Skip to main content
Journal of Anatomy logoLink to Journal of Anatomy
. 2010 Nov;217(5):488–500. doi: 10.1111/j.1469-7580.2010.01299.x

Development of the cornea of true moles (Talpidae): morphogenesis and expression of PAX6 and cytokeratins

F David Carmona 1, Jingxing Ou 1,2, Rafael Jiménez 3, J Martin Collinson 1
PMCID: PMC3035857  PMID: 20979588

Abstract

Corneal development and structure were studied in the Iberian mole Talpa occidentalis, which has permanently closed eyelids, and the European mole Talpa europaea, in which the eyes are open. The vertebrate cornea typically maintains a three-layered structure – a stratified epithelium with protective and sensory function, an avascular, hypocellular, collagenous stroma, and an endothelium with both barrier and transport functions that regulates corneal hydration, hence maintaining transparency. Compared to mouse, both mole species had significant corneal specializations, but the Iberian mole had the most divergent phenotype, with no endothelium and a flattened monolayer epithelium. Nevertheless, normal epithelial cell junctions were observed and corneal transparency was maintained. Corneas of European moles have a dysmorphic phenotype that recapitulates the human disorder keratoconus for which no mouse model exists. Mole corneas are vascularized – a situation only previously observed in the manatee Trichechus– and have non-radial patterns of corneal innervation indicative of failure of corneal epithelial cell migration. The transcription factor Pax6 is required for corneal epithelial differentiation in mice, but was found to be dispensable in moles, which had mosaic patterns of PAX6 localization uniquely restricted, in European moles, to the apical epithelial cells. The apparently stalled or abnormal differentiation of corneas in adult moles is supported by their superficial similarity to the corneas of embryonic or neonatal mice, and their abnormal expression of cytokeratin-12 and cytokeratin-5. European moles seem to have maintained some barrier/protective function in their corneas. However, Iberian moles show a more significant corneal regression likely related to the permanent eyelid fusion. In this mole species, adaptation to the arid, harder, Southern European soils could have favoured the transfer of these functions to the permanently sealed eyelids.

Keywords: cornea, mole, Pax6, Talpa, vision

Introduction

Fossorial animals adapted to live in the dark frequently evolve severely reduced visual systems in which a wide range of defective eye phenotypes are described (Nevo, 1979; Burda et al. 1990; Němec et al. 2007).

Talpid moles (Eulipotyphla) are insectivores that show striking adaptations to low-light environments (Gorman & Stone, 1990). Despite the fact that they represent valuable animal models of evolutionary adaptation to subterranean life, relatively few studies have been performed on the visual system of this group. Mole eyes exhibit clear signs of anatomical regression, although the main ocular structures are present (Quilliam, 1966; Sato, 1977; Carmona et al. 2008, 2010; Glösmann et al. 2008). The most widely distributed talpid mole is the European mole (Talpa europaea Linnaeus 1758), which has minute eyes open to the world due to the presence of functional eyelids. Although their visual system shows evident degenerative features, such as a nucleated and disorganized lens and photoreceptors with an unconventional morphology (Quilliam, 1966; Glösmann et al. 2008), they can successfully discriminate between dark and light under experimental conditions (Lund & Lund, 1965, 1966; Johannesson-Gross, 1988). On the contrary, other talpid species, for example the Iberian mole (Talpa occidentalis Cabrera 1907) lacks eyelids and their eyes remain covered by a skin layer even in the adult (Dubost, 1968; Nevo, 1979; Kryštufek, 1994; Carmona et al. 2008). However, the skin enclosing the eye is thin and relatively translucent and we have shown in a recent study that Iberian moles have also retained a photo-avoidance response (Carmona et al. 2010).

Very little is known about the molecular and developmental mechanisms underlying the regression of the visual system in mammals. There are in addition many human eye diseases whose aetiology might be better understood by the study of natural models of disease. Examples of such eye abnormalities are corneal vascularization (Klintworth, 1991) and keratoconus (KC), which is a non-inflammatory disease of corneal thinning and ectasia, whose incidence is about 1 in 2000 individuals (Rabinowitz, 1998; Gajecka et al. 2009). Both conditions reduce visual acuity in affected individuals (Klintworth, 1991; Li et al. 2007).

In animals studied to date, corneal development begins very early during embryogenesis, with major developmental changes associated with eyelid opening (see review see Zieske, 2004). In rodents, in which eyelid reopening takes place after birth, the undeveloped corneal epithelium is composed of one or two layers of flattened cells. Just before the opening of the eyelids, the epithelium thickness increases to two to three layers. However, immediately following eyelid reopening, epithelium stratification reaches six to seven cell layers or more, and the basal cells change their shape to a cuboidal, and subsequently a columnar appearance (Chung et al. 1992; Song et al. 2003). Shortly prior to eyelid reopening, the corneal endothelium establishes its barrier function. There is closure of inter-cellular spaces and assembly of Descemet's membrane (Murphy et al. 1984; Bahn et al. 1986; Cintron et al. 1988). The tight junction barrier function of the epithelium is required for control of ion flow and to seal the ocular surface against environmental and pathogenic insult, and that of the endothelium is required to control the water content of the cornea – together they facilitate the maintenance of a high level of stromal collagen order, which is required to maintain transparency (Forrester et al. 2002). In most vertebrates the adult corneal epithelium is heavily innervated by sensory axons that project centripetally and swirl towards the centre of the cornea, recapitulating the presumed homeostatic centripetal migration of the epithelial cells (Collinson et al. 2002; Nagasaki & Zhao, 2003; Leiper et al. 2009).

Because Iberian moles have permanently closed eyes whereas European moles have functional eyelids, corneal morphology and function may differ in these two mole species. We have studied the morphology, ultrastructure, cytoskeleton, innervation, vascularization and gene expression pattern of developing and adult corneas from both Iberian and European moles. Our results show that the cornea is significantly more developed in T. europaea than in T. occidentalis, and that it is vascularized in both species, a quite exceptional trait amongst mammals that has been only described in the Florida manatee (Harper et al. 2005). Moreover, all European moles analysed exhibited a keratoconus-like corneal morphology. We show that moles challenge some of the accepted paradigms of corneal differentiation and that some of the ‘rules’ of corneal structure and function need to be reassessed in light of evolutionary adaptation.

Materials and methods

Material analysed

Twenty-six specimens of T. occidentalis and eight specimens of T. europaea were analysed in the present study. All samples were obtained from wild animals using live traps as described previously (Carmona et al. 2008). Iberian mole captures were performed in poplar plantations located near Santafé and Chauchina (Granada province, Andalusia, Spain), and European moles were captured in pasture lands from Dinnet (Aberdeenshire, Scotland). Radio-tracking of lactating females, which are recognizable by the condition of their teats, was carried out to collect infant and juvenile moles. Embryos and fetuses were obtained from pregnant females after confirming pregnancy by abdominal palpation. Developmental staging from stage 3 (s3) (= E9.5 mouse) was established according to Barrionuevo et al. (2004) and Carmona et al. (2009). Animal handling followed the guidelines and approval of the University of Granada's ‘Ethical Committee for Animal Experimentation’ and the University of Aberdeen Ethical Review Committee. Several mouse samples from the CBA/Ca strain, provided by the Medical Research Facility of the University of Aberdeen, were used as controls.

Immunohistochemistry and histological techniques

Immediately after animal death, mole eyes and embryonic heads were fixed in 4% paraformaldehyde (PFA), embedded in paraffin wax and sectioned following standard procedures. Sections were then de-waxed using histoclear, rehydrated and washed with phosphate-buffered saline (PBS). Periodic acid-Schiff (PAS) staining was carried out with the PAS staining kit – for detection of aldehyde and mucosubstances from Merck (Cat. No. 1.01646). Haematoxylin and eosin staining, immunofluorescence and 3,3′-diaminobenzidine tetrahydrochloride (DAB) staining were performed as described previously (Carmona et al. 2008). Table 1 summarizes the antibodies used in this study. There are no commercial antibodies raised against mole proteins. We have found, from extensive study using over 40 antibodies (this study; Carmona et al. 2008, 2010), that those antibodies that recognize both human and mouse proteins also recognize the mole homologues and, in general, antibodies are selected on this basis.

Table 1.

Primary antibodies used in this study

Gene product Antibody source Working dilution References
β-Tubulin III Rabbit polyclonal, raised against human protein 1/500 Sigma (T2200)
Caveolin Rabbit monoclonal, raised against human protein 1/1000 BD Transduction Laboratories (610059)
Cytokeratin 5/8 Mouse monoclonal, raised against human protein 1/300 Santa Cruz (sc-8021)
Cytokeratin 12 Goat polyclonal, raised against mouse protein 1/300 Santa Cruz (sc-17101)
Desmoplakin Rabbit polyclonal, raised against human protein 1/500 Serotec (AHP320)
PAX6 Mouse monoclonal, raised against chicken protein 1/400 Developmental Studies Hybridoma Bank, University of Iowa (USA)
Type II basic keratins Mouse monoclonal raised against human protein 1/500 Abcam (ab6401) Neomarkers (MS-342)

Corneal flat-mount immunofluorescence

Corneas were dissected out from PFA-fixed eyes and washed with PBS. Corneal epithelia were permeated by partial digestion with 10 mm HCl, 1% pepsin in a 37 °C water bath for 10 min. Samples were subsequently neutralized with 0.1 m boric acid pH 8.5 for 10 min at room temperature (RT) and washed with Tris-buffered saline (12.5 mm Tris pH 7.6, 0.9% NaCl), 0.1% Tween-20 (TBS-T). Corneas were then exposed to primary antibodies in TBS-T, 5% bovine serum (BSA) overnight at 4 °C, washed again with TBS-T and incubated in secondary antibodies diluted in TBS-T, 5% BSA for 2.5 h at RT. Corneas were finally flat-mounted in medium containing DAPI.

Transmission electron microscopy

For ultrastructural analysis, dissected eyes from four adult Iberian moles and three adult European moles (2–3 years old in all cases) were fixed in 2.5% glutaraldehyde in PBS at 4 °C for 3 days. Fixed samples were processed by the Electron Microscopy Facility of the Institute of Medical Sciences, University of Aberdeen. Ultra-thin 80-nm sections were collected onto 200-mesh fine bar copper grids, stained using uranyl acetate and lead citrate, and examined with a Philips CM10 TEM and imaged with a Gatan Bioscan CCD camera.

Results

Corneal morphology in talpid moles with and without eyelids

External examination showed that, despite the permanent eyelid closure in the Iberian mole, European and Iberian moles have eyes of similar size (Fig. 1A–C,E). However, there was a significant difference in the morphology of the cornea between both talpid species. Whereas T. occidentalis exhibited a relatively smooth cornea with the typical uniform curve (Fig. 1F), all the corneas analysed from T. europaea specimens (n = 4) showed a conical shape with a thinner central area (Fig. 1D), which resembles the keratoconus corneal phenotype described in humans (Rabinowitz, 1998).

Fig. 1.

Fig. 1

External morphological features of the eye of T. europaea and T. occidentalis. Unlike T. europaea, which has open eyes (A), the Iberian mole shows permanently enclosed eyes under the skin (B). The minute eye of T. europaea (C) exhibits a cone-shaped cornea (D). In T. occidentalis, the skin enclosing the eye is relatively thin (E) and the cornea has the typical curved shape (F). Scale bar in (B) represents: (A,B) 10 mm, (C,E) 2 mm, and (D,F) 500 μm.

Eyelid morphogenesis and closure during prenatal development in T. occidentalis

Histological analysis and external examination of the Iberian mole eye indicated that eyelid morphogenesis starts early in development (Fig. 2). Eyelids were first observable at stage s4c (arrowheads in Fig. 2), which is equivalent to about E11.5 in mice. Subsequently, a complete closure occurred (as in all mammals studied to date), lasting around 2 days, from stage s5b (18 dpc, equivalent to about E13.5 in mice) to s6 (∼20 dpc, equivalent to mouse E15.5). By the s7 stage (∼22 dpc), the eyes are completely enclosed under the skin, a situation which is maintained during the rest of the animal's life. In contrast to mice, in which the eyelids fuse from the dorsal and ventral sides (Findlater et al. 1993), eyelid closure occurs concentrically in the Iberian mole, in a sphincter-like movement (see arrows in Fig. 2).

Fig. 2.

Fig. 2

Eyelid morphogenesis and fusion in the Iberian mole. First morphological evidence of eyelid development appears at stage s4c (arrowheads) (approximately equivalent to E11.5 in mice). The diameter of the eyelid perimeter is significantly reduced at s5b (equivalent to E13.5). As development progresses, eyelids close uniformly around eyelid circumference (arrow in s5c, equivalent to mouse E14.5), moving across the corneal surface. Eyelid closure process continues until s6, in which a small hole is still seen (arrow, see inset). The junctional region is not centred in relation to the eye position. From s7 (late fetus) to adulthood, the eyes remain permanently enclosed under a uniform layer of epidermal cells. External view of the eye of mole embryos and fetuses are shown in the left columns, and haematoxylin-stained wax sections in the right columns. Scale bar: 500 μm in all images.

Histological organization of the cornea in T. europaea and T. occidentalis

Haematoxylin and eosin staining of tissue sections showed a clear dissimilarity between the corneal histology of T. occidentalis and T. europaea (Fig. 3). The adult Iberian mole presented a clearly undeveloped cornea, very similar to that observed in 2 days post-coitum (dpc) mice (Fig. 3). In both cases, the corneal epithelium was composed of flattened cells and it was one to two cell layers thick, in contrast to the six to seven cell layers observed in the adult mouse (Fig. 3). In contrast, the cornea of T. europaea was thicker than that of T. occidentalis (Fig. 3). The stroma was more disorganized and the corneal epithelium was two to three cell layers thick, with the basal cells showing a cuboidal appearance. Moreover, the corneal epithelium of the European mole exhibited a keratinized eosinophilic superficial layer that was never seen in the Iberian mole or in the mouse.

Fig. 3.

Fig. 3

Haematoxylin and eosin staining on corneal wax sections. (Adult mouse) Normal healthy adult mouse cornea with a multilayered epithelium (composed of six to seven cell layers), a stroma of connective tissue, and a monolayer endothelium. (P2 mouse) Developing cornea from a 2 dpp infant mouse with an undeveloped monolayered epithelium (Talpa occidentalis) The adult Iberian mole cornea is very similar to that of P2 mice. (Talpa europaea) The cornea of the European mole shows an intermediate status. The stroma is thicker and a two to three layer epithelium with cuboidal basal cells is observed. However, the corneal surface is bulged, there is a considerable thinning in the central area, and a keratinized eosinophilic outermost layer is observed. Mole corneas were obtained from adult animals on breeding territory and therefore in at least their second or third calendar years. Scale bar: 50 μm in all images.

PAS staining revealed that Iberian moles lack a posterior basement (Descemet's) membrane (Fig. 4). However, although it was thinner than that in the adult mouse, a Descemet's membrane was present in all adult European moles analysed (arrows in Fig. 4).

Fig. 4.

Fig. 4

Periodic acid-Schiff staining on mice and talpid mole adult corneas. Stain of aldehyde and mucosubstances revealed a posterior basement (Descemet's) membrane in the adult mouse and in T. europaea (arrows). No Descemet's membrane is evident in the cornea of T. occidentalis. Scale bar: 50 μm in all images.

Transmission electron microscopy was performed to investigate corneal ultrastructure in T. occidentalis (n = 4) and T. europaea (n = 2). The Iberian mole showed a well organized stroma with the typical collagen fibril structure and many stromal keratocytes (Fig. 5A). However, no endothelium or Descemet's membrane was observed in any of the samples analysed (Fig. 5B). The corneal epithelial cells had prominent nuclei, and cell–cell borders were attached by desmosomes and adherens junctions. Collagen fibril diameter was not measured. An anterior basement membrane was present, in which hemidesmosomes were also well-developed (Fig. 5C,D). On the other hand, the European mole exhibited a more disorganized stromal histology with considerably fewer keratocytes than those in the Iberian mole (Fig. 5E). Despite being stratified, the corneal epithelium in T. europaea also appeared poorly organized and showed two superficial electron-dense layers. The corneal stroma was present in addition to the anterior basement membrane as well as a thin corneal endothelium and a narrow Descemet's membrane (Fig. 5E–G). Although it was significantly more developed than that of the Iberian mole, the cornea of T. europaea appeared clearly regressed when compared with the wild-type mouse (Fig. 5H,G).

Fig. 5.

Fig. 5

Ultrastructural study of the talpid mole cornea. TEM photomicrographs of adult corneas from T. occidentalis (A–D), T. europaea (E–G) and the mouse (H, I). (A) Full-thickness TEM section of the Iberian mole cornea. (B) The corneal stroma in the Iberian mole shows a well organized collagen fibril structure with no endothelium. (C) The Iberian mole corneal epithelium is composed of flattened cells with prominent nuclei and, sometimes, a second layer of external electron-dense cells (arrow). No anterior limiting lamina is evident. (D) Higher magnification of the Iberian mole corneal epithelium showing smooth cell boundaries in which desmosomes (d), hemidesmosomes (hd) and adherens junctions (aj) are observed. A basement membrane is also observed (black arrow). (E) The European mole cornea exhibits a thicker epithelium and a more disorganized corneal stroma with fewer fibroblasts than that of the Iberian mole. (F) The European mole corneal epithelium contains cuboidal basal cells with round nuclei and it is limited by two superficial electron-dense keratinized layers (arrow) and an interwoven feltwork of fine collagen fibrils of about 1 μm thickness. (G) A corneal endothelium (black arrow) and a Descemet's membrane (white arrow) are present in the cornea of the European mole. (H) The mouse corneal epithelium appeared clearly more developed, with large columnar basal cells lying on a prominent anterior basement membrane (arrow). (I) The murine endothelial cells contain larger nuclei (black arrow) and the Descemet's membrane is significantly thicker than that of the European mole – compare (I) with (G). ep, epithelial cells. Scale bar in (G) represents: (A) 10 μm, (B, C) 2.5 μm, (D) 0.75 μm, (E) 7.5 μm, (F) 5 μm, (G) 1 μm, (H) 1.3 μm, (I) 4 μm.

Corneal vasculature and innervation in talpid moles

The vascular system of adult mole corneas was studied by flat-mount immunofluorescence for the blood vessel marker caveolin 1 (Fig. 6). In healthy wild-type mice, three radially oriented blood vessel branches located in the limbus encircled the cornea, with no other sign of vascularization throughout the cornea (Fig. 6A). However, all the corneas analysed from adult Iberian moles (n = 6) and European moles (n = 3) showed vessel ramifications across the whole cornea (Fig. 6B,C), including the centre areas (Fig. 6D). The blood vessels appeared disorganized and were randomly oriented, with no evident common pattern between the different individuals. Capillaries containing erythrocytes were evident histologically in corneal stromas of young postnatal Iberian moles (stage s11 – 7–12 days old) through to adulthood (Fig. 6E,F). The corneal vasculature network was significantly more profuse in T. occidentalis than in T. europaea.

Fig. 6.

Fig. 6

Corneal vascularization in talpid moles. (A) CAV1 immunostaining on flat-mounted mouse corneas revealed blood vessels only in the limbus but not in the cornea. However, corneal vascularization is clearly observed in both Iberian mole (B) and European moles (C), although it appears more profuse in the Iberian mole. (D) Higher magnification of vascularization at the centre of the cornea in T. occidentalis showing erythrocytes in capillaries. (E) Erythrocytes (autofluorescent red) visualized in capillaries in the cornea of a young Iberian mole 7–12 days old. (F) TEM photomicrograph from an Iberian mole specimen. Arrow points to a vessel containing an erythrocyte. Scale bar in (A) represents: (A) 100 μm, (B,C) 250 μm; (D) 40 μm; (E) 8 μm; (F) 3 μm.

To assess the pattern of innervation in the basal epithelial nerve plexus of the cornea, immunostaining for the neuronal marker β-tubulin III was performed on flat-mounted corneas from adult moles as well as infant and adult mice (Fig. 7). In mice, the corneal nerves appeared disorganized 2 days after birth (Fig. 7A). A week later (10 dpp), the number of axons was considerably increased, although no clear organization was observed yet (Fig. 7B). In the adult mouse, a whorl-like organization with a pattern of centripetal extension from the limbus towards the centre of the cornea was observed (Fig. 7C), as described previously (Yu & Rosenblatt, 2007; Leiper et al. 2009). Corneal innervation of the adult Iberian mole was composed of randomly oriented nerve fibres with no evidence of centripetal projection and swirling pattern (n = 6; Fig. 7D), very similar to that observed in 10 dpp mice (Fig. 7A). On the other hand, the European mole cornea was even less densely innervated than that of the Iberian mole and the fibres appeared more disorganized (n = 2; Fig. 7E), resembling that of P2 mice (Fig. 7A).

Fig. 7.

Fig. 7

Pattern of β-tubulin III-expressing nerve fibres in flat-mounted corneas from mice and talpid moles. (A) Disorganized corneal innervation pattern in a 2 dpp mouse. (B) No clear nerve fibre organization is detected in the cornea of 10 dpp mice yet, but a considerable increase in the number and length of axons is observed. (C) Typical whorl-like centripetal projection pattern of axons from the limbus towards the centre of the cornea in the adult mouse. (D) Corneal innervation in the adult Iberian mole showing no centripetal projection or swirling pattern. (E) The European mole has fewer corneal nerve fibres, which appear even more disorganized. Scale bar in (D) represents: (C) 150 μm, and 75 μm in the rest.

Corneal epithelial cytoskeleton in T. occidentalis

Keratins represent the major cytoskeletal element in the corneal epithelium (Kivela & Uusitalo, 1998; Kinoshita et al. 2001). To analyse the corneal epithelial keratin architecture in the more degenerated epithelium of the Iberian mole, we used antibodies that recognize K12, type II basic keratins (Type II K) and desmoplakin (DPK). DPK is a desmosomal protein directly binding to epidermal Type II K to promote desmosome assembly and reinforcement of membrane attachments (Kouklis et al. 1994). Ou et al. (2010) have previously shown that loss of cell membrane DPK is a feature of the disease state of aniridia-related keratopathy in a mouse Pax6+/− model. Cytokeratin-12 (K12) is the definitive marker for fully differentiated corneal epithelium, and is strongly expressed in mouse corneas (Kurpakus et al. 1990). Immunostaining was performed on wax sections and flat-mounted corneas (Fig. 8). No positive K12 signal was observed in any of the Iberian mole samples analysed (n = 8 from eight different individuals; Fig. 8A, top row). However, developing corneas from s8 stage fetuses, stage s10 infant moles, as well as adult Iberian mole corneas were positive for an antibody which recognizes K5/8 cytokeratins (Fig. 8A, lower rows). Flat-mount immunofluorescence on adult corneas revealed that type II basic keratin filaments were assembled into a dense filamentous network in the cytoplasm of corneal epithelial cells (Fig. 8B). On the other hand, desmoplakin was strongly detected at the cell–cell borders (Fig. 8B).

Fig. 8.

Fig. 8

Cytokeratin and desmoplakin structures in the corneal epithelium of mice and Iberian moles. (A) 3,3′-Diaminobenzidine tetrahydrochloride-immunostaining for K12 and K5/8 in developing and adult mice and Iberian moles. No specific K12 signal was detected in the adult cornea of T. occidentalis. K12 was strongly expressed in the multilayered epithelium of the adult mouse. K5/8-positive immunolabelling was observed in the cornea of Iberian mole fetuses (s8), infant (s10) and adult, as well as in mouse embryos (E18). (B) Flat-mount immunofluorescence for type II basic keratins and desmoplakin on an adult cornea of T. occidentalis. A well assembled network of type II basic keratin filaments is observed (left image). Desmoplakin showed a normal intracellular distribution with stronger specific signal at the cell–cell borders and a weaker signal in the cell body (centre image). Scale bar represents: (A) 150 μm in the mouse adult image and 75 μm in the rest; (B) 10 μm in all images.

PAX6 expression in talpid mole corneas

In most vertebrates Pax6/PAX6 is expressed throughout the ocular surface epithelium during development and into adulthood and has been shown to be required for corneal differentiation (Collinson et al. 2003). We studied PAX6 localization by immunohistochemistry in developing and adult eyes of both European and Iberian moles (Fig. 9). Embryos of stage 5b (18 dpc, approximates to E13.5 in mouse) from both mole species exhibited a normal PAX6 spatial expression pattern in the cornea, with every corneal cell showing similar high PAX6 immunoreactivity (Fig. 9A,B). In contrast, different levels of PAX6 were observed amongst the corneal epithelial cells of Iberian moles from the postnatal stage s10 (5 dpp) until adulthood, with PAX6-negative cells occasionally seen interspersed with PAX6-positive cells (Fig. 9C,G). In European mole eyes, PAX6 was detected in the cornea but also in the nucleated lens (n = 4; Fig. 9D), as described previously for T. occidentalis (Carmona et al. 2008). Nevertheless, the corneal epithelium of adult European moles exhibited a significantly more accentuated mosaic of PAX6 localization than that observed in the Iberian mole. Surprisingly, the most external flattened cells showed the highest PAX6-immunoreactivity, with most of the basal cuboidal cells being PAX6-negative or exhibiting a very weak signal, above all those situated in the centre of the cornea (Fig. 9E). This is the opposite pattern of that observed in adult mice (Fig. 9H).

Fig. 9.

Fig. 9

Immunostaining for PAX6 on developing and adult eyes of T. occidentalis and T. europaea. (A,B) Stage 5b mole embryos (18 dpc) of T. occidentalis (A) and T. europaea (B) exhibit a normal PAX6 spatial expression pattern in the eye. (C) Infant Iberian moles of stage s10 (5 dpp) show mosaic PAX6 expression, with some PAX6-negative cells inserted amongst other epithelial cells with different levels of PAX6-immunosignal (arrowheads in the inset). (D) In the adult European mole, PAX6 is detected in the corneal epithelium as well as in all the nuclei of the cellular lens, including both those of the anterior epithelial cells and those of the posterior lens fibres. (E) Higher magnification of the adult European mole cornea shows a more accentuated mosaic PAX6 expression in the corneal epithelium; surprisingly, the stronger PAX6-specific signal is observed in the scattered outermost flattened epithelial cells, with most basal cuboidal cells showing weak or no immunoreactivity. (F) The opposite situation is observed in wild-type adult mice, in which the strongest signal is detected in the basal cells. (G) The corneal epithelium of the adult Iberian mole does not differ significantly from that of the s10 infant mole. Scale bar in (G) represents: (A,B) 150 μm, (C) 75 μm, (D) 100 μm, (E–G) 50 μm.

Discussion

In this study we have been able to show that the corneal epithelial thickness, stratification and keratinization are reduced in moles, with corneal vascularization, atypical patterns of innervation, and changes in stromal organization. When compared to other sighted mammals, these features are unusual and presumably reflect adaptations to the moles’ fossorial habits. It is interesting that the severity (compared to mouse or man) of the epithelial and endothelial phenotypes correlates in the two mole species with the state of the eyelids (open or sealed) as discussed below. Mole adaptations, however, also challenge some of the conclusions about corneal development and function that have arisen from studies on conventional model organisms. In particular, the corneal endothelium is normally required for corneal transparency, substantially by effecting stromal hypohydration. Understanding the mechanism by which transparency is maintained in Iberian moles in the absence of an endothelium would potentially be a finding of clinical interest.

Vascular and nervous abnormalities in mole corneas

The observation of disorganized irregular patterns of innervation in mole species may be symptomatic of disrupted ocular surface function. In mice and humans, the swirling centripetal pattern of basal epithelial innervation follows and recapitulates the patterns of centripetal corneal epithelial cell migration that begin in early life and continue throughout adulthood (Collinson et al. 2002; Nagasaki & Zhao, 2003; Leiper et al. 2009; Marfurt et al. 2010). These patterns of corneal epithelial cell migration are associated with ocular surface maintenance, probably reflecting the migration of progeny of stem cells residing at the limbus, the boundary between the cornea and conjunctiva (Cotsarelis et al. 1989). The loss of regular innervation patterns in mole species may reflect a loss of the corneal epithelial cell migration that appears to guide sensory axons in mouse and humans. In the Iberian mole, where the cornea is permanently protected by a skin layer and corneal abrasion is thus not likely to occur, it is intuitively apparent that mechanisms for corneal homeostasis may be modified. This hypothesis does not explain, however, the situation in the European mole, for which the cornea is potentially exposed to abrasion, although it is possible that the European mole keeps its eyes shut underground, and that mechanical protection is thus provided by the eyelids. Corneal innervation has two main functions: (i) to maintain a blink reflex and (ii) to support corneal epithelial cells by release of neurotrophic factors such as Substance P, NGF, and BDNF (Baker et al. 1993; Reid et al. 1993; Garcia-Hirschfeld et al. 1994). It is possible that neither of these functions is particularly important in moles, hence reducing the selective pressure to maintain nervous innervation levels equivalent to that of other mammals.

The observation that the corneas of both mole species are vascularized is surprising. The small numbers of immature postnatal moles available for analysis means that we do not currently know exactly when vascularization incursion occurs during early life; however, large blood vessels were observable histologically in Iberian moles 1–2 weeks old. Although corneal vascularization is a normal component of the response to injury and is commonly observed following infection or genetic disease, only the manatee has been shown previously to have vascularized corneas in the uninjured state, and this is associated with low transparency and poor vision (Klintworth, 1991; Harper et al. 2005). The corneas of both mole species are completely transparent. The gatekeeper for corneal vascularization is a soluble isoform of VEGFR1 (s-FLT1), which is absent in corneas of manatees and genetically compromised Pax6+/− and Corn1−/− corneas (Ambati et al. 2006). We were unable to obtain antibodies that recognized VEGFR1 in Talpa, but we speculate that the soluble isoform, s-FLT1, which acts as a dominant negative gatekeeper to bind and sequester corneal VEGF, is absent. S-flt1 is absent in Pax6+/− mouse and PAX6+/− human corneas (Ambati et al. 2006), so the mosaic expression of PAX6 in mole corneas may lead to loss of s-FLT1 and corneal vascularization.

Corneal specification in the Iberian mole

Using mouse chimeras, Collinson et al. (2003) showed that PAX6 was absolutely required for cells to contribute to the ocular surface epithelium during development, suggesting a requirement for PAX6 activity during corneal epithelial differentiation. Whereas PAX6 expression is widely conserved in the eyes of vertebrates (Ashery-Padan & Gruss, 2001), the mosaic pattern of PAX6 in mole corneas clearly indicates that its role in defining the ocular surface epithelium has not been conserved. This could be due to mutations in the regulatory regions of genes required for corneal specification, which render them insensitive to the presence of the PAX6 protein. Alternatively, it is also possible that the mole cornea does not contain a fully differentiated corneal epithelium (compared with other mammals) and does not express a full suite of corneal markers. Evidence from our work supports the second possibility.

It is likely that Type II keratins and DPK interact to integrate the K12 framework, a definitive marker of corneal epithelium in most vertebrates, and regulate cell adhesion and epithelial morphogenesis (Kouklis et al. 1994; Gumbiner, 1996). Our results show that the Iberian mole has a relatively well conserved Type II K organization and DPK was clearly aggregated at the cell–cell junctions, as described previously for wild-type mice (Ou et al. 2010). Our data suggest that K12 is absent in T. occidentalis; however, because its expression in other mammals is restricted solely to the corneal epithelium, we are unable to confirm that the antibody recognizes a mole epitope. However, cytokeratin-5, which in mouse is expressed in the embryonic corneal epithelium but is restricted to the conjunctiva and limbus in adults (Lu et al. 2006), was expressed in the corneal epithelium of the adult Iberian mole. Overall, these data suggest incomplete specification of the Iberian mole corneal epithelium. While we have argued that the corneas of adult moles show features that would be characteristic of immaturity, the expression of cytokeratin-5 in the epithelium, together with reduced stratification and increased stromal vascularization, may argue that there is some degree of conjunctivalization. The presence of goblet cells is a definitive marker of the conjunctival epithelium and incursion of goblet cells onto the corneal surface is taken as clinical evidence of conjunctivalization (Ramaesh et al. 2005). We stained mole corneal epithelia by periodic acid – Schiff reagent staining but found no goblet cells, arguing that full conjunctivalization of the corneas has not occurred.

The lens orchestrates the development of the anterior chamber of the eye (Coulombre & Coulombre, 1964; Ashery-Padan et al. 2000). It is possible that the disruption of postnatal lens development, due to a misregulation of the key genes PAX6 and FOXE3, that we have previously reported for T. occidentalis (Carmona et al. 2008) is responsible for the altered phenotype of the corneal epithelium in moles.

Consequences of eyelid fusion in talpid moles

Unlike the European mole, which has open eyes, the Iberian mole eyelids remain fused during its entire life, making these two closely related taxa very interesting models to study the influence of eyelid opening on cornea development. We predicted that the cornea of Iberian moles would have lost its protective function, and the changes in morphology we observed (monolayered epithelium composed of flattened cells and lack of endothelium and posterior basement membrane) are consistent with that view. The intermediate, less degenerated morphology of the European mole (with a two to three layer thick corneal epithelium containing cuboidal basal cells, and an endothelium which has assembled a thin Descemet's membrane) indicates that this mole species may have retained some protective function against either physical damage or oxidative stress. The eye morphology of both mole species is almost identical, with the most overt differences observed only in the cornea (Quilliam, 1966; Carmona et al. 2008). Hence, the more defective corneal phenotype of the Iberian mole could certainly be the consequence of a failure in the eyelid reopening process. The possible evolutionary advantage that a permanent eyelid closure may confer to most of the mole species of the genus Talpa is discussed below.

The biological significance of eyelid fusion in moles from Southern Europe

The perception of short wavelengths, which are the main components of day-light, may have a significant relevance in the ecology of a mole, as it can alert the animal to possible predator presence or damage in its tunnel network. However, some mole species, including Talpa caeca, Talpa romana, Talpa stankovici and T. occidentalis, show a permanent eyelid fusion even in adulthood (Dubost, 1968; Nevo, 1979; Kryštufek, 1994; Carmona et al. 2008). Nevertheless, the skin overlying the eyes in these species is delicate, thin and without fur; here and in previous publications (Carmona et al. 2008, 2010), we have shown that the Iberian mole eye is functional. It has a transparent cornea, a transparent β-crystallin-containing lens and a laminated retina that has retained all the main cell types for colour vision and circadian sensitivity, with RGC axons projecting contralaterally to the brain through an optic chiasm. The animal retains a photoavoidance response despite permanent eyelid closure. Hence, it seems that the thinness of the epidermis around the eye allows short- to medium-wavelength light to reach the retina and trigger the pathways of visual function.

Permanent eye closure may represent a valuable adaptive trait in subterranean species, as it could be very useful to avoid eye wounds and infections (a hypothesis proposed originally by Darwin in The Origin of Species). The fact that T. europaea has functional eyelids (Quilliam, 1966; Glösmann et al. 2008), whereas other closely related species from Western Europe have not (Dubost, 1968; Nevo, 1979; Kryštufek, 1994; Carmona et al. 2008), could be related to the geographical distribution of these species and their adaptation to different soil conditions. Talpa europaea inhabits a wide area of central and Northern Europe, living mostly in pasture, woodland and gardens. On the other hand, T. occidentalis, T. romana, and T. stankovici occupy the Iberian, Italian and Balkan Mediterranean peninsulas, respectively, and T. caeca has been reported in Southern montane areas of the Western Alps, Apennines and Tracia Mountains (Gorman & Stone, 1990; Mitchell-Jones et al. 1999; Loy et al. 2005). As Mediterranean soils are drier and dustier than those of the humid habitats of T. europaea, mainly during the summer months, permanent eyelid fusion could have conferred an important evolutionary advantage to this group of mole species in Southern Europe, where the ocular surface might be more prone to abrasion.

European moles as possible models for human eye diseases

Diseases affecting the human cornea represent a significant cause of blindness and visual impairment worldwide (Whitcher et al. 2001). In most cases, the current knowledge of their molecular basis is mainly based on the study of transgenic or mutant mice (Cvekl & Tamm, 2004). In this paper, we have reported that European moles exhibit corneas which morphologically resemble the human keratoconus phenotype, and that both T. europaea and T. occidentalis have vascularized corneas, another pathological feature present in many human ocular disorders (Klintworth, 1991). The use of talpid moles as an experimental model for the study of ocular diseases could contribute to understanding their molecular causes. One animal's adaptation is another animal's pathology.

Concluding remarks

The comparison of mouse to European mole to Iberian mole reveals a relatively straightforward progressive loss of corneal structure and barrier function associated, presumably, with a progressive decrease in visual acuity and proportionate increase in the role of the eyelids in protecting the ocular surface. The more general interest of Iberian moles in particular for the field of ocular biology is in their apparent contrast with accepted models of eye development. Although they are dark-adapted animals, they have colour vision and their photoreceptors retain distinctive characteristics of diurnal species (Carmona et al. 2010). Their corneas are transparent with a relatively normal collagen fibril structure in spite of vascularization and the lack of either an endothelium or a properly stratified epithelium – both of which are required for transparency in other species. Furthermore, their lenses avoid cataract in spite of very severe cellular dysgenesis and retention of lens fibre nuclei (Carmona et al. 2008). How moles manage to avoid the blindness apparent in humans with similar cellular corneal and lens defects is unknown, but the explanation could impact on the understanding and management of human disease.

Acknowledgments

The authors would like to thank D. G. Lupiáñez, F. M. Real, J. E. Martín and R.K. Dadhich (University of Granada, Spain) for assistance in capture tasks, and the Andalusian Consejería de Medio Ambiente for capture permits. This work was supported by the Alfonso Martín Escudero Foundation and Junta de Andalucía through Group PAI BIO-109. J.M.C.'s laboratory is funded by Biotechnology and Biological Sciences Research Council (BB/E015840/1).

References

  1. Ambati BK, Nozaki M, Singh N, et al. Corneal avascularity is due to soluble VEGF receptor-1. Nature. 2006;443:993–997. doi: 10.1038/nature05249. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Ashery-Padan R, Gruss P. Pax6 lights-up the way for eye development. Curr Opin Cell Biol. 2001;13:706–714. doi: 10.1016/s0955-0674(00)00274-x. [DOI] [PubMed] [Google Scholar]
  3. Ashery-Padan R, Marquart T, Zhou X, et al. Pax6 activity in the lens primordium is required for lens formation and for correct placement of a single retina in the eye. Genes Dev. 2000;14:2701–2711. doi: 10.1101/gad.184000. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Bahn CF, Glassman RM, Maccallum DK, et al. Postnatal development of corneal endothelium. Invest Ophthalmol Vis Sci. 1986;27:44–51. [PubMed] [Google Scholar]
  5. Baker KS, Anderson SC, Romanowski EG, et al. Trigeminal ganglion neurons affect corneal epithelial phenotype: influence on type VII collagen expression in vitro. Invest Ophthalmol Vis Sci. 1993;34:137–144. [PubMed] [Google Scholar]
  6. Barrionuevo FJ, Zurita F, Burgos M, et al. Developmental stages and growth rate of the mole Talpa occidentalis (Insectivora, Mammalia) J Mammal. 2004;85:120–125. [Google Scholar]
  7. Burda H, Bruns V, Müller M. Sensory adaptations in subterranean mammals. Prog Clin Biol Res. 1990;335:269–293. [PubMed] [Google Scholar]
  8. Carmona FD, Jiménez R, Collinson JM. The molecular basis of defective lens development in the Iberian mole. BMC Biol. 2008;6:44. doi: 10.1186/1741-7007-6-44. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Carmona FD, Lupiáñez D, Martín E, et al. The spatio-temporal pattern of testis organogenesis in mammals. Insights from the mole. Int J Dev Biol. 2009;53:1035–1044. doi: 10.1387/ijdb.072470fc. [DOI] [PubMed] [Google Scholar]
  10. Carmona FD, Glösmann M, Ou J, et al. Retinal development and function in a ‘blind’ mole. Proc R Soc B. 2010;277:1513–1522. doi: 10.1098/rspb.2009.1744. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Chung EH, Bukusoglu G, Zieske JD. Localization of corneal epithelial stem cells in the developing rat. Invest Ophthalmol Vis Sci. 1992;33:2199–2206. [PubMed] [Google Scholar]
  12. Cintron C, Covington HI, Kublin CL, et al. Morphogenesis of rabbit corneal endothelium. Curr Eye Res. 1988;7:913–929. doi: 10.3109/02713688808997248. [DOI] [PubMed] [Google Scholar]
  13. Collinson JM, Morris L, Reid AI, et al. Clonal analysis of patterns of growth, stem cell activity, and cell movement during the development and maintenance of the murine corneal epithelium. Dev Dyn. 2002;224:432–440. doi: 10.1002/dvdy.10124. [DOI] [PubMed] [Google Scholar]
  14. Collinson JM, Quinn JC, Hill RE, et al. The roles of Pax6 in the cornea, retina, and olfactory epithelium of the developing mouse embryo. Dev Biol. 2003;255:303–312. doi: 10.1016/s0012-1606(02)00095-7. [DOI] [PubMed] [Google Scholar]
  15. Cotsarelis G, Cheng S-Z, Dong G, et al. Existence of slow-cycling limbal epithelial basal cells that can be preferentially stimulated to proliferate: implications on epithelial stem cells. Cell. 1989;57:201–209. doi: 10.1016/0092-8674(89)90958-6. [DOI] [PubMed] [Google Scholar]
  16. Coulombre AJ, Coulombre JL. Lens development. I. Role of the lens in eye growth. J Exp Zool. 1964;156:39–48. doi: 10.1002/jez.1401560104. [DOI] [PubMed] [Google Scholar]
  17. Cvekl A, Tamm ER. Anterior eye development and ocular mesenchyme: new insights from mouse models and human diseases. Bioessays. 2004;26:374–386. doi: 10.1002/bies.20009. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Dubost G. Les mammifères souterrains. Rev Ecol Biol Sol. 1968;5:99–197. [Google Scholar]
  19. Findlater GS, McDougall RD, Kaufman MH. Eyelid development, fusion and subsequent reopening in the mouse. J Anat. 1993;183:121–129. [PMC free article] [PubMed] [Google Scholar]
  20. Forrester JV, Dick AD, McMenamin PG, et al. The Eye: Basic Sciences in Practice. Edinburgh: Saunders; 2002. [Google Scholar]
  21. Gajecka M, Radhakrishna U, Winters D, et al. Localization of a gene for keratoconus to a 5.6-Mb interval on 13q32. Invest Ophthalmol Vis Sci. 2009;50:1531–1539. doi: 10.1167/iovs.08-2173. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Garcia-Hirschfeld J, Lopez-Briones LG, Belmonte C. Neurotrophic influences on corneal epithelial cells. Exp Eye Res. 1994;59:597–605. doi: 10.1006/exer.1994.1145. [DOI] [PubMed] [Google Scholar]
  23. Glösmann M, Steiner M, Peichl L, et al. Cone photoreceptors and potential UV vision in a subterranean insectivore, the European mole. J Vis. 2008;8(4) doi: 10.1167/8.4.23. [DOI] [PubMed] [Google Scholar]
  24. Gorman ML, Stone RD. The Natural History of Moles. Ithaca, NY: Cornell University Press; 1990. [Google Scholar]
  25. Gumbiner BM. Cell adhesion: the molecular basis of tissue architecture and morphogenesis. Cell. 1996;84:345–357. doi: 10.1016/s0092-8674(00)81279-9. [DOI] [PubMed] [Google Scholar]
  26. Harper JY, Samuelson DA, Reep RL. Corneal vascularization in the Florida manatee (Trichechus manatus latirostris) and three-dimensional reconstruction of vessels. Vet Ophthalmol. 2005;8:89–99. doi: 10.1111/j.1463-5224.2005.00351.x. [DOI] [PubMed] [Google Scholar]
  27. Johannesson-Gross K. Brightness discrimination of the mole (Talpa europaea L.) in learning experiments applying a modified tube-maze method. Z Saugetierkd. 1988;53:193–201. [Google Scholar]
  28. Kinoshita S, Adachi W, Sotozono C, et al. Characteristics of the human ocular surface epithelium. Prog Retin Eye Res. 2001;20:639–673. doi: 10.1016/s1350-9462(01)00007-6. [DOI] [PubMed] [Google Scholar]
  29. Kivela T, Uusitalo M. Structure, development and function of cytoskeletal elements in non-neuronal cells of the human eye. Prog Retin Eye Res. 1998;17:385–428. doi: 10.1016/s1350-9462(98)00001-9. [DOI] [PubMed] [Google Scholar]
  30. Klintworth GK. Corneal Angiogenesis – A Comprehensive Clinical Review. New York: Springer; 1991. [Google Scholar]
  31. Kouklis PD, Hutton E, Fuchs E. Making a connection: direct binding between keratin intermediate filaments and desmosomal proteins. J Cell Biol. 1994;127:1049–1060. doi: 10.1083/jcb.127.4.1049. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Kryštufek B. The taxonomy of blind moles (Talpa caeca and T. stankovici, Insectivora, Mammalia) from south-east Europe. Bonn Zool Beitr Bd. 1994;45:1–16. [Google Scholar]
  33. Kurpakus MA, Stock EL, Jones JC. Expression of the 55-kD/64-kD corneal keratins in ocular surface epithelium. Invest Ophthalmol Vis Sci. 1990;31:448–456. [PubMed] [Google Scholar]
  34. Leiper LJ, Ou J, Walczysko P, et al. Control of patterns of corneal innervation by Pax6. Invest Ophthalmol Vis Sci. 2009;50:1122–1128. doi: 10.1167/iovs.08-2812. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Li X, Yang H, Rabinowitz YS. Longitudinal study of keratoconus progression. Exp Eye Res. 2007;85:502–507. doi: 10.1016/j.exer.2007.06.016. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Loy A, Colangelo P, Annesi F, et al. Origin and evolution of Western European moles (genus Talpa, Insectivora): a multidisciplinary approach. Mammal Study. 2005;30:S13–S17. [Google Scholar]
  37. Lu H, Zimek A, Chen J, et al. Keratin 5 knockout mice reveal plasticity of keratin expression in the corneal epithelium. Eur J Cell Biol. 2006;85:803–811. doi: 10.1016/j.ejcb.2006.04.001. [DOI] [PubMed] [Google Scholar]
  38. Lund RD, Lund JS. The visual system of the mole Talpa europaea. Exp Neurol. 1965;13:302–316. doi: 10.1016/0014-4886(65)90118-4. [DOI] [PubMed] [Google Scholar]
  39. Lund RD, Lund JS. The central visual pathways and their functional significance in the mole (Talpa europaea) J Zool. 1966;149:95–101. [Google Scholar]
  40. Marfurt CF, Cox J, Deek S, et al. Anatomy of the human corneal innervation. Exp Eye Res. 2010;90:478–492. doi: 10.1016/j.exer.2009.12.010. [DOI] [PubMed] [Google Scholar]
  41. Mitchell-Jones AJ, Amori G, Bogdanowicz W, et al., editors. The Atlas of European Mammals. London: Academic Press; 1999. [Google Scholar]
  42. Murphy C, Alvarado J, Juster R, et al. Prenatal and postnatal cellularity of the human corneal endothelium. A quantitative histologic study. Invest Ophthalmol Vis Sci. 1984;25:312–322. [PubMed] [Google Scholar]
  43. Nagasaki T, Zhao J. Centripetal movement of corneal epithelial cells in the normal adult mouse. Invest Ophthalmol Vis Sci. 2003;44:558–566. doi: 10.1167/iovs.02-0705. [DOI] [PubMed] [Google Scholar]
  44. Němec P, Cveková P, Burda H, et al. Visual systems and the role of vision in subterranean rodents: diversity of retinal properties and visual system designs. In: Begall S, Burda H, Schleich CE, editors. Subterranean Rodents: News from Underground. Berlin: Springer; 2007. pp. 129–160. [Google Scholar]
  45. Nevo E. Adaptive convergence and divergence of subterranean mammals. Annu Rev Ecol Syst. 1979;10:269–308. [Google Scholar]
  46. Ou J, Lowes C, Collinson JM. Cytoskeletal and cell adhesion defects in wounded and Pax6+/− corneal epithelia. Invest Ophthalmol Vis Sci. 2010;51:1415–1423. doi: 10.1167/iovs.09-4023. [DOI] [PubMed] [Google Scholar]
  47. Quilliam AT. The problem of vision in the ecology of Talpa europaea. Exp Eye Res. 1966;5:63–78. doi: 10.1016/s0014-4835(66)80022-2. [DOI] [PubMed] [Google Scholar]
  48. Rabinowitz YS. Keratoconus. Surv Ophthalmol. 1998;42:297–319. doi: 10.1016/s0039-6257(97)00119-7. [DOI] [PubMed] [Google Scholar]
  49. Ramaesh T, West JD, Ramaesh K, et al. Developmental and cellular factors underlying corneal epithelial dysgenesis in the Pax6+/− mouse model of aniridia. Exp Eye Res. 2005;81:224–235. doi: 10.1016/j.exer.2005.02.002. [DOI] [PubMed] [Google Scholar]
  50. Reid TW, Murphy CJ, Iwahashi CK, et al. Stimulation of epithelial cell growth by the neuropeptide substance P. J Cell Biochem. 1993;52:476–485. doi: 10.1002/jcb.240520411. [DOI] [PubMed] [Google Scholar]
  51. Sato Y. Comparative morphology of the visual system of some Japanese species of Soricoidea (Superfamily) in relation to life habits. J Hirnforsch. 1977;18:531–546. [PubMed] [Google Scholar]
  52. Song J, Lee YG, Houston J, et al. Neonatal corneal stromal development in the normal and lumican-deficient mouse. Invest Ophthalmol Vis Sci. 2003;44:548–557. doi: 10.1167/iovs.02-0592. [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Whitcher JP, Srinivasan M, Upadhyay MP. Corneal blindness: a global perspective. Bull World Health Org. 2001;79:214–221. [PMC free article] [PubMed] [Google Scholar]
  54. Yu CQ, Rosenblatt MI. Transgenic corneal neurofluorescence in mice: a new model for in vivo investigation of nerve structure and regeneration. Invest Ophthalmol Vis Sci. 2007;48:1535–1542. doi: 10.1167/iovs.06-1192. [DOI] [PubMed] [Google Scholar]
  55. Zieske JD. Corneal development associated with eyelid opening. Int J Dev Biol. 2004;48:903–911. doi: 10.1387/ijdb.041860jz. [DOI] [PubMed] [Google Scholar]

Articles from Journal of Anatomy are provided here courtesy of Anatomical Society of Great Britain and Ireland

RESOURCES