed embryos or adult lenses by placing into 10% sucrose in PBS (w/v), 20% sucrose for 1 h at RT each (or overnight at 4 °C) and overnight in 30% sucrose at 4 °C.
2. Embed tissue in OCT in base molds, and then freeze onto chucks with OCT. Cryosection at 12-14 µm and collect onto a Superfrost/Plus microscope slides. Warm sections onto slide for 10 min on a slide warmer prewarmed to ~35 °C.
3. Wash sections three times with PBS for 10 min each. Label sections with Phalloidin-Alexa Flour 546 (1:200) with DAPI (1:1,000) or WGA-Alexa Flour 594 (1:200), followed by three PBS washes. Mount with anti-fade mounting medium, and seal coverslip with nail polish.

9. Imaging

1. Acquire images with a confocal microscope. Acquire z-stacks or optical slices using a 60x N/A 1.2 water-immersion objective, or similar, at specific regions from the anterior to posterior lens pole. Use the following acquisition settings: 1.2 airy disc using the 561 nm laser, Texas red filter for phalloidin-Alexa Flour 546 or mApple, or the 405 laser with the UV filter for DAPI.
2. Correct z-intensity laser power to counteract signal loss with depth into the lens. This allows for a strong signal at the posterior pole of fixed and live 3 dpf embryos, and strong signal in equatorial optical slices in adult fish lenses.
3. Compile and view images using image processing software.

10. Measurement of Lens Nucleus Localization in the Anterior-posterior Axis

1. Orient freshly excised lenses axially in PBS in a 35 mm dish with a coverglass bottom, with poles and sutures oriented parallel to the plane of focus. Look for a difference in the refractive index, which usually occurs at the interface of the lens cortex and lens nucleus to identify the lens nucleus.
   NOTE: Healthy wild type adult lenses are extremely transparent making it difficult to see the lens sutures and nucleus, or to determine lens orientation. In this case, a slight nick with forceps to the lens capsule causes minor damage, which makes the sutures and lens nucleus apparent in about 10 min helping to orient the lens for this measurement. However, the lenses become unusable for downstream applications as they are now damaged.
2. Take images of lenses with the lens nucleus in focus under bright field illumination with or without DIC optics using a dissection microscope with a camera attached. Image a micrometer under the same magnification for calibration.
   1. Click on the live view button, adjust the exposure settings to visualize the lens nucleus and lens periphery, and take an image by clicking the snap shot button. Save as a tif file.
   2. Use image processing software to calibrate the acquired lens images. Select the straight-line tool and draw a line of known length on the micrometer image, click analyze | set scale. Enter known distance, units and select 'global' calibration, and click ok.
3. Measure the distance of the center of the lens nucleus to the anterior pole (a - r).
   1. Use the straight-line tool in image processing software to draw a line across the imaged center of the lens nucleus in an axial orientation. Take the center of this line as the center of the lens nucleus.
   2. Draw another line from this point to the anterior pole of the lens and select 'measure' in the 'analyze' menu to measure the distance (a - r), where a is the radius of the lens and r is the distance from the center of the lens to the center of the nucleus.
   3. Draw another line from the anterior to the posterior poles and measure this distance as the lens diameter (2a). Copy these lengths from the 'results' window and transfer to a spreadsheet or statistics program.
      NOTE: It is easier to precisely measure from the center of the nucleus to the anterior pole, than measuring from the center of the lens nucleus to the center of the lens. This measurement is converted by the formula in step 10.3.5 to report the distance of the center of the lens nucleus to the center of the lens. Always use the adult nucleus for measurements, even if the embryonic nucleus is evident.
   4. Divide the diameter by 2, to calculate the lens radius, a.
   5. Calculate r/a, as , which is the normalized localization of the lens nucleus with respect to the lens radius.
      NOTE: For a nucleus placed closer to the anterior pole, r/a will be >0.0, while a centrally localized nucleus will have an r/a of 0.0.
4. Graph the log of the normalized axial nucleus measurement as a function of the standard length to determine changes in lens nucleus localization during zebrafish development.

Representative Results

Adult zebrafish eye anatomy closely resembles that of mammals (Figure 1A). Despite some differences between zebrafish and mammalian eyes, such as having a ciliary zone instead of a ciliary body¹⁷, differences in optical properties¹⁸, and differences in morphogenesis during embryonic development¹⁹, the zebrafish eye is an excellent model for studying eye development and understanding ophthalmic disease^20,21,22. A simple epithelium lines the anterior pole of the lens, which at the equator undergoes a life-long process of proliferation, elongation, and differentiation into fiber cells, constituting the bulk of the lens. Mature fiber cells (Figure 1B, light blue) first differentiate in the embryo, are devoid of organelles and are never replaced. Differentiating fiber cell tips meet at the poles to form the anterior and posterior sutures. These cells are then internalized by newly differentiated fiber cells. Like all vertebrate lenses, the zebrafish lens thus has a gradient of development with the oldest cells located in the center of the lens nucleus. The zebrafish lens placode forms at 16 h postfertilization (hpf), which elongates to form the primary lens fibers at 18 hpf, and secondary fibers form the transition zone at 28-36 hpf. The posterior suture is visible at 48 hpf, which is before the anterior suture becomes visible¹⁰.

The precise development and maintenance of regular lens architecture is essential for its optics. Here we describe methods for analysis of zebrafish lens cell morphology in vitro and in vivo, including both advantages and limitations, which we developed to analyze phenotypes resulting from loss of-function mutants of two zebrafish Aqp0s, Aqp0a and Aqp0b¹. For assessment of adult lens morphology in vitro, lenses were dissected (Figure 2) and immediately fixed. Embryonic assessment was carried out in whole fixed embryos (Figure 3) and compared to live transgenic embryos (Figure 4). Dissected and fixed adult lenses (Figure 5) were compared to transgenic adult lenses imaged live (Figure 6). Differences in detection and visualization of specific parts of the lens using these methods are summarized (Table 1).

Phalloidin was found to be superior for labeling lens fiber cell membranes in zebrafish compared to wheat germ agglutinin (WGA). WGA is a lectin that binds to N-acetyl-D-glucosamine and sialic acid, and WGA conjugated to fluorophores is typically used to label lens fiber cell membranes in human²³, bovine²⁴, ovine²⁵, mouse²⁶, rat²⁷ and zebrafish^28,29. Here we use phalloidin on whole zebrafish lenses (Figure 3, Figure 5).

Embryonic and larval zebrafish larvae up to 7 dpf are small and permeable enough to allow penetration of phalloidin into the lens outer cortex. By 3 dpf the lens nucleus is compact and devoid of organelles, anterior sutures form at the anterior pole (Figure 3A,B), and phalloidin is excluded from the lens nucleus in an equatorial optical plane (Figure 3C,D). Phalloidin is likely excluded from the nucleus due to the high compaction of fiber cell membranes, consistent with previous studies²⁸. However, the outer cortex is visualized in great detail with this staining (Figure 3C-F). In cross section, fiber cells take on a flattened hexagonal shape, with longer broad sides, and shorter narrow sides. Phalloidin strongly labels both the narrow and broad fiber cell membranes (Figure 3E-F) highlighting the compaction of the cortical fiber cells between broad sides. This organization is largely unaffected in aqp0a/b double mutants (Figure 3B, D, F and H).

Mosaic expression of the transgene (mApple tethered to the cell membrane by the CAAX motif driven by a 300 bp promoter region of the human βB1 crystallin promoter) strongly expresses in lenses beginning at 2 dpf. The transgene is randomly integrated into the genome resulting in heterogenous expression of mApple. We show examples of a 3 dpf lens with strong expression in the cortex (Figure 4) and expression in parts of the nucleus (Figure 4D). The membrane marker is not limited by permeability like phalloidin, thus it labels the lens nucleus. Mosaics (Figure 4) generated by DNA injections that are only expressed in a small subset of cells are useful for analyzing individual cell morphologies compared to stable lines, which label every cell that expresses βB1 crystallin (Figure 6). In Tg(βB1cry:mAppleCAAX) mosaics the morphology of the anterior suture can be visualized in z-projections taken at the anterior pole (Figure 4A,B). Note that the morphology of cell membranes differs from fixed lenses labelled with phalloidin (Figure 3A,B), and presence of subdomains in broad cell membranes (Figure 4a' white arrows) previously identified in lenses from other species³⁰ is visible. However, weaker labeling of narrow fiber cell membranes results in an inability to see the striking convergence of cells at the anterior suture (Figure 4A,B arrow) seen in fixed lenses labeled with phalloidin (Figure 3A,B arrows).

Interestingly, optical slices taken at the lens equator also reveal a different labelling pattern, with obvious disruptions to cell volume in the aqp0a/b double mutants compared to WT (Figure 4C-F). This is likely because the transgenic labels broad fiber cell membranes more effectively than phalloidin. We were able to obtain z-projections through the posterior pole to analyze integrity of the posterior suture with the use of Z-correction, which increased laser power with depth into tissue. Clear analysis of the posterior suture (Figure 4G,H) in intact fish was limited to the first 5 dpf, and after that the lens was too dense, resulting in loss of signal in the posterior part of the lens.

Fixed dissected adult lenses labeled with phalloidin revealed striking detail in the highly ordered architecture of fiber cell rows converging to form anterior sutures (Figure 5A, arrow) and morphology of the cortex (Figure 5C-H), due to very strong labeling of narrow fiber cell membranes by phalloidin. Maximum Z-projections of the anterior pole revealed loss of a tight anterior suture in aqp0a/b double mutant lenses (Figure 5B, arrow), compared to WT (Figure 5A, arrow), which evident as a mass of membranes and cell nuclei in optical slices 30 µm from the anterior pole (Figure 5D). In deeper optical slices, phalloidin labeling was excluded from the deeper lens cortex and nucleus (Figure 5E,F). The overall organization of fiber cell rows in the outer cortex was somewhat affected by lack of a tight anterior suture in aqp0a/b double mutant lenses, making it impossible to position lenses exactly perpendicular to the imaging plane (Figure 5F) compared to WT lenses (Figure 5E). However, high power images of fiber cell rows taken in the equatorial plane revealed that cells in the outer cortex still formed orderly rows, visible due to strong narrow fiber cell labeling (Figure 5G,H). It was impossible to discern individual fiber cell broad membranes due to a combination of their tight compaction, and weak signal. Since these lenses were dissected, they could be mounted with the posterior side facing the objective, allowing z-projections of the posterior sutures, which showed no differences between genotypes as posterior fiber cell tips formed tight sutures (Figure 5I,J arrows).

Z-projections of lenses of live anesthetized adult stable Tg(βB1cry:mAppleCAAX) fish reveal cortical membrane morphology in the anterior cortex (Figure 6A,a'), as well as the extent of cell convergence at the anterior pole (Figure 6B). It was more difficult to mount the fish with lens sutures perpendicular to the imaging plane, compared to fixed dissected lenses. Stable lines carrying the transgene express mApple in the lens nucleus (Figure 6C,D). Optical slices at the equator revealed a strong signal indicating extensive compaction of the nucleus when imaged with the same laser power as the anterior cortex (Figure 6C). Interestingly, no cellular resolution of the outer cortex was evident in adults, unlike in embryos. This is likely due to the iris blocking a clear signal from the lens periphery. It may be possible to obtain a stronger lens cortical signal by dilation of the pupil prior to imaging. With reduced laser power, it was possible to distinguish some cell layers in the lens nucleus (Figure 6D). The adult zebrafish lens is very dense, and it was impossible to obtain signal from the posterior pole in vivo.

We have shown that the zebrafish lens nucleus moves in the optical axis from an initial anterior position in larval lenses to a central position in adults¹. These movements are highlighted in axial lens sections, as phalloidin and WGA are excluded from the lens nucleus (Figure 7A-C). We have shown that Aqp0a is required for this centralization process¹, but this mechanism is likely to be affected by other proteins. The localization of the center of the lens nucleus in relation to its position along the anterior-posterior axis was measured by placing dissected whole lenses in their axial orientation and imaging under DIC illumination, thus highlighting slight differences in refractive properties (Figure 7E). The sutures usually have a different refractive index than the surrounding cortex, so can be used as a guide for orientation. Young larval lenses have more obvious sutures (posterior sutures in very young lenses), and lens nuclei, which have a different refractive index, are obviously asymmetric, making it easier to orient lenses at these stages. The relative distance of the center of the lens nucleus from the center of the lens (r) is normalized to the lens radius (a). This is then graphed with relation to zebrafish development, for which standard length measurement is used¹³. In WT, the lens nucleus starts off closer to the anterior lens pole , and then centralizes  in adulthood (Figure 7F).

Figure 1: Zebrafish adult eye and lens diagram. Anterior is taken as where the light enters the eye, which in the zebrafish is actually lateral. (A) The zebrafish lens together with the cornea focus light onto the retina. In aquatic animals, the cornea plays a minor role in light refraction, but instead, the lens has a higher refractive index¹⁸. The lens separates the anterior chamber filled with aqueous humor and posterior chamber filled with vitreous humor. (B) The zebrafish lens is more spherical than mammalian lenses. Fiber cell tips meet at point or umbilical sutures⁸ at the anterior and posterior poles. Differentiating fiber cells in the lens cortex have nuclei and organelles, while mature fiber cells in the lens nucleus (light blue) are devoid of organelles and nuclei. Please click here to view a larger version of this figure.    

Figure 2: Zebrafish lens dissection. (A) Eyes are dissected from anesthetized fish and placed posterior side up (B). (C) Eyes are immobilized by forceps via the optic disc (od) and two to three radial incisions are made in retina and sclera from the optic disc to the zonules. (D) Fold the flaps out to reveal the lens. (E) Rotate the eye to face cornea up, immobilize the lens indirectly via the cornea and pull away the sclera-retina. (F) Trim away remaining retina and ciliary zone/zonules from the lens. Please click here to view a larger version of this figure.

Figure 3: Embryonic lens morphology in vitro. Fixed and permeabilized embryos labelled with phalloidin (red) and cell nuclei with DAPI (blue) reveal membrane morphology and suture integrity at the poles. Z-projections reveal that WT and aqp0/b double mutant lenses exhibit very regular fiber cell arrangement that meet at a very tight anterior suture (arrows) at 3 dpf (A, B). Optical slices taken at the equator reveal tight rows of hexagonal fiber cells packed in the outer cortex in both genotypes (C-F). Both, narrow and broad sides of fiber cell membranes are labelled as shown in (E). Posterior sutures (arrows) are formed and tight in both genotypes (G, H). Please click here to view a larger version of this figure.

Figure 4: Embryonic lens morphology in vivo. Live imaging of anesthetized 3 dpf mosaic F0 injected Tg(βB1cry:mAppleCAAX) lenses reveal anterior sutures (arrows) in z-projections (A, B). (a') Membrane subdomains are evident as puncta (white arrows) in broad fiber cell membranes. Narrow fiber cell membranes are also evident (black arrows). WT lenses reveal tightly packed lens cortical fiber cells (C, E), compared to disrupted cortex of aqp0a/b double mutants^- with evident swollen cells (D, F). Focussing on posterior (G, H) sutures (arrows) reveals no difference between the genotypes (G, H). Please click here to view a larger version of this figure.

Figure 5: Adult lens morphology in vitro. Excised, fixed and permeabilized lenses from fish over 23 mm standard length were labelled with phalloidin (red) and DAPI (blue). 150 µm Z-projection at the anterior pole reveal strict arrangement of fiber cell tips converging at a tight suture (arrow) in WT (A), but not aqp0a/b double mutants (arrow), which instead have a mass of membranes and nuclei (B). An optical slice taken 42 µm from the anterior pole reveal ordered cells converging in the center in WT (C), but a mass of nuclei where the suture should be in mutant lenses (D). Optical slices through the equator of the lens, at 130 µm from the anterior pole reveal tightly packed rows of fiber cells in WT (E, G) and mutants (F, H). Posterior sutures (arrows) are formed and tight in both genotypes (I, J). Please click here to view a larger version of this figure.

Figure 6 Adult lens morphology in vivo. Live imaging of anesthetized adult stable Tg(βB1cry:mAppleCAAX) WT lenses. Z-projections at the anterior cortex reveal cortical morphology (A, a'), and anterior suture (arrow), with cells converging to a point in a single optical slice (B, arrow). The cortex can be visualized at higher laser power in an optical slice through the equator (C), compared to the stronger signal form the lens nucleus (D). Please click here to view a larger version of this figure.

Figure 7: Zebrafish lens nucleus centralization. Fixed and cryoprotected embryos (A) and lenses (B, C) were axially cryo-sectioned and labelled with phalloidin-546 (red in A, B), DAPI (blue in A, B) or WGA-594 (red in C). The lens nucleus is devoid of cell nuclei, and is placed closer to the anterior (a) lens at 3 dpf (A), and in 1 month old lenses (B), compared to being centrally placed in adult lenses (C). Dissected lenses from a 1 month old zebrafish of 4.1 mm standard length were oriented equatorially (D), or axially (E). The relative localization of the lens nucleus in the axis was calculated using the following strategy: (E) the distance of the center (*) of the lens nucleus (white dotted line) was measured to the anterior pole, as this more practical than measuring the distance (r) to the center of the lens (plus). This was normalized to the lens radius (a), which was intially measured as the axial (ant.-pos.) lens diameter. The log of the normalized axial nucleus localization was graphed as a function of zebrafish development, measured by the standard length (F). Please click here to view a larger version of this figure.

Feature	IHC embryo	Tg embryo	Whole IHC adult lens	Tg adult
Live	N	Y	N	Y
Anterior suture	Y	Somewhat	Y	Somewhat
Posterior suture	Y	Y	Y	N
Cortical cellular detail	Somewhat	Y	Y	Anterior somewhat
Lens nucleus detail	N	Y (stable)	N	Y (stable)
Secondary label	Y - Antibodies	Live only with only Tg	Y - Antibodies	Live only with only Tg
Broad/Narrow fiber label	Both	Broad	Mostly narrow	Broad

Table 1: Summary of comparison of in vivo vs in vitro methods for analysis of lens morphology in zebrafish. Y = Yes, N = No

Discussion

Analysis of zebrafish lens morphology is the initial step in understanding phenotypes in mutants, or the effects of pharmacological interventions aimed at studying biology of the ocular lens. We outline methods to analyze lens sutures, cortical fiber cell morphology and aspects of the lens nucleus. These approaches are a combination of in vitro and in vivo (compared in Table 1). The in vitro methods allow for greater detail of the outer cortical cell morphology, as well as access to the posterior suture in adult lenses.

In vivo analysis is possible with the use of transgenic markers. The transgene we introduce here is lens membrane specific, compared to the existing Q01 zebrafish line, which while labelling the lens membranes also labels other cell types in the eye, including amacrine cells, complicating interpretation¹¹. The in vivo signal is limited by the pigmented epithelium of the eye, which forms during the second day of embryogenesis, so embryos need to be PTU treated for analysis at this and subsequent stages. In adults, the iris obscures clear analysis of the lens cortex, but allows for in vivo analysis of the anterior lens cortex. Live imaging of fish lenses allows for longitudinal studies of lens morphology in the same animal. This can be an extremely useful tool for studying effects on dynamic processes, such as cell volume disruption in response to osmotic or pharmacological perturbations. An unexpected finding is that we can clearly visualize the membrane subdomains in broad fiber cell membranes in the live larval transgenics, but not in fixed lenses. This highlights the difference in labelling by the transgene and phalloidin, as well as the fact that these subdomains may be affected by the fixation process.

Analysis of the molecular mechanisms of lens nucleus centralization using the methods we describe may uncover mechanisms essential for the developmental of lens optical properties. Although, to date, axial lens nucleus asymmetry has only been reported in zebrafish, by studying this process we are likely to gain insights into mechanisms required for the development of lens optics in other species. In the future, the transgenic animals can be used in combination with other applications - such as overexpression studies, with the use of a green transgenesis marker. These could be used to study effects of overexpressing a specific protein in the lens, or for rescue of phenotypes in existing mutants.

Materials

Name	Company	Catalog Number	Comments
1-phenyl-2-thiourea (PTU)	Sigma	P7629	CAUTION – very toxic
4% Paraformaldehyde aqueous solution	Electron Microscopy Sciences	RT 157-4	CAUTION – health hazard, combustible
Confocal microscope	Nikon	Eclipse Ti-E
Cryostat	Leica	CM3050S	Objective and chamber temperature set to -21 °C
DAPI	Invitrogen	D1306	CAUTION – irritating to eyes and skin
Dimethyl Sulfoxide (DMSO)	Fisher Scientific	D128	CAUTION – combustible, penetrates skin
Disposable base mold	VWR Scientific	15154-631
Disposable Pasteur glass pipets	Fisherbrand	13-678-20A
Dumont #5 forceps	Dumont & Fils		Keep forceps sharpened
Ethyl 3-aminobenzoate methanesulfonate salt (Tricaine)	Sigma-Aldrich	A5040	CAUTION - toxic
Glass bottom microwell dish (35mm Petri dish, 14 mm microwell, #1.5 coverglass)	MatTek Corporation	P35G-1.5-14-C
Glycerol	Sigma	G2025
huβB1cry:mAppleCAAX DNA construct	Addgene	ID:122451
ImageJ	Wayne Rasband, NIH	v1.51n
Low Melt Agarose (LMA)	Apex	902-36-6
NIS-Elements	Nikon	V 4.5
NIS-Elements AR software	Nikon
Olympus  with a model 2.1.1.183 controller	Olympus Corp	DP70
Olympus microscope	Olympus Corp	SZX12
Phalloidin-Alexa Flour 546	Thermo Fisher	A22283
Phenol Red indicator (1% w/v)	Ricca Chemical Company	5725-16
Phosphate buffered saline (PBS)	Fisher Scientific	BP399
Photoshop	Adobe	CS5 v12.0
Photoshop software	Adobe	CS5 v12.0
Plan Apo 60x/1.2 WD objective	Nikon
Power source	Wild Heerbrugg	MTr 22	Or equivalent power source
Slide warmer model No. 26020FS	Fisher Scientific	12-594
Sodium Hydroxide beads	Fisher Scientific	S612-3	CAUTION - corrosive/irritating to eyes and skin, target organ - respiratory system, corrosive to metals
Superfrost/Plus microscope slide	Fisher Scientific	12-550-15
Sylgard 184 silicone Dow Corning	World Prevision Instruments	SYLG184
Tissue-Tek O.C.T. Compound	Sakura Finetek	4583
Triton X-100 BioXtra	Sigma	T9284	CAUTION – Toxic, hazardous to aquatic environment, corrosive
Vannas micro-dissection scissors	Ted Pella Inc	1346	Sharp/sharp straight tips
Vectashield antifade mouting medium	Vector laboratories	H-1000
Wheat Germ Agglutinin (WGA)-Alexa Flour-594	Life Technologies	W11262