Human Molecular Genetics, 2000, Vol. 9, No. 7 1021-1032
© 2000 Oxford University Press
Haploinsufficiency of the transcription factors FOXC1 and FOXC2 results in aberrant ocular development
1The Howard Hughes Medical Institute and 2The Jackson Laboratory, Bar Harbor, ME, USA, 3Department of Cell Biology, Vanderbilt University Medical Center, Nashville, TN, USA, 4Department of Anatomy and Cell Biology, University of Cape Town, Observatory 7925, Cape Town, South Africa, Departments of 5Pediatrics and 6Ophthalmology, University of Iowa, Iowa City, IA, USA and 7Department of Ophthalmology, Tufts University School of Medicine, Boston, MA, USA
Received 22 November 1999; Revised and Accepted 1 February 2000.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Anterior segment developmental disorders, including Axenfeld–Rieger anomaly (ARA), variably associate with harmfully elevated intraocular pressure (IOP), which causes glaucoma. Clinically observed dysgenesis does not correlate with IOP, however, and the etiology of glaucoma development is not understood. The forkhead transcription factor genes Foxc1 (formerly Mf1) and Foxc2 (formerly Mfh1) are expressed in the mesenchyme from which the ocular drainage structures derive. Mutations in the human homolog of Foxc1, FKHL7, cause dominant anterior segment defects and glaucoma in various families. We show that Foxc1+/– mice have anterior segment abnormalities similar to those reported in human patients. These abnormalities include small or absent Schlemm’s canal, aberrantly developed trabecular meshwork, iris hypoplasia, severely eccentric pupils and displaced Schwalbe’s line. The penetrance of clinically obvious abnormalities varies with genetic background. In some affected eyes, collagen bundles were half normal diameter, or collagen and elastic tissue were very sparse. Thus, abnormalities in extracellular matrix synthesis or organization may contribute to development of the ocular phenotypes. Despite the abnormalities in ocular drainage structures in Foxc1+/– mice, IOP was normal in almost all mice analyzed, on all genetic backgrounds and at all ages. Similar abnormalities were found in Foxc2+/– mice, but no disease-associated mutations were identified in the human homolog FKHL14 in 32 ARA patients. Foxc1+/– and Foxc2+/– mice are useful models for studying anterior segment development and its anomalies, and may allow identification of genes that interact with Foxc1 and Foxc2 (or FKHL7 and FKHL14) to produce a phenotype with elevated IOP and glaucoma.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
The glaucomas are blinding diseases that involve death of retinal ganglion cells, atrophy of the optic nerve, and typically are associated with elevated intraocular pressure (IOP) (1). Some glaucomas are associated with developmental anomalies of the anterior chamber that would be expected to result in decreased drainage of aqueous humor, leading to IOP elevation. For example, congenital glaucoma and glaucoma associated with iridogoniodysgenesis anomaly involve maldevelopment of the irido-corneal angle through which the aqueous humor must pass before exiting the eye (2–5). These abnormalities may result from a primary defect in the migration and/or differentiation of the mesenchymal cells that contribute to the anterior segment of the eye, including the corneal stroma, iris stroma and trabecular meshwork (TM) (a structure in the aqueous humor drainage pathway) (6–8). Other anterior segment developmental disorders associated with glaucoma that appear to involve defective differentiation of head mesenchyme include Axenfeld–Rieger anomaly (ARA) and its syndromic forms involving skeletal and dental dysgenesis (9), and iris hypoplasia (10). The presence or absence of specific developmental anomalies varies considerably between patients, and it is becoming evident that the disorders mentioned above represent part of a spectrum of disease phenotypes (11–16). The variability of phenotypes may result from: (i) different mutations in the same gene(s); (ii) interaction between specific mutations and genetic background; (iii) stochastic developmental events that modify the phenotype of specific mutations; and (iv) mutations in different genes of related function. The severity of clinically observable dysgenesis does not correlate well with IOP (9,11,17), and the etiology of glaucoma development is not well defined, highlighting the need to identify causative genes. The generation and study of mouse models with mutations in these genes will help in understanding the etiology of these glaucomas and in identifying modifier genes.
Forkhead transcription factors have a characteristic forkhead or winged helix DNA-binding domain and are required for various developmental processes (18). The highly related murine genes Foxc1 (forkhead box C1, formerly Mf1) and Foxc2 (formerly Mfh1) encode two such transcription factors that have almost identical DNA-binding domains and similar embryonic expression patterns (19–21). Sites of expression of Foxc1 and Foxc2 include endothelial cells and mesenchyme cells of the head (19,21–23). To study the in vivo functions of FOXC1 and FOXC2, null alleles of their genes were generated (21,23,24). Foxc1 homozygous mutants (Foxc1–/–) die perinatally with hemorrhagic hydrocephalus, most likely as a result of abnormal development of the arachnoid through which cerebrospinal fluid must drain (21). The spontaneous mouse mutation causing congenital hydrocephalus (ch) is a truncating allele of Foxc1 (21,25). In addition to the hydrocephaly, homozygous Foxc1–/– and Foxc1ch/ch mutants have severe skeletal, cardiovascular and ocular abnormalities, including absent anterior chamber and open eyelids at birth (21,25–27). Foxc2–/– null mutants die prenatally or perinatally with multiple skeletal and cardiovascular defects (23,24), but no detailed analysis of eye development has been performed in these mutants.
Due to the severe ocular phenotypes in Foxc1–/– homozygotes, the expression of Foxc1 in periocular mesenchyme cells that give rise to ocular drainage structures and the association of a chromosomal region including the equivalent human gene [Forkhead, Drosophila, homolog-like 7 (FKHL7)] with several dominant glaucoma-related phenotypes (28–31), it seemed reasonable that various ocular tissues, including the drainage structures, may not develop normally in Foxc1 heterozygotes. To test this, and to determine whether haploinsufficiency of Foxc1 causes elevated IOP and glaucoma, we have performed clinical, physiological, histological and electron microscopic analyses of the eyes of Foxc1 heterozygotes (Foxc1+/– and Foxc1ch/+). Since this study began, mutations in FKHL7 have been identified in patients with various glaucoma-related phenotypes (13,32).
Our clinical studies show that both Foxc1+/– and Foxc1ch/+ mice have multiple, clinically obvious abnormalities, including eccentric irregularly shaped pupils and displaced Schwalbe’s line. (Schwalbe’s line is the anatomical location where Descemet’s membrane, which lines the posterior corneal surface, thickens at its termination.) The penetrance of these abnormalities depends on genetic background. Similar clinical findings were reported recently in Foxc1ch/+ mice on the CHMU/Le genetic background (25). Our histological and ultrastructural studies show for the first time that Foxc1+/– and Foxc1ch/+ heterozygotes also have abnormal aqueous humor drainage structures, including small or absent Schlemm’s canal (SC) (the aqueous humor drainage channel) and abnormal TM. However, the mice do not have elevated IOP on the genetic backgrounds studied. Thus, the Foxc1 mutant mice represent a useful model to understand human anterior segment anomalies, and they may allow identification of genes that interact with Foxc1 (or FKHL7) to give a more severe phenotype resulting in elevated IOP and glaucoma. Additionally, we evaluated for the first time the role of Foxc2 in anterior segment development and show that heterozygous mice have similar defects to those in Foxc1+/– mice. Moreover, the more severe iris abnormalities in double heterozygotes (Foxc1+/– Foxc2+/–) compared with strain-matched Foxc1+/– and Foxc2+/– single heterozygotes, and the presence of ciliary body abnormalities in all double heterozygotes but only in one Foxc2+/– single heterozygote, suggest that both FOXC1 and FOXC2 cooperatively may activate genes important for anterior segment development. Finally, we performed sequence analysis on FKHL14 (the human equivalent of Foxc2) in families with ARA, but found no disease-associated changes.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Table 1 summarizes the different genetic backgrounds and genotypes studied, and various phenotype penetrances on these backgrounds.
|
Clinical examinations
Initial slit-lamp and ophthalmoscopic analysis revealed no obvious abnormalities in 3-month-old Foxc1+/– heterozygotes on a pure 129/SvEvTac (129) background (n = 23) (Fig. 1). However, at 11 months of age, two of eight heterozygotes on this background had obvious abnormalities. The left eyes of both affected mice had abnormally shaped eccentric pupils, and a focal region of the peripheral cornea of the right eye of one mouse was opaque and resembled sclera (scleralization). To determine whether genetic background can modify the mutant phenotype, the Foxc1 mutation was crossed to different mouse strains. One of twenty Foxc1+/– mice on a mixed (129 x Black Swiss) (129BS) background had abnormal processes extending from the iris (iris strands) and an abnormal pupil that was both irregular in shape and eccentrically located. The eyes of all heterozygotes appeared normal when the mutation was backcrossed for two generations to Mus castaneous/Ei (CAST) (n = 17). On the other hand, multiple abnormalities frequently were evident as the mutation was moved to a C57BL/6J (B6) background (Fig. 1). The severity and combination of abnormalities in affected B6 heterozygous Foxc1 eyes varied both between and within animals. Abnormalities included irregular pupils that had a normal location, pupils that were severely misplaced, irregularly shaped pupils with iris tears and iris strands that frequently were attached to the cornea, displaced and/or enlarged Schwalbe’s line and scleralization of the peripheral cornea. After backcrossing to strain B6 for two generations (N2), 4 of 12 Foxc1+/– mice had such abnormalities. At N3 and higher backcross generations, most (20/21) heterozygous animals were affected, whereas no abnormalities were seen in Foxc1+/+ mice (0/16). Of the 24 affected B6 mice studied, 17 had anomalies in both eyes, whereas seven had clinically detectable abnormalities in only one eye. Similar findings were present in Foxc1ch/+ heterozygotes on the CHMU/Le background (Fig. 1). Of 51 heterozygotes analyzed on this background, 48 had clinically detectable anomalies (40 bilateral and 8 unilateral), whereas 7 of 7 Foxc1+/+ were phenotypically normal. The CHMU/Le heterozygotes tended to have more severe iris abnormalities and irido-corneal adhesions than those in the B6 mice. The irides of a few Foxc1ch/+ mice had very large pupils that resembled the phenotype associated with aniridia. Although we did not analyze many mice at multiple ages, the phenotype sometimes became worse with age on both B6 and CHMU/Le backgrounds. (However, no mice that appeared normal at a young age showed abnormalities at an older age.) Pupils that were located centrally and appeared normal or mildy irregular at 6–8 weeks of age were sometimes severely irregular and very eccentric when examined again at 6–8 months. No sex differences in phenotype were observed.
|
Histological analyses
Histological examinations revealed abnormalities in the eyes of Foxc1 mutant heterozygotes on all genetic backgrounds analyzed (129, B6, 129BS, CAST and CHMU/Le). The severity of lesions varied within individual eyes and between eyes. In most eyes, the irido-corneal angle had morphologically normal and abnormal regions. Although we analyzed many sections from different locations within each eye (see Materials and Methods), these samples still only represent a relatively small portion of the total eye, so it is not possible to quantitate the relative proportions of normal and abnormal areas in whole eyes. Abnormalities of the irido-corneal angle included small or absent SC, large blood vessels and iris strands, hypoplastic or absent TM, and TM that appeared compressed. In one Foxc1+/– heterozygote (with a 129BS background), SC was absent in all analyzed sections. Morphometric analysis comparing the width of the inner wall of the SC of Foxc1+/– mice on the pure 129 background (see Materials and Methods) confirmed that the SC was significantly smaller than normal (mean ± SEM: Foxc1+/– 22.2 ± 4.1 µm, n = 6 eyes; Foxc1+/+ 61.2 ± 5.9 µm, n = 5; ANOVA P = <0.0001; range of values for individual mice in each group: Foxc1+/– 6.8–35.9 µm, Foxc1+/+ 43.4–75.9 µm). Some areas that lacked a TM were occupied by cells (Fig. 2) that resembled mesenchymal precursor cells that normally give rise to the TM (33, unpublished data). All analyzed eyes from Foxc1+/–, and Foxc1ch/+ heterozygotes exhibited developmental dysgenesis, whereas all eyes from non-carrier control mice matched for age and strain were normal. Some Foxc1ch/+ eyes had a focally hypoplastic ciliary body with short and thin ciliary processes. The ciliary processes were sometimes aberrantly located, such as at the posterior end of the pars plana. In contrast, no obvious ciliary body anomalies were identified in Foxc1+/– heterozygotes on any genetic background analyzed. The restriction of ciliary body abnormalities to heterozygotes on the CHMU/Le background and the generally more severe clinical phenotypes of these mice suggests that this background is relatively susceptible to some anterior segment developmental phenotypes. Although both Foxc1– and Foxc1ch are believed to be null alleles, our experiments do not distinguish between differential effects of these mutations or genetic background. Overall, there were no obvious differences in the severity of drainage structure malformations between the various genetic backgrounds studied.
|
Since Foxc2 is highly related to Foxc1 and is also expressed in periocular mesenchyme (unpublished data), we tested whether Foxc2 participates in anterior segment development. Foxc2+/– mice (129BS) were found to have similar abnormalities to Foxc1+/– heterozygotes (Fig. 2), whereas Foxc2+/+ littermates did not. Again, there was variability in the severity of abnormalities within and between eyes. In addition to the similar irido-corneal angle phenotypes, both Foxc1 and Foxc2 heterozygotes exhibited comparable hypoplastic iris development with abnormally thin iris stroma and iris pigment epithelium (Fig. 3).
|
The phenotypic similarities between Foxc1+/– and Foxc2+/– heterozygotes prompted us to investigate overlapping functions of both gene products during anterior segment development by analyzing double heterozygotes (Foxc1+/–Foxc2+/–). All comparisons between single (n = 7 mice for both Foxc1 and Foxc2) and double (n = 6) heterozygotes were made on the 129BS background. Double heterozygotes exhibited the same range and intermittent pattern of angle abnormalities as each class of single heterozygotes. Interestingly, however, the iris of double heterozygotes was more severely affected than in either single heterozygote. The mesenchyme-derived iris stroma was almost completely lacking in some areas (Fig. 3). Additionally, all Foxc1+/–Foxc2+/– mice had intermittent ciliary body abnormalities that were not present in single heterozygotes, except for one Foxc2+/– eye (Fig. 2). Some phenotypes were specific to double heterozygotes. All Foxc1+/–Foxc2+/– mice had corneal vascularization (Fig. 3) and some had open eyelids at birth (data not shown). Corneal ulcers or immune cell infiltrates (Fig. 2) were present in some of the double heterozygotes and were probably a consequence of corneal damage in young pups whose corneas were not protected by their open eyelids.
Gene expression
To identify genes that may be affected by a reduction in forkhead gene expression, we used in situ hybridization to assess the expression in Foxc1–/– eyes of various genes that previously were implicated in anterior segment or neural crest development. We limited these initial studies to Foxc1 mutants since the similar phenotypes resulting from FOXC1 or FOXC2 deficiency, and the almost identical DNA-binding domains of these proteins suggested that FOXC1 and FOXC2 activate a number of common genes during the normal development of this ocular region. The genes we analyzed encode paired-like homeodomain transcription factor 2 (PITX2) (34,35), LIM homeobox transcription factor 1B (LMX1B) (36,37), platelet-derived growth factor 
(PDGFA) and platelet-derived growth factor receptor, polypeptide (PDGFRA) (38,39). As has been reported for the developing cornea (26), we found no obvious expression differences in the embryonic irido-corneal angle between wild-type and Foxc1–/– mice for any of these genes (data not shown).
Electron microscopy
To investigate further the development of the irido-corneal angle, we analyzed mutant eyes by electron microscopy. Since the phenotypes of Foxc1 and Foxc2 heterozygous mutant mice are very similar, we limited these studies to Foxc1+/– and wild-type mice. Heterozygous but not wild-type mice had a number of abnormalities that varied between eyes. These studies confirmed the presence of both relatively normal and abnormal regions within individual heterozygous eyes. In normal regions, the TM consisted of robust trabecular beams with abundant organized collagen and elastic tissue cores that were lined with endothelial-like trabecular cells. These beams were separated by inter-trabecular spaces through which the aqueous humor passes as it drains from the eye. In these unaffected regions, SC also had a normal appearance and was lined with endothelial cells. In contrast, in other regions, the TM was either missing or had developed abnormally. There was sometimes a general paucity of extracellular matrix (ECM) components, including organized collagen bundles and elastic tissue (Fig. 4, compare c and d). In some of these areas, the cells lining the anterior chamber did not resemble any adult cells usually associated with this region and may therefore represent precursor cells that failed to complete differentiation. In other areas, the TM next to the anterior chamber had a normal appearance, but just internal to the SC canal it lacked inter-trabecular spaces and was dense and continuous (Fig. 4e and f). In many places where the TM was abnormal, the SC had a generally normal appearance but had very few giant vacuoles (structures that transmit aqueous humor into the lumen of the canal). In other places, SC was completely absent (Fig. 4j).
|
Intraocular pressure
To assess the relationship between these morphological abnormalities and IOP, we compared the IOPs of Foxc1+/– and Foxc1ch/+ mice with those of their wild-type age- and strain-matched counterparts. We initially assessed the IOPs of young mice. No significant differences in IOP were detected in 6- to 8-week-old B6 N2 mice (Foxc1+/– 12.3 ± 0.4 mmHg, n = 10; Foxc1+/+ 13.5 ± 0.5 mmHg, n = 10; ANOVA P = 0.08) or 8- to 12-week-old CAST N2 mice (Foxc1+/– 14.2 ± 0.4 mmHg, n = 14; Foxc1+/+ 14.7 ± 0.4 mmHg, n = 17; ANOVA P = 0.4). We tested whether IOP increased with age by analyzing 7.5-month-old B6 N4 animals. Again, we found no significant difference between heterozygous and wild-type mice (Foxc1+/– 12.6 ± 0.4 mmHg, n = 11; Foxc1+/+ 13.6 ± 0.4 mmHg, n = 11; ANOVA P = 0.3). Due to the variability in phenotype between eyes, we reasoned that analyzing many mice might identify some eyes with elevated IOP. Histological analysis of these eyes might then provide valuable insight into the association in human eyes between specific anterior segment malformations and high or normal IOP. We therefore determined the IOPs of many normal and heterozygous mice at different ages. This included 59 Foxc1+/– B6
N3 mice ranging from 3 to 8 months old, and 60 Foxc1ch/+ CHMU/Le mice ranging from 2 to 18 months old (11 were 14 months old). Of all of the analyzed mice, only a single 17-month-old Foxc1ch/+ had elevated IOP (18.6 mmHg, which was >2 SDs above the normal CHMU/Le mean of 11.2 ± 0.7 mmHg). Six additional 17- or 18-month-old Foxc1ch/+ mice had normal IOPs that were all <12 mmHg. The developmental aberrations in Foxc1+/– and Foxc1ch/+ mice, therefore, rarely appear to cause elevated IOP on any genetic background studied.
Other abnormalities
Cataracts were present in a few Foxc1 heterozygous mice on the B6 and CHMU/Le backgrounds. Corneal opacities occurred in ~25% of Foxc1ch/+ mice of the CHMU/Le strain and their incidence increased with age. Corneal vascularization was also present in some old Foxc1ch/+ heterozygotes. Due to the pleiotropic effects resulting from the absence of FOXC1, we assessed whether heterozygosity for Foxc1 mutations affects tissues other than the anterior segment. With the exception of one Foxc1ch/+ eye, no posterior segment or optic nerve abnormalities were identified in any of the analyzed mice on the 129, B6, CAST and CHMU/Le backgrounds. In this eye, retinal areas with completely normal photoreceptors adjoined regions that completely lacked photoreceptor cell bodies, inner segments and outer segments. 129BS mice with or without any Foxc1 and Foxc2 mutations had a severe global retinal degeneration that confounded analysis of the effects of these forkhead gene mutations on this background.
Analysis of FKHL14
We analyzed the human homolog of Foxc2 (FKHL14) to assess its potential role in human ARA. FKHL14 consists of a single exon mapping to chromosome 16q22–24 (20,40). The FKHL14 gene was analyzed in 32 probands with ARA not associated with mutations in other genes that are known to cause these conditions. The complete coding region was sequenced but no frameshift, premature termination or protein-altering changes were found.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Multiple effects of FOXC1 and FOXC2 deficiency on ocular drainage structure development
We provide evidence here that mutations in either Foxc1 or Foxc2 result in dominantly inherited developmental dysgeneses of the mouse irido-corneal angle that are very similar to those reported in human patients with ARA and congenital glaucoma (2,3,9). Some patients with ARA, congenital glaucoma or iris hypoplasia have mutations in the human homolog of Foxc1 (FKHL7) (13,32). The Foxc1 and Foxc2 mutant mice are, therefore, valuable models for studying anterior segment development.
The ocular abnormalities found in these mice are variable both within and between eyes, with the existence of both normal and abnormal regions within affected eyes. The SC was often small and had fewer giant vacuoles than were present typically in normal regions of the same mutant eye or in wild-type eyes. The TM did not appear to complete development or was completely missing at some locations. The paucity of TM tissue at some locations may reflect abnormal migration of mesenchymal cells into the angle or abnormal retention of these cells at these locations. Sometimes the corneo-scleral transition zone (where the TM and SC usually reside) appeared more like sclera, or pigmented cells were present in this normally extensively pigment-free region, suggesting development of multipotent mesenchyme cells along inappropriate paths.
The SC consists of an endothelium surrounded by a basal lamina, and seems to develop from a small intrascleral venous plexus (41–44). It is proposed that the mammalian TM, iris stroma and corneal stroma derive from neural crest cells (6,9,45). However, lineage-tracing studies to demonstrate conclusively the origin of various ocular structures have not been conducted in mammals. In all of the mutant mice, SC malformations were only present in regions where the TM was also abnormal, and at locations where TM was absent there was typically no SC. At many locations with aberrant TM, small endothelial-lined channels were present in the general region where SC should be located, whereas in other regions where the TM had started to form but had apparently stalled SC often appeared normal. This suggests that the presumptive TM cells interact with the precursors of the SC vessels or the developing vessels themselves, and that developmental changes in these future TM cells are important in directing the appropriate growth and/or trajectory of the SC vessels.
Abnormalities in FOXC1 mutant mice are not sufficient to cause elevated IOP and glaucoma
Developmental disorders of the anterior chamber are variably associated with elevated IOP and glaucoma. For example, approximately half of the patients with ARA develop elevated IOP and glaucoma (11). The age at onset of this glaucoma is variable, and elevated IOP often manifests during childhood or young adulthood as opposed to infancy. A significant issue for our understanding of glaucoma in these conditions is identifying the abnormalities that lead or predispose to elevated IOP.
In spite of the abnormalities in aqueous humor drainage structures in Foxc1 mutant mice, the IOP was normal in almost all analyzed mice, on all genetic backgrounds and at all ages. There are several possible explanations for this observation. First, there may still be sufficient normal tissue within each eye to maintain normal IOP. The presence of normal tissue may also explain the absence of glaucoma in many patients with ARA. A second possibility is that abnormalities in ocular drainage structures and/or TM and SC cell metabolism, caused by mutations in genes such as FKHL7, predispose affected human eyes to the IOP-elevating effects of environmental factors or aging. The mice that we analyzed may not have been exposed to these factors or may have had genetic backgrounds that were resistant to their effects. Glaucoma may manifest itself on strain backgrounds that we did not investigate. Further studies are needed to evaluate this possibility, and may lead to the identification of modifier genes that interact with Foxc1 or Foxc2 during angle development. Several other genes including CYP1B1 (46), PITX2 (34), PITX3 (47), Lmx1b (37) and PAX6 (48) cause anterior segment abnormalities, ARA or congenital glaucoma in humans or mice. It is not known whether any of the developmental pathways affected by these genes overlap or interact with each other or with FOXC1- and FOXC2-activated pathways during anterior segment formation. Crosses between mice with mutations in these genes will be valuable in addressing these issues.
Ciliary body abnormalities may influence IOP
The ciliary body produces the aqueous humor by a combination of plasma filtration and active transport (49). The ciliary epithelium secretes antioxidant enzymes into the aqueous humor that may protect the TM from damage that could elevate IOP. There is also evidence that it acts as a neuroendocrine tissue in IOP regulation by secreting various molecules that may have a paracrine effect on the drainage tissues (50). Focal ciliary body malformations were present in double heterozygotes, some Foxc1ch/+ mice and one Foxc2+/– mouse, although it is not known whether they affected aqueous humor production. It is possible that there are more widespread metabolic abnormalities in ciliary body tissues that may affect the produced volume or composition of aqueous humor. Metabolic abnormalities in ciliary body, TM or SC cells may gradually affect the composition of the ECM through which the aqueous humor drains. The ECM is believed to be an important determinant of the resistance to aqueous humor outflow from the eye (51–53). Depending on the nature of these abnormalities, they could act to either increase or decrease IOP. If such abnormalities occur in human patients, they could affect IOP and contribute to glaucoma. Thus, the combination of both ciliary body and drainage angle phenotypes may contribute to variability in the age of onset or presence of IOP elevation. The study of mouse mutants affecting anterior segment development will help to address this possibility.
No mutations that alter FKHL14 were found in ARA patients
The malformations we observed in FOXC2-deficient mice prompted us to examine the complete coding region of the Foxc2 homolog, FKHL14, in human patients with ARA. No frameshift, premature termination or protein-altering changes were identified, suggesting that FKHL14 is not a common cause of ARA. However, we cannot rule out the presence of mutations in non-analyzed regions that may alter gene expression. It may even be possible that deficiency of FOXC2 is not the sole cause of the mouse phenotypes, since a selectable marker promoter remains in the mutant Foxc2 gene (23) and may interfere with expression of the neighboring forkhead gene Fkh6 (40). However, in situ hybridization studies showed no difference in the expression of Fkh6 RNA between Foxc2–/– embryos and wild-type controls (23).
FOXC1 and FOXC2 may activate common downstream genes cooperatively
Foxc1 and Foxc2 are co-expressed extensively in developing mouse tissues, and the encoded proteins have almost identical DNA-binding domains (19–23,54). The very similar malformations of ocular drainage structures in Foxc1+/–, Foxc2+/– and Foxc1+/–Foxc2+/– mice suggest that the two genes function in the same cells as components of two signaling pathways converging on a common gene(s). The normal expression of Foxc1 in Foxc2–/– mutants, and of Foxc2 in Foxc1–/– mice (27), argues against a direct action of FOXC2 on Foxc1 and vice versa. Similarly, in situ hybridization revealed no obvious expression differences of Foxc1 in the developing eyes of Foxc2+/– mice and Foxc1+/–Foxc2+/– double heterozygotes or of Foxc2 in Foxc1+/– mice and double heterozygotes (unpublished data). The ciliary body abnormalities and open eyelids at birth in double but not single heterozygotes, and the more severe iris phenotype in double heterozygotes compared with each single heterozygote, revealed a genetic interaction between these genes in some ocular and periocular tissues.
Recent studies show that this interaction extends to the developing cardiovascular system. Double but not single heterozygotes have severe cardiovascular anomalies similar to those in each homozygous mutant (27). This suggests that the overall level of both FOXC1 and FOXC2 is important for the normal development of these tissues. It seems likely that important target genes are not activated appropriately if this level falls below a critical threshold (as in double heterozygotes and homozygous mutants). Various models for the genetic interaction of Foxc1 and Foxc2 were discussed by Winnier et al. (27).
FOXC1 and FOXC2 have some unique functions in anterior segment development
Double heterozygotes do not show all the abnormalities found in Foxc1 homozygous mutants, suggesting that Foxc1 and Foxc2 do not have completely equivalent functions. For example, Foxc1–/– mice do not form a corneal endothelium or anterior chamber (26) whereas double heterozygotes do. This suggests that, in addition to any overlapping functions, FOXC1 and FOXC2 may activate some unique gene(s), may activate some gene(s) to different levels or may both be required to activate some gene(s) appropriately during ocular development. These possibilities can be explored as more is learned about the catalog of genes involved in anterior chamber morphogenesis. Foxc2 and Pax1 (a paired-box transcriptional activator) act synergistically during vertebral column development, and Foxc2 Pax1 double heterozygotes have severe vertebral column abnormalities and spina bifida (55). Various Pax family genes are important for ocular development (48,56,57), and it is possible that forkhead genes cooperate with Pax genes during anterior segment morphogenesis.
Stochastic developmental events result in phenotypic variability
With current knowledge, any attempts to explain the intermittent anomalies of the drainage angle and ciliary body within affected eyes are speculative. The phenotypic variability, even in inbred strains (129 and CHMU/Le), probably reflects stochastic developmental events related to the timing and level of expression of important developmental control genes. It is possible that decreased steady-state levels of FOXC1 or FOXC2 decrease the probability that an important downstream regulator(s) of development is activated correctly at the appropriate time, and in the correct location. Phenotypic variability may result from correct activation of this regulator in some groups of cells and incorrect or no activation in groups of cells at separate locations. Monoallelic expression also may contribute to phenotypic variability. Interestingly, Pax5 is expressed from only one allele at specific stages during B cell development. The allele-specific expression of Pax5 is stochastic, reversible and independent of parental origin of the alleles. This allele-specific expression was proposed as a common mechanism contributing to the dominant (haploinsufficient) developmental defects caused by Pax mutations (58). Monoallelic expression of IL4 and Ly49 family genes is also under stochastic developmental control (59,60). The focal abnormalities in Foxc1+/– and Foxc2+/– eyes may be explained if the normal and mutant alleles undergo stochastic developmental expression with predominant expression of the mutant gene among cells occupying future abnormal regions. Further experiments are needed to test this.
ECM abnormalities may contribute to various phenotypes in Foxc1 and Foxc2 mutants
The abnormalities in Foxc1 and Foxc2 mutant mice may involve abnormal cell–cell adhesion, cell migration and/or delayed or interrupted differentiation of mesenchymal cells in the irido-corneal drainage angle, including abnormal production of ECM material. The TM of the wild-type mouse has a complex ECM including various collagens, proteoglycans, fibronectin, laminin, elastin and vitronectin (53,61). The collagen and elastic tissue in trabecular beams normally have a characteristic organized structure when viewed by electron microscopy. In contrast, the ECM components of Foxc1+/– trabecular beams are often disorganized, and in some locations there is very little collagen and elastic tissue. In other places, the collagen bundles are approximately half normal diameter. ECM abnormalities have been implicated in other phenotypes caused by Foxc1 deficiency. The ECM in the developing sternocostal cartilages, sternum and kidneys of Foxc1ch/ch mutants is disorganized and there is a deficiency of collagen fibrils and proteoglycans (62–65). This suggests that Foxc1 regulates the synthesis and/or organization of ECM, and that the ECM abnormalities lead to or exacerbate the developmental defects by altering cellular behavior (66–68). Due to the very similar overall phenotype in Foxc2+/– mice, this also may be true for Foxc2. Further experiments are needed to determine whether these forkhead transcription factors directly or indirectly alter ECM composition and structure. Alternatively, but not exclusively, these transcription factors may regulate the expression of cell surface receptors that participate in cell–matrix and cell–cell interactions affecting both cell migration and ECM composition/assembly. Future efforts to understand pathways of anterior segment differentiation will include the identification of genes acting both upstream and downstream of Foxc1 and Foxc2.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Investigators conducting all analyses were unaware of the genotypes of the mice. Throughout the text, Foxc1– refers to the Foxc1lacz allele (21), Foxc1ch refers to the congenital hydrocephalus allele (21,69) and Foxc2– refers to the Foxc2tm1blh allele (23). All of these mutations are believed to result in null alleles.
Animal husbandry
All experiments were performed in compliance with the ARVO statement for use of animals in ophthalmic and vision research. 129BS mice were bred at Vanderbilt University Medical Center and transferred to the Jackson Laboratory. All other mice were bred and maintained at the Jackson Laboratory. Mice were housed in cages containing white pine bedding and covered with polyester filters. The environment was kept at 21°C with a 14 h light–10 h dark cycle. Mice were fed NIH31 (6% fat) chow ad libitum, and their water was acidified to pH 2.8–3.2. The B6, CAST and CHMU/Le strains were obtained from the Jackson Laboratory whereas the 129/SvEvTac strain was obtained from Taconic Farms (Germantown, NY). The colony was monitored for specific pathogens by the Jackson Laboratory’s routine surveillance program (http://www.jax.org for specific pathogens ).
Clinical examinations
Anterior chambers were examined with a slit-lamp biomicroscope. Slit-lamp photographs were taken using a 40x objective lens. An indirect ophthalmoscope and a 60 or 90 diopter lens was used to visualize the retinas and optic nerves. For this analysis, pupils were dilated with a drop of 1% cyclopentolate (70).
Histological analysis
Eyes were fixed, processed and then embedded in Historesin (Leica, Heidelberg, Germany) as previously reported (71,72). Eyes were sectioned sagittaly at 1.5 µm thickness. Initially, 24–56 sections from each of 3–5 different ocular locations were collected, stained with hematoxylin and eosin, and analyzed for each eye. The lens was used as a landmark in collecting these sections. Collected regions included both temporal and nasal lens peripheries, central lens and regions on each side of the lens center that were halfway towards the lens periphery. The sections were stained with hematoxylin and eosin and the pathological changes were assessed by analyzing these sections. Due to the potential for artifacts in the delicate tissues analyzed, conclusions were drawn from only high quality sections, and abnormalities had to be present in multiple sections from the same region to be regarded as real. The following numbers of eyes were analyzed for each genotype on each genetic background: 129 Foxc1+/–, eight; 129 Foxc1+/+, eight; B6 Foxc1+/–, six; B6 Foxc1+/+, eight; CAST Foxc1+/–, three; CAST Foxc1+/+, two; 129BS Foxc1+/–, seven; 129BS Foxc2+/–, seven; 129BS Foxc1+/–Foxc2+/–, six; 129BS+/+, five; CHMU/Le Foxc1ch/+, eight, CHMU/Le Foxc1+/+, four. Typically, only one eye from an individual mouse was analyzed. For morphological analysis of SC, the extent of the inner wall was traced on a captured image at a magnification of ~800x and measured using a Quantimet 600 Image Analysis system (Leica). The canal on each side of the first five high quality sections with little obvious distortion was measured at each of the first three collected ocular positions. Thus, 30 measurements (2 x 5 x 3) were made for each eye. The average of these measurements was used for statistical comparisons. To avoid subjective interpretations of developmental stages, any small vessels (endothelial lined channels) that were located where SC should be were counted as SC.
In situ hybridization
Whole heads (14.5 days post-coitum) or eyes (17.5 days post-coitum) were fixed in 4% paraformaldehyde in phosphate buffer overnight, washed in phosphate buffer and serially dehydrated into 100% methanol. Tissues were embedded in paraffin and sectioned at 7 µm thickness. Section in situ hybridization was carried out as previously described (26,73,74).
Electron microscopy
Mice were euthanized and eyes immediately were enucleated and fixed with 0.8% paraformaldehyde and 1.2% glutaraldehyde in 0.1 M phosphate buffer pH 7.2 at 4°C. The eyes were fixed for 1 h and the anterior segment removed and cut into 1 x 2 mm blocks that included cornea, iris, TM and ciliary body. Fixation was continued at 4°C for 12 h and the tissues were washed in phosphate buffer and then post-fixed with 1% osmium tetroxide, dehydrated and embedded in epon–araldite resin (75). Sections of 1 µm were cut for orientation; thin sections were stained with uranyl acetate and lead citrate (76).
Intraocular pressure
IOPs were measured as previously described (77). The pressure was recorded for a 1 min period following stabilization after the microcannula entered the eye. Mice of different genotypes were included in each measurement period, as were C57BL/6J mice. The IOPs of C57BL/6J mice are very consistent over time (unpublished data), and so these animals were interspersed with experimental mice to ensure that calibration had not drifted and that the system was functioning optimally.
Analysis of FKHL14
The majority of the probands were of Caucasian descent and had ARA. A sample of whole blood (5–10 ml) was obtained from probands following informed consent using protocols approved by the Instutional Review Board at the University of Iowa. The entire coding region was sequenced. The coding region was amplified as four fragments with primer pairs, 5F/6R (fragment 1), 6F/7R (fragment 2), 8F/2R (fragment 3) and 10F/4R (fragment 4). Fragment 1 was sequenced using primers 5F and 6R. Fragment 2 was sequenced using primers 6F and 7R. Fragment 3 was sequenced using primers 2R, 8F, 8R and 9F. Fragment 4 was sequenced using primers 3F, 3R, 4R and 10F. PCR and sequencing conditions were reported previously (13). The primer sequences in 5'
3' orientation were:
2R, CCTCCTTCTCCTTGGACACGT;
3F, AGGAGGAGCGGGCCCACCTCA;
3R, GCAGCGCGCTCTCGGGGCTCA;
4R, GCACCACCAGGCCCGCGCGTC;
5F, CCGCCGGGCGGAGAGCTGAGC;
6F, GAGCGGCCGGCGCACATGTGC;
6R, TGTGGCCGGGGAGGTGGTTCA;
7R, GGGAGGTCCCGGGACACGTCA;
8F, CGCGCCCTCCAGGATGCCGAT;
8R, GTCCTTAGGCGCCGCGGGCTG;
9F, GGCCGCTCCTACGCGCCCTAC; and
10F, CGCTCAACGAGTGCTTCGTCA.
|
ACKNOWLEDGEMENTS |
|---|
We thank Val Sheffield for his assistance in analyzing the FKHL14 gene, Felicia Farley for help with references, Jennifer Smith and Joyce Worcester for help with figures, and Michael Anderson, Alexander Chervonsky and Derry Roopenian for critical reading of the manuscript. Supported in part by a grant from the Glaucoma Research Foundation (D.Y.N.) and CA34196. S.W.M.J. is an Assistant Investigator and B.L.M.H. is an Investigator of the Howard Hughes Medical Institute.
|
FOOTNOTES |
|---|
+ To whom correspondence should be addressed at: The Howard Hughes Medical Institute, The Jackson Laboratory, 600 Main Street, Bar Harbor, ME 04609, USA. Tel: +1 207 288 6496; Fax: +1 207 288 6079; Email: swmj@jax.org
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
1 Ritch, R., Shields, M.B. and Krupin, T. (1996) Classification of the glaucomas. In Ritch, R., Shields, M.B. and Krupin, T. (eds), The Glaucomas, Clinical Science, Vol. 2. Mosby Year Book, St Louis, MO, pp. 717–725.
2 Tawara, A. and Inomata, H. (1981) Developmental immaturity of the trabecular meshwork in congenital glaucoma. Am. J. Ophthalmol., 92, 508–525.[ISI][Medline]
3 Anderson, D.R. (1981) The development of the trabecular meshwork and its abnormality in primary infantile glaucoma. Trans. Am. Ophthalmol. Soc., 79, 458–485.[Medline]
4 Jerndal, T. (1972) Dominant goniodysgenesis with late congenital glaucoma: a re-examination of Berg’s pedigree. Am. J. Ophthalmol., 74, 28–33.[ISI][Medline]
5 Weatherill, J.R. and Hart, C.T. (1969) Familial hypoplasia of the iris stroma associated with glaucoma. Br. J. Ophthalmol., 53, 433–438.
6 Kupfer, C. and Kaiser-Kupfer, M.I. (1978) New hypothesis of developmental anomalies of the anterior chamber associated with glaucoma. Trans. Ophthalmol. Soc. UK, 98, 213–215.[ISI][Medline]
7 Kupfer, C. and Kaiser-Kupfer, M.I. (1979) Observations on the development of the anterior chamber angle with reference to the pathogenesis of congenital glaucomas. Am. J. Ophthalmol., 88, 424–426.[ISI][Medline]
8 Tripathi, B.J., Tripathi, R.C. and Wisdom, J.E. (1996) Embryology of the anterior segment of the human eye. In Ritch, R., Shields, M.B. and Krupin, T. (eds), The Glaucomas, Vol. 1. Mosby Year Book, St Louis, MO, pp. 3–38.
9 Shields, M.B. (1983) Axenfeld–Rieger syndrome: a theory of mechanism and distinctions from the iridocorneal endothelial syndrome. Trans. Am. Ophthalmol. Soc., 81, 736–784.[Medline]
10 Heon, E., Sheth, B.P., Kalenak, J.W., Sunden, S.L., Streb, L.M., Taylor, C.M., Alward, W.L., Sheffield, V.C. and Stone, E.M. (1995) Linkage of autosomal dominant iris hypoplasia to the region of the Rieger syndrome locus (4q25). Hum. Mol. Genet., 4, 1435–1439.
11 Shields, M.B., Buckley, E., Klintworth, G.K. and Thresher, R. (1985) Axenfeld–Rieger syndrome. A spectrum of developmental disorders. Surv. Ophthalmol., 29, 387–409.[ISI][Medline]
12 Waring, G.O., Rodrigues, M.M. and Laibson, P.R. (1975) Anterior chamber cleavage syndrome. A stepladder classification. Surv. Ophthalmol., 20, 3–27.[Medline]
13 Nishimura, D.Y., Swiderski, R.E., Alward, W.L.M., Searby, C.C., Patil, S.R., Benner, S.R., Kanis, A.B., Gastier, J.M., Stone, E.M. and Sheffield, V.C. (1998) The forkhead transcription factor gene FKHL7 is responsible for glaucoma phenotypes which map to 6p25. Nature Genet., 19, 140–147.[ISI][Medline]
14 Kulak, S.C., Kozlowski, K., Semina, E.V., Pearce, W.G. and Walter, M.A. (1998) Mutation in the RIEG1 gene in patients with iridogoniodysgenesis syndrome. Hum. Mol. Genet., 7, 1113–1117.
15 Alward, W.L.M., Semina, E.V., Kalenak, J.W., Heon, E., Sheth, B.P., Stone, E.M. and Murray, J.C. (1998) Autosomal dominant iris hypoplasia is caused by a mutation in the Reiger syndrome (RIEG/PITX2) gene. Am. J. Ophthalmol., 125, 98–100.[ISI][Medline]
16 Semina, E.V., Reiter, R.S. and Murray, J.C. (1997) Isolation of a new homeobox gene belonging to the Pitx/Rieg family: expression during lens development and mapping to the aphakia region on mouse chromosome 19. Hum. Mol. Genet., 6, 2109–2116.
17 Pearce, W.G. and Kerr, C.B. (1965) Inherited variation in Rieger’s malformation. Br. J. Ophthalmol., 49, 530–537.
18 Kaufmann, E. and Knöchel, W. (1996) Five years on the wings of fork head. Mech. Dev., 57, 3–20.[ISI][Medline]
19 Hiemisch, H., Schutz, G. and Kaestner, K.H. (1998) The mouse Fkh1/Mf1 gene: cDNA sequence, chromosomal localization and expression in adult tissues. Gene, 220, 77–82.[ISI][Medline]
20 Miura, N., Iida, K., Kakinuma, H., Yang, X.L. and Sugiyama, T. (1997) Isolation of the mouse (MFH-1) and human (FKHL 14) mesenchyme fork head-1 genes reveals conservation of their gene and protein structures. Genomics, 41, 489–492.[ISI][Medline]
21 Kume, T., Deng, K.Y., Winfrey, V., Gould, D.B., Walter, M.A. and Hogan, B.L. (1998) The forkhead/winged helix gene Mf1 is disrupted in the pleiotropic mouse mutation congenital hydrocephalus. Cell, 93, 985–996.[ISI][Medline]
22 Hiemisch, H., Monaghan, A.P., Schutz, G. and Kaestner, K.H. (1998) Expression of the mouse Fkh1/Mf1 and Mfh1 genes in late gestation embryos is restricted to mesoderm derivatives. Mech. Dev., 73, 129–132.[ISI][Medline]
23 Winnier, G.E., Hargett, L. and Hogan, B.L. (1997) The winged helix transcription factor MFH1 is required for proliferation and patterning of paraxial mesoderm in the mouse embryo. Genes Dev., 11, 926–940.
24 Iida, K., Koseki, H., Kakinuma, H., Kato, N., Mizutani Koseki, Y., Ohuchi, H., Yoshioka, H., Noji, S., Kawamura, K., Kataoka, Y. et al. (1997) Essential roles of the winged helix transcription factor MFH-1 in aortic arch patterning and skeletogenesis. Development, 124, 4627–4638.[Abstract]
25 Hong, H.K., Lass, J.H. and Chakravarti, A. (1999) Pleiotropic skeletal and ocular phenotypes of the mouse mutation congenital hydrocephalus (ch/Mf1) arise from a winged helix/forkhead transcription factor gene. Hum. Mol. Genet., 8, 625–637.
26 Kidson, S.H., Kume, T., Deng, K.Y., Winfrey, V. and Hogan, B.L.M. (1999) 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., 211, 306–322.[ISI][Medline]
27 Winnier, G.E., Kume, T., Deng, K.Y., Rogers, R., Bundy, J., Raines, C., Walter, M.A., Hogan, B.L.M. and Conway, S.J. (1999) Roles for the winged helix transcription factors MF1 and MFH1 in cardiovascular development revealed by nonallelic noncomplementation of null alleles. Dev. Biol., 213, 418–431.[ISI][Medline]
28 Mirzayans, F., Mears, A.J., Guo, S.W., Pearce, W.G. and Walter, M.A. (1997) Identification of the human chromosomal region containing the iridogoniodysgenesis anomaly locus by genomic-mismatch scanning. Am. J. Hum. Genet., 61, 111–119.[ISI][Medline]
29 Gould, D.B., Mears, A.J., Pearce, W.G. and Walter, M.A. (1997) Autosomal dominant Axenfeld–Rieger anomaly maps to 6p25. Am. J. Hum. Genet., 61, 765–768.[ISI][Medline]
30 Morissette, J., Falardeau, P., Dubois, S., Bergeron, J., Vonck, P., Cote, G., Anctil, A., Amyot, M., Blondeau, P., Bergeron, E. et al. (1997) A common gene for developmental and familial open-angle glaucomas confined on chromosone 6p25. Am. J. Hum. Genet., 61, A286.
31 Jordan, T., Ebenezer, N., Manners, R., McGill, J. and Bhattacharya, S. (1997) Familial glaucoma iridogoniodysplasia maps to a 6p25 region implicated in primary congenital glaucoma and iridogoniodysgenesis anomaly. Am. J. Hum. Genet., 61, 882–888.[ISI][Medline]
32 Mears, A.J., Jordan, T., Mirzayans, F., Dubois, S., Kume, T., Parlee, M., Ritch, R., Koop, B., Kuo, W.L., Collins, C. et al. (1998) Mutations of the forkhead/winged-helix gene, FKHL7, in patients with Axenfeld–Rieger anomaly. Am. J. Hum. Genet., 63, 1316–1328.[ISI][Medline]
33 Pei, Y.F. and Rhodin, J.A. (1970) The prenatal development of the mouse eye. Anat. Rec., 168, 105–125.[Medline]
34 Semina, E.V., Reiter, R., Leysens, N.J., Alward, W.L., Small, K.W., Datson, N.A., Siegel Bartelt, J., Bierke Nelson, D., Bitoun, P., Zabel, B.U. et al. (1996) Cloning and characterization of a novel bicoid-related homeobox transcription factor gene, RIEG, involved in Rieger syndrome. Nature Genet., 14, 392–399.[ISI][Medline]
35 Gage, P.J. and Camper, S.A. (1997) Pituitary homeobox 2, a novel member of the bicoid-related family of homeobox genes, is a potential regulator of anterior structure formation. Hum. Mol. Genet., 6, 457–464.
36 Vollrath, D., Jaramillo Babb, V.L., Clough, M.V., McIntosh, I., Scott, K.M., Lichter, P.R. and Richards, J.E. (1998) Loss-of-function mutations in the LIM-homeodomain gene, LMX1B, in nail–patella syndrome. Hum. Mol. Genet., 7, 1091–1098.
37 Pressman, C.L., Chen, H. and Johnson, R.L. (2000) Lmx1b, a LIM homeodomain class transcription factor is necessary for normal development of multiple tissues in the anterior segment of the murine eye. Genesis, 26, 15–25.[ISI][Medline]
38 Morrison Graham, K., Schatteman, G.C., Bork, T., Bowen Pope, D.F. and Weston, J.A. (1992) A PDGF receptor mutation in the mouse (Patch) perturbs the development of a non-neuronal subset of neural crest-derived cells. Development, 115, 133–142.[Abstract]
39 Reneker, L.W. and Overbeek, P.A. (1996) Lens-specific expression of PDGF-A alters lens growth and development. Dev. Biol., 180, 554–565.[ISI][Medline]
40 Kaestner, K.H., Bleckmann, S.C., Monaghan, A.P., Schlondorff, J., Mincheva, A., Lichter, P. and Schutz, G. (1996) Clustered arrangement of winged helix genes Fkh-6 and MFH-1: possible implications for mesoderm development. Development, 122, 1751–1758.[Abstract]
41 Mann, I. (1964) The Development of the Human Eye. Grune & Stratton, New York, NY.
42 Wulle, K.G. (1972) The development of the productive and draining system of the aqueous humour in the human eye. Adv. Ophthalmol., 26, 269–355.
43 Smelser, G.K. and Ozanics, V. (1971) The development of the trabecular meshwork in primate eyes. Am. J. Ophthalmol., 71(suppl.), 366–385.
44 Hamanaka, T., Bill, A., Ichinohasama, R. and Ishida, T. (1992) Aspects of the development of Schlemm’s canal. Exp. Eye Res., 55, 479–488.[ISI][Medline]
45 Beauchamp, G.R. and Knepper, P.A. (1984) Role of the neural crest in anterior segment development and disease. J. Pediatr. Ophthalmol. Strabismus, 21, 209–214.[Medline]
46 Stoilov, I., Nurten Akarsu, A. and Sarfarazi, M. (1997) Identification of cytochrome P4501B1 as the gene mutated in primary congenital glaucoma families linked to the GLC3A locus on 2p21. Am. J. Hum. Genet., 61, A21.
47 Semina, E.V., Ferrell, R.E., Mintz Hittner, H.A., Bitoun, P., Alward, W.L., Reiter, R.S., Funkhauser, C., Daack Hirsch, S. and Murray, J.C. (1998) A novel homeobox gene PITX3 is mutated in families with autosomal-dominant cataracts and ASMD. Nature Genet., 19, 167–170.[ISI][Medline]
48 Hanson, I.M., Fletcher, J.M., Jordan, T., Brown, A., Taylor, D., Adams, R.J., Punnett, H.H. and van Heyningen, V. (1994) Mutations at the PAX6 locus are found in heterogeneous anterior segment malformations including Peters’ anomaly. Nature Genet., 6, 168–173.[ISI][Medline]
49 Fatt, I. and Weissman, B.A. (1992) The aqueous humor. Physiology of the Eye. An Introduction to the Vegetative Functions. Butterworth-Heinemann, Stoneham, UK, pp. 17–30.
50 Coca Prados, M., Escribano, J. and Ortego, J. (1999) Differential gene expression in the human ciliary epithelium. Prog. Retin. Eye Res., 18, 403–429.[ISI][Medline]
51 Acott, T.S. (1992) Trabecular extracellular matrix regulation. In Drance, S.M., Van Buskirk, E.M. and Neufeld, A.H. (eds), Pharmacology of Glaucoma. Williams & Wilkins, Baltimore, MD, pp. 125–157.
52 Francois, J. (1975) The importance of the mucopolysaccharides in intraocular pressure regulation. Invest. Ophthalmol., 14, 173–176.
53 Acott, T.S. and Wirtz, M.K. (1996) Biochemistry of aqueous outflow. In Ritch, R., Shields, M.B. and Krupin, T. (eds), The Glaucomas, Basic Science, Vol. 1. Mosby Year Book, St Louis, MO, pp. 281–305.
54 Swiderski, R.E., Reiter, R.S., Nishimura, D.Y., Alward, W.L.M., Kalenak, J.W., Searby, C.S., Stone, E.M., Sheffield, V.C. and Lin, J.J.C. (1999) Expression of the Mf1 gene in developing mouse hearts: implication in the development of human congenital heart defects. Dev. Dyn., 216, 16–27.[ISI][Medline]
55 Furumoto, T.A., Miura, N., Akasaka, T., Mizutani Koseki, Y., Sudo, H., Fukuda, K., Maekawa, M., Yuasa, S., Fu, Y., Moriya, H. et al. (1999) Notochord-dependent expression of MFH1 and PAX1 cooperates to maintain the proliferation of sclerotome cells during the vertebral column development. Dev. Biol., 210, 15–29.[ISI][Medline]
56 Dahl, E., Koseki, H. and Balling, R. (1997) Pax genes and organogenesis. Bioessays, 19, 755–765.[ISI][Medline]
57 Jordan, T., Hanson, I., Zaletayev, D., Hodgson, S., Prosser, J., Seawright, A., Hastie, N. and van Heyningen, V. (1992) The human PAX6 gene is mutated in two patients with aniridia. Nature Genet., 1, 328–332.[ISI][Medline]
58 Nutt, S.L., Vambrie, S., Steinlein, P., Kozmik, Z., Rolink, A., Weith, A. and Busslinger, M. (1999) Independent regulation of the two Pax5 alleles during B-cell development. Nature Genet., 21, 390–395.[ISI][Medline]
59 Riviere, I., Sunshine, M.J. and Littman, D.R. (1998) Regulation of IL-4 expression by activation of individual alleles. Immunity, 9, 217–228.[ISI][Medline]
60 Held, W. and Kunz, B. (1998) An allele-specific, stochastic gene expression process controls the expression of multiple Ly49 family genes and generates a diverse, MHC-specific NK cell receptor repertoire. Eur. J. Immunol., 28, 2407–2416.[ISI][Medline]
61 Yue, B.Y. (1996) The extracellular matrix and its modulation in the trabecular meshwork. Surv. Ophthalmol., 40, 379–390.[ISI][Medline]
62 Breen, M., Richardson, R., Bondareff, W. and Weinstein, H.G. (1973) Acidic glycosaminoglycans in developing sterno-costal cartilage of the hydrocephalic (ch+-ch+) mouse. Biochim. Biophys. Acta, 304, 828–836.[Medline]
63 Richardson, R.R. and Reyes, M.G. (1984) Renal dysplasia and chondrodysplasia in the ch hydrocephalic mouse: a cellular model of defective differentiation and organization. Mt Sinai J. Med., 51, 188–196.[Medline]