Molecular Vision 2007; 13:2289-2300 <http://www.molvis.org/molvis/v13/a259/> Received 20 August 2007 | Accepted 8 December 2007 | Published 18 December 2007

Download Reprint

Perinatal ablation of the mouse lens causes multiple anterior chamber defects

Yan Zhang,¹ Paul A. Overbeek,² Venkatesh Govindarajan¹
 
 

¹Department of Surgery, Creighton University, Omaha, NE; ²Department of Molecular & Cellular Biology, Baylor College of Medicine, Houston, TX

Correspondence to: Venkatesh Govindarajan, Creighton University, Department of Surgery, 254 Criss III, Cancer Center, 2500 California Plaza, Omaha, NE, 68178; Phone: (402) 280-1819; FAX: (402) 280-3817; email: g.v@creighton.edu

Abstract

Purpose: The purpose of this study was to reassess the role of the lens as an "embryonic organizer" of ocular tissues.

Methods: We ablated the lens in mice by lens-specific expression of an attenuated version of diphtheria toxin A subunit(Tox176) driven by a modified crystallin promoter. Alterations in the differentiation programs of ocular tissues were examined by hematoxylin and eosin staining, in situ hybridization, and immunohistochemistry.

Results: Transgenic mice in the family OVE1757 exhibited severe microphakia. Apoptotic lens fibers were seen by embryonic day 15 (E15) and the lenses were completely ablated by post natal day 8. Multiple defects were seen in the anterior chamber. Corneal endothelial cells did not differentiate properly. The mesenchymal cells that would normally give rise to the endothelial layer were found to express N-cadherin, but they failed to form tight junctions and undergo a mesenchymal-to-epithelial transition. Although early specification of the presumptive ciliary body and iris was detected, subsequent differentiation of the iris was blocked. No dramatic changes were seen in the development of the retina.

Conclusions: These results support the hypothesis that an intact lens is essential for proper differentiation of both the corneal endothelium and the iris and that the lens "organizes" the development of tissues in the anterior chamber.

Introduction

Since the lens, cornea, and retina develop through coordinated inductive interactions, the eye has been used as a model system to investigate mechanisms of instructive induction during embryogenesis. During early ocular development, an inductive signal from the optic vesicle leads to the upregulation of transcription factors, such as Pax6 and Sox2 in the surface ectoderm, and formation of the lens placode [1]. The placode then invaginates and forms the lens. The anterior part of the lens is composed of undifferentiated cuboidal epithelial cells. The nearby surface ectoderm subsequently differentiates to form the corneal epithelium. Periocular mesenchymal cells migrate into the region between the surface ectoderm and the anterior part of the lens and are sorted through a process that is poorly understood to give rise to the corneal stromal keratocytes and the corneal endothelium. In chicken embryos, the corneal endothelium and stroma form as a result of two waves of mesenchymal migration; the first wave of migration gives rise to the corneal endothelium and the second wave to corneal stromal keratocytes [2-5]. During maturation, the corneal endothelial precursors undergo a mesenchymal-to-epithelial transition. The differentiated corneal endothelium is critical for maintaining corneal dehydration and transparency [6,7].

Concurrent with the differentiation of the corneal endothelium, the presumptive iris and ciliary body become histologically distinct at the anterior margins of the neuroretina. The ciliary epithelium is composed of two layers; the inner layer is derived from the neuroectoderm and the outer layer from the anterior pigmented epithelium. Periocular mesenchymal cells migrate along the anterior margins of the presumptive ciliary body and iris and differentiate to form the iris and ciliary stroma. The iris stroma is composed of pigmented dendritic melanocytes and clump cells [8]. Secreted signals such as BMP4 and BMP7 and transcription factors such as Otx1 have been shown to be essential for the proper differentiation of the ciliary body and iris [9,10]. However, the molecular signals that determine the boundary between the iris and ciliary body are not known.

Previous studies in non-mammalian models indicate that the embryonic lens plays a role in specifying the proper development of surrounding ocular tissues. Removal or reorientation of the lens in chicks results in delayed differentiation of the corneal endothelial cell layer [11,12]. In addition, the epithelial cells of older embryonic lenses are able to support the formation of a corneal endothelium [4]. Chick lenses have also been shown to be capable of inducing the expression of markers of the presumptive iris and ciliary body in developing neural retina [13]. Lens ablation studies in zebrafish show that an intact lens is essential for proper retinal lamination [14]. There is also evidence that an alteration in an inductive signal from the lens is a major cause for the loss of the cornea and iris and subsequent eye regression in the cave-dwelling fishes Astyanax mexicanus [15].

In mice, lens ablation experiments performed by generation of transgenic mice with lens-specific expression of diphtheria toxin A subunit (DTA) or Tox176 (an attenuated version of DTA; more details below) lead to aphakia, cataracts and/or anopthalmia [16-21]. Alterations in corneal and iridial architecture have been reported in these mice [18]. However, these studies have focused mostly on the efficacy of the toxin for cell ablation in vivo and alterations in the differentiation programs of the affected tissues were mostly assessed by histological assays. Changes in marker gene expression were generally not analyzed and so the molecular alterations in cell fate determination have remained undefined.

Diphtheria toxin is generated as a precursor protein. Proteolytic cleavage generates A and B peptides; the A peptide contains the toxic moiety and it inactivates elongation factor-2 (EF-2) by ADP-ribosylation. The B peptide binds the cell surface and causes the toxin to be internalized. In the absence of the B peptide, the A peptide is not internalized. Therefore, expression of the A peptide (DTA) is toxic specifically in cells where it is expressed. The toxicity of DTA is sufficiently high that one molecule of DTA in the cell may be sufficient to kill the cell [22]. A point mutant of DTA, Tox176, that contains a Gly to Asp substitution at residue 128 has been generated and is estimated to be 30 fold less cytotoxic than wild-type DTA. Tox176 has been shown to be effective in ablation of lens cells [17].

We initiated our studies with the aim of ablating both the lens epithelial and fiber cells in mice during early ocular development. A hybrid promoter (DREAM) that was previously found to be active in both lens epithelial and fiber cells was used [23]. We generated two transgenic lines that carry the DREAM-Tox176 transgene. Unexpectedly, transgene expression in the DREAM-Tox176 mice was found to be localized only to the lens fibers. Nonetheless, Tox176 expression in the fibers led to embryonic microphakia and complete ablation of the lens by postnatal day 8 (P8). Although expression of the transgene was detected only in the lens, altered development of most tissues in the anterior segment was observed. Corneal endothelial precursors failed to undergo a mesenchymal-to-epithelial transition to form a mature endothelium. Interactions between the anterior ocular neuroectoderm and ocular mesenchyme were disrupted, and differentiation of the iris was affected. These results support the hypothesis that the lens plays an important role in initiation/promotion of differentiation programs of multiple tissues in the anterior chamber of the eye.

Methods

Generation of DREAM-Tox176 transgenic mice

Plasmid Tox176, encoding an attenuated version of diphtheria toxin (a gift from Ian Maxwell, University of Colorado Health Sciences Center, Denver, CO), was digested with EcoRI and XhoI and the Tox176 cDNA was inserted into the EcoRI and SmaI restriction sites of the DREAM vector between the minx intron and the 3' untranslated region derived from the αB-crystallin gene [23]. The injection fragment was generated by KpnI and NotI digestion and was microinjected into individual pronuclei of 1-cell stage FVB/N mouse embryos. Injected embryos were transferred into pseudopregnant ICR strain female mice. Animals were handled following the guidelines provided in the US Public Health Service Policy on Humane Care and Use of Laboratory Animals. The embryos were allowed to develop to term and potential DREAM-Tox176 transgenic mice were identified by isolating genomic DNA from tail biopsies and screening by PCR, using primers specific for the Tox176 sequences: 5'-GAT GTT GTT GAT TCT TCT-3' and 5'-ACG GTT CAG TGA GAC TTA-3'. The PCR cycle conditions were as follows: denaturation at 94 °C for 45 s, annealing at 50 °C for 45 s, and extension at 72 °C for 60 s for 35 cycles.

Histological Analyses

Embryos were obtained by mating of FVB/N females to heterozygous Tox176 transgenic males. Pregnant females were sacrificed at appropriate time points and transgenic offspring were identified by PCR. Heads or eyes of transgenic mice were removed, fixed in 10% formalin, dehydrated, embedded in paraffin, sectioned (5-7 μm) and used for histological analyses, in situ hybridizations and immunohistochemistry.

In situ hybridizations

To analyze expression of the Tox176 transgene, a [³⁵S]UTP-labeled riboprobe specific to the Tox176 sequences of the transgene was generated. The Tox176 antisense riboprobe was synthesized using NcoI-digested Tox176 cDNA and T3 RNA polymerase (Promega, Madison, WI). The Pitx2 and Lmx1b antisense probes were synthesized using XhoI-digested mouse Pitx2 and HindIII-digested Lmx1b cDNAs (from Dr. Randy Johnson, UT MD Anderson Cancer Center, Houston, TX) and T3 RNA polymerase. The BMP4 and BMP7 antisense probes were synthesized using EcoRI-digested mouse BMP4 and HindIII-digested mouse BMP7 cDNAs using SP6 and T7 RNA polymerases, respectively. The N-cadherin antisense probe was synthesized using HindIII-digested N-cadherin cDNA (from Dr. Lixing Reneker, University of Missouri, Columbia, MO) and T7 RNA polymerase. In situ hybridizations were performed using the same hybridization and washing conditions as described previously [24]. The hybridized slides were soaked in Kodak NTB-2 emulsion, dried and exposed for three to seven days at 4 °C. Following development and fixation, the slides were counterstained with hematoxylin. Bright and dark-field images were captured separately using a Nikon Eclipse E600 microscope. Silver grains in the dark field images were pseudo-colored red using ADOBE Photoshop CS and overlaid on corresponding bright-field images.

Immunohistochemistry

Immunohistochemistry on paraffin-embedded tissue sections was performed as follows: slides containing ocular sections were deparaffinized and rehydrated. The antigens were retrieved by microwave treatment in 10 mM sodium citrate buffer (pH 6.0). Following antigen retrieval, the tissue sections were blocked with 10% normal horse serum for 30 min, at room temperature. The slides were then incubated with anti-Pax6 (1:250; Covance, Berkeley, CA), anti-K12 (1:200; Cosmo Bio Company, Tokyo, Japan), anti-E-cadherin (1:1,000; BD Transduction Lab, Franklin Lakes, NJ), anti-β-crystallin (1:400; Santa Cruz Biotechnology, Santa Cruz, CA), anti-TGFβ (1:250; Santa Cruz Biotechnology), anti-phospho-SMAD2 (1:1,000; Chemicon, Billerica, MA), anti-Otx1 (1:1,000; Hybridoma Bank, Iowa City, IA), anti-Opticin (1:500; R&D systems, Minneapolis, MN), anti-active caspase 3 (1:500; R&D Systems), anti-α-smooth muscle action (1:5,000; Sigma, St. Louis, MO), or anti-Z01 (1:750; Invitrogen, Carlsbad, CA) antibody overnight at 4 °C. Following brief washes in PBS, the slides were incubated with the appropriate biotinylated-secondary antibodies: anti-rabbit IgG (1:200; Vector laboratories, Burlingame, CA) for Pax6, K12, β-crystallin, TGF-β, phospho-SMAD2, active caspase 3, and ZO1, anti-mouse IgG (1:200; Vector laboratories) for E-Cadherin, Otx1 and α-smooth muscle actin and anti-goat IgG (1:200; Vector laboratories) for opticin for 30 min at 37 °C. Antigen-antibody complexes were then detected using streptavidin-linked Alexa 594 (Invitrogen) at 1:1000 dilution. Sections were mounted using Prolong fade media containing DAPI (Invitrogen). Images were captured using a Nikon Eclipse E600 microscope.

Electron microscopy

Eyes were isolated from P2 pups and fixed in 3% glutaraldehyde in 0.1 M Sorensen's buffer (20 mM NaH₂PO₄.H₂O, 80 mM Na₂HPO₄, pH 7.4) overnight. After fixation, samples were washed in 0.1 M Sorensen's buffer overnight, post fixed in OsO₄, dehydrated, and embedded in Embed 812 (Electron Microscopy Sciences, Fort Washington, PA). Sections (0.7 μm) were taken and stained with toludine blue. Approximately 80-90 nm thin sections were stained with uranyl acetate in 50% ethanol and Reynolds lead citrate before viewing.

Results

With the goal of ablating all cells of the embryonic lens, we generated transgenic mice that express an attenuated version of diphtheria toxin A, Tox176, driven by a hybrid promoter (DREAM) that is typically active both in the undifferentiated epithelial and differentiated fiber cells of the lens (Figure 1A). This promoter contains the chicken δ-crystallin regulatory enhancer placed upstream from the mouse αA-crystallin promoter followed by a noncoding minx intron (from adenovirus) and the polyA region from αB-crystallin [23]. The DREAM-Tox176 transgene was microinjected and two transgenic founders, OVE1757 and OVE1758 were generated. Stable families were established from these founders. Transgenic mice in both OVE1757 and OVE1758 families showed ocular abnormalities. OVE1757 transgenic mice exhibited microphthalmia (Figure 1B) and OVE1758 mice showed cataracts (data not shown).

To examine transgene expression, in situ hybridizations were performed (Figure 2). Tox176 transcripts were found to be localized to the lens fiber cells in OVE1757 transgenic embryos (Figure 2D,F). Transgene expression was not detected at E11.5, but was seen at E13.5 and E15.5 (Figure 2D,F). Transgene expression was not detectable in the OVE1758 family at E15.5 (data not shown). In the OVE1758 family, changes in lenticular morphology were not seen until P21 (data not shown). Since prenatal lens ablation was not seen in the OVE1758 family, we decided to use family OVE1757 for our developmental studies and only those results are presented here.

To begin to assess the changes in ocular development and morphology, sections of Tox176 transgenic embryos were stained with hematoxylin and eosin (Figure 3). Nontransgenic embryos were used as controls. Ocular abnormalities were not detected at E11.5 (Figure 3A,B). By E13.5, pyknotic nuclei indicative of apoptosis were seen in the lens fiber cells (Figure 3D,F, arrows). At E15.5 and P1, the lenses were smaller (Figure 3H,J,N,P) than the nontransgenic controls and by P8, there was almost a complete loss of the lens (Figure 3T). Activated caspase-3 (Figure 3Z, arrows) was detected in the degenerating lens fiber cells and condensed nuclei were seen in electron micrographs of the Tox176 transgenic lenses (data not shown). Alterations in corneal architecture were apparent by E13.5 (Figure 3D) and were more pronounced by E15.5 (Figure 3L). At E15.5, the Tox176 corneas lacked a differentiated corneal endothelial layer in contrast to nontransgenic corneas (Figure 3, compare L to K, R to Q, V to U). In addition, disorganization of the corneal stroma was seen at E15.5 (Figure 3L, arrow heads). At birth, a disorganized group of mesenchymal cells was seen between the cornea and the lens (Figure 3R, red arrow). At P8, there was a complete absence of the anterior/posterior and vitreal chambers (Figure 3T). Extensive convolutions were seen in the retina of the Tox176 mice at P8 (Figure 3T), although the different layers of the retina were still discernable (Figure 3X).

Alterations in the differentiation programs of the lens and cornea were examined by in situ hybridization and immunohistochemistry (Figure 4A-R). In the lens, expression of Pax6 and E-cadherin is normally restricted to the undifferentiated epithelial cells [25,26]. Expression levels of these markers in the Tox176 transgenic lenses were similar to wild-type at E15.5 (Figure 4A,B,E,F) but were significantly reduced at P1 (Figure 4C,D,G,H) suggesting that the integrity of the lens epithelial cells was compromised by P1. Expression of β-crystallin, a fiber differentiation marker, is initiated at a more anterior location in the Tox176 lenses relative to wild-type (Figure 4I,J, arrowheads) suggesting premature induction of lens fiber differentiation. Expression of Keratin-12 (K12), a marker of corneal epithelial differentiation, was unaltered in the Tox176 transgenic corneas (Figure 4K,L). Expression of Pitx2 and Lmx1b, genes normally expressed in the mesenchymal cells of the corneal stroma, was altered but not lost in the Tox176 transgenic corneas (Figure 4M-P). Due to the increased cell density in the corneal region adjacent to the lens (Figure 2L), expression of Pitx2 and Lmx1b appears elevated in these cells (Figure 4N,P). Expression of N-cadherin, a marker for differentiating corneal endothelial cells [4], was seen both at E15.5 (Figure 4Q,R, arrowheads) and P1 (data not shown) in the transgenic corneas suggesting that the onset of differentiation of the corneal endothelial precursors was not altered.

Corneal endothelial precursors normally undergo a mesenchymal to epithelial transition and develop tight junctions by E18.5 [8,27]. To assess whether mesenchymal cells adjacent to the lens in the Tox176 corneas formed tight junctions, we examined postnatal day 2 (P2) corneas by electron microscopy (Figure 5). In the nontransgenic corneas, flattened corneal endothelial cells were connected to each other by tight and adherens junctions (Figure 5A, arrows). In contrast, in the Tox176 corneas, tight and adherens junctions were not seen (Figure 5B,C). Bundles of collagen fibers could be seen between and around the mesenchymal cells (Figure 5B, arrows) in the posterior part of the Tox176 corneas. In addition, expression of Z01, a critical component of tight junctions, was not detected in the Tox176 corneas (Figure 5E). These results suggest that the corneal endothelial precursors in the Tox176 mice initiate, but fail to complete, the mesenchymal to epithelial transition.

The effects of lens ablation on iris and ciliary development were assessed by histological and immunohistochemical assays (Figure 6). At E15.5, the presumptive ciliary body and iris were histologically distinct at the anterior margins of the Tox176 retinas similar to wild-type (Figure 6A,B). A distinctive iris epithelium was present by P1 in the nontransgenic eyes (Figure 6C), but was missing in the Tox176 eyes (Figure 6D). By P8, ciliary folds had formed in the Tox176 eyes. However, a distinctive iris was still not seen (Figure 6F). Alterations in ciliary body differentiation were examined by analyses of markers Otx1, opticin, BMP4, and BMP7 (Figure 6I-T). Otx1, a transcription factor, is expressed in the presumptive ciliary body and iris and has been shown to be essential for development of these tissues [9]. Opticin, a glycoprotein, is expressed initially in the presumptive ciliary body and iris and later, localized to the ciliary body [28]. The secreted molecules, BMP4 and 7 are expressed in the ciliary body and iris and have been shown to be necessary for proper differentiation of these tissues [10]. With the exception of opticin, expression of all these markers was unaltered in the Tox176 eyes (Figure 6I-T). Normally, opticin expression at E15.5 (Figure 6M) is seen in all the cells of the presumptive ciliary body and iris but at P1 (Figure 6O), is restricted to the inner ciliary epithelium. In the Tox176 eyes, opticin expression is maintained in all the cells of the presumptive ciliary body and iris both at E15.5 and P1 (Figure 6N,P) suggesting that the ciliary body-iris boundary is lost in these mice. However, expression of α-smooth muscle actin, normally expressed in the iris sphincter muscles derived from the iris epithelium [10], was unaltered in the Tox176 eyes (Figure 6U,V, white arrows).

To assess the source of the disorganized group of mesenchymal cells seen in the region between the lens and the cornea in the Tox176 eyes, we mated the Tox176 transgenic mice to C57/BL6 mice to obtain pigmented offspring. Histological analyses of these offspring show that many of the mesenchymal cells in the region anterior to the lens are pigmented (Figure 6H,X, im) suggesting that these cells are abnormally migrating iridial melanocytes (Figure 6G,W). Pigmentation in the outer ciliary epithelium appeared unaffected (Figure 6H,X).

TGFβ2 is normally expressed in lens fibers [29]. As TGFβ2 null mice lack a distinctive corneal endothelium similar to the Tox176 corneas [30], a critical role for TGFβ2 in the initiation and/or progression of the corneal endothelial differentiation program has been proposed. We examined the expression of TGFβ2 and phosphorylation of its downstream signaling effector SMAD2 (pSMAD2) in Tox176 eyes by immunohistochemistry (Figure 7). At E15.5 when the corneal endothelial precursors normally form junctional contacts, TGFβ2 expression was nearly normal in the Tox176 lens fibers (Figure 7B,D). Phosphorylated SMAD2 could also be detected in the nuclei of the mesenchymal cells closest to the lens epithelium at E15.5 (Figure 7F,H) similar to wild-type mice (Figure 7E,G). These results suggest that loss of signaling by TGFβ2 is unlikely to be the cause of the corneal alterations in the transgenic mice.

Discussion

The studies presented here were undertaken to further evaluate the role of the lens in coordinating the differentiation of the tissues in the anterior chamber of the eye. We found that lens ablation, by targeted expression of Tox176, led to defects in differentiation of the corneal endothelium, ciliary body, iris, and the anterior chamber mesenchyme. Early markers for the differentiation of periocular mesenchymal cells including Pitx2, Lmx1b and N-Cadherin were expressed in the Tox176 corneas suggesting that the initial steps in the corneal stromal/endothelial differentiation programs were not altered. However, the corneal endothelial precursors failed to form tight junctions and failed to mature into an epithelium. In addition, abnormal migration of the iridial mesenchymal cells was observed. Our results therefore, support the notion that the embryonic lens provides signals that are needed for the proper differentiation of other tissues in the anterior chamber.

Previous transgenic lens ablation studies focused mostly on the efficacy of DTA- or Tox176-initiated cell ablation and descriptions of changes in ocular morphology. Alterations in fate determination of ocular tissues were not assessed by molecular marker analyses. Though these reports included general descriptions of perturbations in ocular development, specific changes such as absence of tight junctional complexes in the corneal endothelial precursors, abnormal iridial mesenchymal migration and loss of iris-ciliary body boundary formation were not included.

The DREAM promoter we have used for these studies can be active in both the lens epithelial and fiber cells [23]. However, transgene expression in our Tox176 lines was detected only in lens fibers but not in epithelial cells. It is unclear why this is so. It may be due to chromosome position effects at the site of transgene integration. The severity of the ocular phenotype correlated with the level of transgene expression. OVE1757 transgenic lenses displayed a higher level of transgene expression (Figure 2). Although it is theoretically possible that the ocular abnormalities in this line are due to insertional inactivation of an endogenous gene, this is highly unlikely for the following reasons: (1) only ocular development is affected in these mice; (2) the lenticular phenotype correlates well with the known function of the Tox176 transgene [17]; and (3) the ocular phenotype is similar to that reported in earlier studies of transgenic mice with DTA expression in the lens [16-21].

Apoptosis and necrosis were observed in the fiber cells of the Tox176 lenses by E15.5 and lens epithelial markers (Pax6 and E-cadherin) showed significantly reduced expression by P1. We presume that the degeneration of the lens epithelial cells during the first week of birth is due to the progressive loss of healthy fiber cells. An alternative possibility is that as the Tox176 lens fibers degenerate, the smaller lens becomes more surrounded by the anterior margins of the retina. As the signal that induces lens fiber differentiation is secreted by the neuroretina [31], it is possible that this altered architecture results in premature fiber differentiation. The anteriorized β-crystallin expression domain in the Tox176 lenses is consistent with this possibility. In addition, our results are consistent with the results of previous lens ablation studies where the fiber cells were specifically targeted using lens fiber-specific promoters (αA- or γ-crystallin promoters) [17,21]. In these studies the loss of the fibers eventually led to complete ablation of the lens epithelial cells and the lens. The lens fibers however, degenerated over a period of almost two weeks. This could be due to the low metabolic activity of the differentiated lens fibers in contrast to the proliferating epithelial cells.

Transgene expression in the Tox176 mice is lens-specific and is not seen in other ocular tissues. Therefore, the inhibition of corneal endothelial and iris/ciliary body differentiation is not due to ectopic transgene expression. Apoptotic cells were not detected either by electron microscopy or by active caspase 3 immunostaining in non-lenticular tissues of the Tox176 mice (data not shown). Though it is possible that lens fiber degeneration could result in the leakage of DTA protein into the anterior segment, the peptide cannot be internalized without DTB peptide. Therefore, it is unlikely that the alterations in the differentiation programs of the corneal endothelium and iris/ciliary body are due to DTA-induced cytotoxicity.

Histological analyses revealed no dramatic differences between the transgenic and nontransgenic retinas suggesting that retinal differentiation is not altered. However, our results suggest that the lens is critical for the proper differentiation of tissues derived from the anterior margins of the retina. Early markers for differentiation of the ciliary body (Otx1, opticin, BMP4, and BMP7) are expressed in the transgenic eyes. These results suggest that (a) transient lens formation is sufficient to initiate early events in ciliary body differentiation or (b) the ciliary body differentiation program is initiated independent of the lens. At later ages, ciliary folds could be seen in the Tox176 eyes. However, the lack of vitreous and aqueous humor suggests loss of, or abnormal, ciliary function. It is also possible that terminal differentiation of the ciliary body is blocked. If so, it leads to the intriguing possibility that the lens directly promotes ciliary body differentiation. The opticin expression data offer interesting clues about iris differentiation. Opticin expression, normally restricted to the ciliary body by P1, is seen in all the cells of the presumptive ciliary body and iris in the Tox176 eyes. Our results suggest that a signal from the lens is essential for delineation of a border between ciliary body and iris. The identity of this secreted signal from the lens is not known.

Corneal stroma and endothelium have different origins in birds and mammals. These cells are of neural crest origin in birds but are of mesoderm and neural crest origins in mammals [32]. However, lens transplantation experiments performed in chicks are consistent with our lens ablation studies in mice. Removal of the lens in chicks results in the loss of corneal endothelial differentiation but replacement of the lens two days after surgery leads to the formation of the corneal endothelial layer [4]. These results suggest that the inductive influence of the lens on the corneal endothelial differentiation program is conserved across divergent species.

The nature of the signals from the lens to the corneal mesenchyme is also not known. It is likely that there is a secreted signal since the migratory periocular mesenchymal cells are not in contact with the lens epithelium. Of the known signals secreted by the lens, TGFβ2 is of special interest as TGFβ2 is expressed by the lens fibers at E13.5 and at E15.5 [29] and TGFβ2 null mice lack a distinctive corneal endothelium similar to the Tox176 mice [30]. Conditional deletion of TGFβRII in the neural crest derived mesenchymal precursors results in the loss of corneal endothelial differentiation [29]. Therefore TGFβ2 is a critical component of the inductive system that initiates and/or maintains the corneal endothelial differentiation program. Our results suggest that loss of TGFβ2 is unlikely to be the reason for the block in corneal endothelial differentiation in the Tox176 mice. The corneal phenotype could be the result of (a) inhibition of expression of another secreted signal from the lens, (b) reduction in the strength of the lens signal as a result of the posterior displacement of the lens, or (c) abnormal migration of the iridial mesenchymal cells into the space between the lens and the posterior cornea. Our data do not allow us to distinguish between these three possibilities.

Abnormal migration of the iridial mesenchymal cells is seen in the Tox176 eyes. One possible explanation for the abnormal migration of these cells is that the lens plays a direct role in preventing their migration by secretion of an unknown signal. Alternatively, the lack of a proper iris might cause the abnormal migration. However, this second possibility seems less likely. The iris normally differentiates perinatally but the abnormal migration of the mesenchyme is observed as early as E15.5 in the Tox176 eyes. Overall, we believe our results suggest that the embryonic lens produces signals that help specify the distinctive cell fates of the anterior ocular mesenchyme. One signal would be recognized uniquely by the corneal mesenchyme, while another signal would be targeted to the iridial mesenchyme. Since the molecular mechanisms that regulate mesenchymal homing and cell fate determination remain poorly defined, our results suggest that the signaling pathways initiated by the embryonic lens warrant further analysis.

Acknowledgements

The authors thank Dongcai Liang (Baylor College of Medicine, Houston, TX) for performing microinjections to generate the transgenic mice. This research was supported by an NEI grant EY017610 (V.G.), revenue from Nebraska cigarette taxes and the Nebraska Tobacco Settlement Biomedical Research Development Fund (V.G.) and Health Future Foundation (V.G.).

References

1. Chow RL, Lang RA. Early eye development in vertebrates. Annu Rev Cell Dev Biol 2001; 17:255-96.

2. Bard JB, Hay ED. The behavior of fibroblasts from the developing avian cornea. Morphology and movement in situ and in vitro. J Cell Biol 1975; 67:400-18.

3. Bard JB, Hay ED, Meller SM. Formation of the endothelium of the avian cornea: a study of cell movement in vivo. Dev Biol 1975; 42:334-61.

4. Beebe DC, Coats JM. The lens organizes the anterior segment: specification of neural crest cell differentiation in the avian eye. Dev Biol 2000; 220:424-31.

5. Fitch J, Fini ME, Beebe DC, Linsenmayer TF. Collagen type IX and developmentally regulated swelling of the avian primary corneal stroma. Dev Dyn 1998; 212:27-37.

6. Stiemke MM, Edelhauser HF, Geroski DH. The developing corneal endothelium: correlation of morphology, hydration and Na/K ATPase pump site density. Curr Eye Res 1991; 10:145-56.

7. Stiemke MM, McCartney MD, Cantu-Crouch D, Edelhauser HF. Maturation of the corneal endothelial tight junction. Invest Ophthalmol Vis Sci 1991; 32:2757-65.

8. Smith RS, John SW, Nishina PM, Sundberg JP, editors. Systematic evaluation of the mouse eye: anatomy, pathology, and biomethods. Boca Raton (FL): CRC Press; 2002.

9. Acampora D, Mazan S, Avantaggiato V, Barone P, Tuorto F, Lallemand Y, Brulet P, Simeone A. Epilepsy and brain abnormalities in mice lacking the Otx1 gene. Nat Genet 1996; 14:218-22.

10. Zhao S, Chen Q, Hung FC, Overbeek PA. BMP signaling is required for development of the ciliary body. Development 2002; 129:4435-42.

11. Genis-Galvez JM. Role of the lens in the morphogenesis of the iris and cornea. Nature 1966; 210:209-10.

12. Genis-Galvez JM, Santos-Gutierrez L, Rios-Gonzalez A. Causal factors in corneal development: an experimental analysis in the chick embryo. Exp Eye Res 1967; 6:48-56.

13. Thut CJ, Rountree RB, Hwa M, Kingsley DM. A large-scale in situ screen provides molecular evidence for the induction of eye anterior segment structures by the developing lens. Dev Biol 2001; 231:63-76.

14. Kurita R, Sagara H, Aoki Y, Link BA, Arai K, Watanabe S. Suppression of lens growth by alphaA-crystallin promoter-driven expression of diphtheria toxin results in disruption of retinal cell organization in zebrafish. Dev Biol 2003; 255:113-27.

15. Yamamoto Y, Jeffery WR. Central role for the lens in cave fish eye degeneration. Science 2000; 289:631-3. Erratum in: Science 2001; 291:2551.

16. Breitman ML, Bryce DM, Giddens E, Clapoff S, Goring D, Tsui LC, Klintworth GK, Bernstein A. Analysis of lens cell fate and eye morphogenesis in transgenic mice ablated for cells of the lens lineage. Development 1989; 106:457-63.

17. Breitman ML, Rombola H, Maxwell IH, Klintworth GK, Bernstein A. Genetic ablation in transgenic mice with an attenuated diphtheria toxin A gene. Mol Cell Biol 1990; 10:474-9.

18. Harrington L, Klintworth GK, Secor TE, Breitman ML. Developmental analysis of ocular morphogenesis in alpha A-crystallin/diphtheria toxin transgenic mice undergoing ablation of the lens. Dev Biol 1991; 148:508-16.

19. Kaur S, Key B, Stock J, McNeish JD, Akeson R, Potter SS. Targeted ablation of alpha-crystallin-synthesizing cells produces lens-deficient eyes in transgenic mice. Development 1989; 105:613-9.

20. Key B, Liu L, Potter SS, Kaur S, Akeson R. Lens structures exist transiently in development of transgenic mice carrying an alpha-crystallin-diphtheria toxin hybrid gene. Exp Eye Res 1992; 55:357-67.

21. Klein KL, Klintworth GK, Bernstein A, Breitman ML. Embryology and morphology of microphthalmia in transgenic mice expressing a gamma F-crystallin/diphtheria toxin A hybrid gene. Lab Invest 1992; 67:31-41.

22. Saito M, Iwawaki T, Taya C, Yonekawa H, Noda M, Inui Y, Mekada E, Kimata Y, Tsuru A, Kohno K. Diphtheria toxin receptor-mediated conditional and targeted cell ablation in transgenic mice. Nat Biotechnol 2001; 19:746-50.

23. Reneker LW, Chen Q, Bloch A, Xie L, Schuster G, Overbeek PA. Chick delta1-crystallin enhancer influences mouse alphaA-crystallin promoter activity in transgenic mice. Invest Ophthalmol Vis Sci 2004; 45:4083-90.

24. Govindarajan V, Ito M, Makarenkova HP, Lang RA, Overbeek PA. Endogenous and ectopic gland induction by FGF-10. Dev Biol 2000; 225:188-200.

25. Govindarajan V, Harrison WR, Xiao N, Liang D, Overbeek PA. Intracorneal positioning of the lens in Pax6-GAL4/VP16 transgenic mice. Mol Vis 2005; 11:876-86 <http://www.molvis.org/molvis/v11/a104/>.

26. Xu L, Overbeek PA, Reneker LW. Systematic analysis of E-, N- and P-cadherin expression in mouse eye development. Exp Eye Res 2002; 74:753-60.

27. Kidson SH, Kume T, Deng K, Winfrey V, Hogan BL. The forkhead/winged-helix gene, Mf1, is necessary for the normal development of the cornea and formation of the anterior chamber in the mouse eye. Dev Biol 1999; 211:306-22.

28. Ramesh S, Bonshek RE, Bishop PN. Immunolocalisation of opticin in the human eye. Br J Ophthalmol 2004; 88:697-702.

29. Ittner LM, Wurdak H, Schwerdtfeger K, Kunz T, Ille F, Leveen P, Hjalt TA, Suter U, Karlsson S, Hafezi F, Born W, Sommer L. Compound developmental eye disorders following inactivation of TGFbeta signaling in neural-crest stem cells. J Biol 2005; 4:11.

30. Saika S, Saika S, Liu CY, Azhar M, Sanford LP, Doetschman T, Gendron RL, Kao CW, Kao WW. TGFbeta2 in corneal morphogenesis during mouse embryonic development. Dev Biol 2001; 240:419-32.

31. Lovicu FJ, McAvoy JW. Growth factor regulation of lens development. Dev Biol 2005; 280:1-14.

32. Gage PJ, Rhoades W, Prucka SK, Hjalt T. Fate maps of neural crest and mesoderm in the mammalian eye. Invest Ophthalmol Vis Sci 2005; 46:4200-8.