- Email: [email protected]
Autosomal-dominant Iridogoniodysgenesis and Axenfeld-Rieger Syndrome are Genetically Distinct
Autosomal . . dominant Iridogoniodysgenesis and Axenfeld . .Rieger Syndrome are Genetically Distinct Michael A. Walter, PhD, Farideh Mirzayans, MSc, Alan J. Mears, PhD, Kristin Hickey, BSc, William G. Pearce, MD Purpose: To determine whether there is a locus for iridogoniodysgenesis (IGO)/ familial iris hypoplasia in the region of the known Axenfeld-Rieger syndrome (ARS) locus at 4q25 and to determine the ocular phenotype within the autosomal-dominant iris hypoplasia group of disorders. Methods: Clinical examinations were performed on 27 members, with 11 affected from one family in which the IGO occurred in association with the nonocular features of ARS, and on 70 members with 30 affected from a second IGO family with ocular features only. Family members were genotyped for markers within the 4q25 region known to contain a locus for ARS. LOO scores were calculated with the MLiNK option of the LINKAGE program. Results: The iris hypoplasia in each IGO family was similar. In the IGO family with only ocular features (IGO anomaly), however, a majority of those affected had a goniodysgenesis with excess tissue in the angle and anomalous angle vascularity. These findings were absent in the IGO family with syndromic features (IGO syndrome). Linkage to the 4q25 region was excluded in the IGO anomaly family, whereas the family with IGO syndrome was found to be completely linked to the 4q25 region (peak LOO score with 04S407 of 7.827 at = 0.00). Conclusions: The authors' results suggest that mutations at the 4q25 locus can result in variable ocular features that also occur in combination with nonocular (dental and jaw) anomalies. Mutation of a separate locus must underlie IGO with ocular features only. A re-evaluation of the relation between the various forms of autosomal-dominant iris hypoplasia, therefore, may be warranted. Ophthalmology 1996; 103:1907-1915
e
Originally received: December 21, 1995. Revision accepted: July 18, 1996. From the Department of Ophthalmology, University of Alberta, Edmonton, Alberta, Canada. Part of the linkage analysis of the IGDS pedigree was presented at the American Society of Human Genetics Meeting, Minneapolis, October 1995. Supported by grants from the Alberta Heritage Fund for Medical Research (AHFMR grant no. 9400216) (Edmonton, AB, Canada), the Medical Research Council of Canada (MRC grant no. MT12916) (Ottawa, ON, Canada), and an AHFMR postdoctoral fellowship (Dr. Mears). Dr. Walker is an MRC and AHFMR scholar. Reprint requests to Michael A. Walter, PhD, Ocular Genetics Laboratory, 671 Heritage Medical Research Center, University of Alberta, Edmonton, Alberta, Canada.
Axenfeld,l in 1920, first described a patient with a prominent annular white line near the limbus at the level of Descemet membrane (embryotoxon posterior). From the iris stroma, a number of delicate fibrillae crossed the anterior chamber to this line. Rieger,2 in 1934, subsequently described two patients with the findings noted by Axenfeld, but also noted marked iris stromal atrophy and congenital pupillary abnormalities (ectopia, dyscoria). In the same year (1934), Rossano 3 described similar changes in a father and son who also showed hypertelorism with a broad flat nose. Rieger,4 in 1935, considered the embryotoxon posterior and iris hypoplasia to be features of the same disorder that he termed dysgenesis mesodermalis corneae et iris. Some of these patients had associated
1907
Ophthalmology
Volume 103, Number 11, November 1996
nonocular developmental defects, especially of the teeth, facial bones, and the periumbilical skin. 5- s The similarity of anterior segment angle defects described by Axenfeld and Rieger has led to the suggestion that these descriptions are part of a spectrum of developmental disorders. 4 ,9,10 Patients with Axenfeld-Rieger anomaly (ARA) present with the characteristic ocular features alone; the phenotype in those with both ocular findings and nonocular features (including maxillary hypoplasia, hypodontia, and failure of involution of the periumbilical skin) is classified as Axenfeld-Rieger syndrome (ARS). Glaucoma occurs in approximately half of the patients with ARA and ARS.lO Various chromosomal aberrations have been found in association with ARS, including involvements of chromosomes 4, 6, 9, 13, 18, and 21 (reviewed by Nielsen and Tranebjaerg ll ). The most consistent evidence points to an ARS locus mapping to 4q25. Several patients with deletions, including 4q25-27, have been described (reviewed by Kulharya and colleagues 12). Motegi and coworkers 13 suggested that ARS is not determined by an abnormality in 4q26. They described a patient with an interstitial deletion involving 4q26 who did not have the features of ARS, although the patient did have coloboma and microphthalmos. Taken together, these studies indicated that a locus in either 4q25 or 4q27 underlies ARS. A recent report of a patient with a translocation t(1; 4) (q23.I; q25) and ARS further narrowed the critical region to 4q25. 14 Genetic linkage analysis confirmed that a locus for ARS maps to 4q25. 15 Tight linkage to the epidermal growth factor (EGF) gene initially supported its role as a candidate gene. Subsequent analysis of the EGF gene in patients with ARS, however, has failed to find mutations that segregate with ARS.16 As the clinically related anterior segmental ocular dysgenesis syndrome (MIM# 107250) also has been mapped to 4q by linkage to the MNS blood group,17 it is possible that anterior segmental ocular dysgenesis syndrome is a variant of ARS. It is clear, however, that ARS and ARA are genetically heterogeneous. Legius and colleagues lS described a family with ARA in three successive generations in which linkage to the 4q25 region was excluded. Although the patients had the ocular malformation of this disorder and maxillary hypoplasia and hypertelorism, they did not show the dental or umbilical anomalies found in typical ARS. The known variability in expression of ARS 19 ,20 has made precise subdivision of the Axenfeld-Rieger-type eye malformations for the purposes of genetic linkage analysis very difficult. Iridogoniodysgenesis (IGD) is a related ocular abnormality characterized by abnormalities in differentiation of the iridocorneal angle tissue ("goniodysgenesis") and maldevelopment of the anterior stromal layer of the iris associated with increased intraocular pressure resulting in juvenile glaucoma. In one IGD family with 23 affected members, secondary glaucoma had developed in 19 by 30 years of age, whereas the remaining four members without glaucoma were younger than 7 years of age at time of examination. 21 Glaucoma onset occurred at less than 11 years in five individuals, between 11 and 20 years 1908
in seven individuals, and between 21 and 30 years in the remaining seven individuals. Iridogoniodysgenesis was first recognized in 1932 by Berg 22 as an autosomally inherited dominant disorder. lerndal 21 later re-examined and expanded Berg's pedigree and confirmed the iris and iridocorneal angle defects characteristic of IGD. Weatherill and Hart,23 in examining a different family, found iris hypoplasia and iridocorneal angle defects in 30 affected individuals. These latter two studies established the slitlamp and gonioscopic features of the iris and anterior chamber angle abnormalities upon which rests the current understanding of IGD. The slit-lamp appearance of the iris stroma is similar in IGD without somatic features (IGD anomaly [IGDAD and in IGD with ARS-like nonocular features (IGD syndrome [IGDSD. 23 - 27 However, the excess angle tissue noted on gonioscopy in the majority of patients with IGDA has not been found in those affected with IGDS. 2s ,27 Iridogoniodysgenesis has been observed, usually in isolated cases, in conjunction with other ocular and nonocular disorders. Iridogoniodysgenesis, with variable amounts of iris stromal hypoplasia, iris processes, and broad peripheral anterior synechiae in the angle, has been observed in association with Prader-Willi syndrome (interstitial deletion of the proximal long arm of chromosome 15), and sporadically with unilateral congenital ectropion uveae, due to iris pigment epithelium proliferation. 29 However, IGD is predominantly observed as an autosomal-dominant, inherited trait. The genetic defect(s) that underlie IGD appear to be the result of aberrant migration of the neural crest cells. Neither IGDA nor IGDS displays the prominent and centrally displaced Schwalbe line with iris adhesions that results in the abnormal shapes and locations of the pupil characteristic of ARA or ARS. Axenfeld-Rieger anomaly, ARS, IGDA, and IGDS, however, all have a goniodysgenesis responsible for the intraocular pressure regulation disturbance that leads to glaucoma. Recently, a family described as having autosomal-dominant iris hypoplasia was found to be linked to the region of the ARS locus at 4q25. 2S This family had ocular features very similar to those of IGD; therefore, we have investigated whether IGD also is linked to the 4q25 region. Although we confirmed linkage of the phenotype in a family with IGDS to the ARS region at 4q25, IGDA was found to be unlinked to this region.
Subjects and Methods Subjects The pedigrees of the two families analyzed in this study are shown in Figure 1. The IGDA family is originally from the Maritime region of Canada, and the IGDS family is from northeastern Alberta. The clinical features of the IGDA family have been reported previousll5,26 and include autosomal-dominant inheritance of IGDA through eight generations. Ocular features include marked iris stromal hypoplasia and iridocorneal angle malformations
IGDS FAMILY
was treated as having an unknown phenotype in linkage analysis.
Figure 1. Pedigrees of two families with iridogoniodysgenes is (IGO). Asterisks = individuals whose DNA was available for DNA studies. Plus symbol = individuals in the lGO syndrome fami ly with isolated maxillary hypoplasia and inguinal hernias in males. Arrow = individual III-6 referred to the text. Question mark = individual lY-29 in the IGO anomaly family whose phenotype was equivocal and who
V
IV
III
II
VIII
VII
VI
£..
V
~.
ro
~
-< ~
~
o· 0...
5:. o ~
o ......., ......
~
~
>'[
F
rt
(1)
~
oro
-l
~
IV
~ (t
IGDAFAMILY
III
II
Ophthalmology
Volume 103 , Number 11 , November 1996
Figure 2. Survey and Koeppe photographs of patients with iridogoniodysgenesis anomaly (lGDA) and iridogoniodysgenesis syndrome (lGDS). A, left eye ofVI-27 (lGDA family) shows marked iris stroma hypoplasia with exposure of the sphincter muscle. B, gonioscopy of the left eye of VI-27 (lGDA family) shows excess trabecular tissue, anomalous angle vascularity and white membrane-like tissue over the iris root. Anomalous vessels appear atthe 11-, 1-, 1:30-,2-, and 3-o'clock positions. Abnormal "yelLowish" tissue can be observed at the ll -o'clock position. (Reprinted with permission from the Canadian}oumal of Ophthalmology.) C, lefr eye of IV6 (lGDS family) shows marked iris stroma hypoplasia with exposure of the sphincter muscle. D, gonioscopy of the left eye of IV -6 (IGDS family) shows open-angle without visible s tructural abnormalities. but with white membrane-like tissue over the iris root. E, photograph of the teeth of IV6 (IGDS family) shows conical and misshaped teeth.
with excess "woolly" tissue in the angle and anomalous angle vascularity (Figs 2A and 2B). Nine of ten affected individuals in this family who were younger than 30 years of age at their initial presentation had abnormal intraocular pressures (>22 mmHg). The remaining affected individual was younger than 1 year of age at the time of examination and therefore may present with glaucoma at a later time. The IGDS family also has been clinically described previousll7 and presented with autosomaldominant inheritance of IGD and somatic abnormalities. Ocular findings included iris stromal hypoplasia (Fig 2C)
1910
and associated juvenile glaucoma. The angle is open without excess tissue or anomalous vascularity (Fig 2D). One eye within the IGDS pedigree had a distinct iris hypoplasia in which the pupil was downwardly displaced by a localized adhesion of the iris to the posterior periphery of the cornea, not unlike that found in ARS. The remaining 21 eyes with ocular defects in this family lacked corneal involvement, which is more characteristic of IGDA. The somatic abnormalities in the IGDS family include maxillary hypoplasia with dental anomalies (Fig 2E), inguinal hernia, redundant perumbilical skin, and, in
Walter et al . Genetic Analysis of lridogoniodysgenesis Table l. Iridogoniodysgenesis Anomaly versus Polymorphic Markers in the 4q25 Region LOD Score at Recombination Fractions «(})
0.00
D4S2623 D4S1616 D4S1611
-00 _00 _00
0.05
0.10
0.15
0.20
0.30
0.40
0.50
-6.03 -7.57 -2.33
-3.32 -4.72 -0.88
-1.98 -3.20 -0.24
-1.20 -2.22 0.08
-0.45 -1.04 0.25
-0.19 -0.39 0.15
0.00 0.00 0.00
LOD = logarithm of odds ratio.
males, hypospadias. Dental anomalies more severe than shown in Figure 2E occur in this family, but in general have been corrected by dental plates before ophthalmologic examination. The karyotype of the IGDS family is normal. 27 The study and collection of blood samples from all individuals included in this report were approved by the Research Ethics Board of the Faculty of Medicine of the University of Alberta. Genotypic and Molecular Studies Blood samples were collected in EDTA-containing tubes, and DNA was prepared from isolated leukocytes by standard organic solvent extraction procedures. Oligonucleotide primers for all markers used in this study were obtained from Research Genetics (Huntsville, AL). Polymerase chain reaction amplification for linkage analyses was done with 35S_dATP incorporated directly into the peR product as described. 30 Polymerase chain reaction primers were annealed at 55°C. Polymerase chain reaction products were separated on 6% denaturing polyacrylamide gels. After electrophoresis, gels were dried and used for autoradiography.
Results Linkage Analysis of lridogoniodysgenesis Anomaly
Linkage Analysis Linkage analysis was conducted with a DOS-compatible Power Macintosh 6100/66 computer (Apple Computers,
Table 2. Statistical Analysis of Linkage Heterogeneity of lridogoniodysgenesis Anomaly and Syndrome*
df Linkage heterogeneity versus linkage homogeneity Linkage homogeneity versus no linkage Linkage heterogeneity versus no linkage
Inc, Cupertino, CA). The Hypercard-based program LINKAGE INTERFACE3l was used to store and export information for linkage analysis with the DOS-computer program LINKAGE. 32 LOD score values were calculated with the MLINK option of the LINKAGE program. The allele frequencies were assumed to be equal for the markers used in this study. Varying the allele frequency, from 0.01 to 0.50, of the D4S407 allele found to be linked to the IGDS phenotype did not affect the linkage results. This indicated that the assumption of equal allele frequency did not bias our results. Although there is no evidence of reduced penetrance for either ARS or IGD, the phenotypes in these families were analyzed as autosomal-dominant traits with incomplete penetrance (95%) and gene frequencies of 0.0001. Varying the estimations of the disease allele frequencies from 0.001 to 0.00001 did not significantly alter the LOD scores. Significance of linkage was evaluated with standard criteria (Zmax > 3). Exclusion of linkage was evaluated with the criteria suggested by Morton 33 (Zmax < -2). Linkage heterogeneity was evaluated with the HOMOG program?2
2
Chi-square
P
14.06
<0.001
4.57
<0.05
18.62
<0.0001
* Statistical analyses of the LOD scores of D4S1616 within the iridogoniodysgenesis anomaly and syndrome families were conducted by the HOMOG program. 32
Investigation of the family with the phenotype of IGDA (no associated nonocular features) with the 4q25 markers D4S2623, D4S1616, and D4S1611 excluded the 4q25 region from containing the locus underlying the phenotype present in this family (Table 1). The latter two markers are the closest markers to ARS 28 ; therefore, this result effectively excludes IGDA from being allelic with ARS. Analysis of the LOD scores of both the IGDA and IGDS family (linkage results discussed below) for the marker D4S1616 with the HOMOG program resulted in significant evidence of linkage heterogeneity (Table 2). Further analysis with additional markers has excluded the entire length of chromosome 4 from containing the IGDA locus (A. J. Mears and M. A. Walter, unpublished data). These results indicate that IGDA is genetically and clinically distinct from ARS. Segregation and Linkage Analysis of a Family with lridogoniodysgenesis Syndrome As is apparent from examination of the IGDS pedigree (Fig 1), it was not initially clear whether two distinct
1911
Ophthalmology
Volume 103 , Number 11, November 1996
Table 3. Iridogoniodysgenesis Syndrome versus Chromosome 4q25 Markers LOD Score at Recombination Fractions (9)
0.00
D4S1570 D4S1564 D4S1571 D4S2623 D4S406 D4S193 D4S407 D4S1616 D4S1613 D4S828 D4S427 D4S430
-14.98 4.47 0.99 2.79* 7.07* 6.80* 7.51 * 4.65* 2.82* 3.37* 6.19* -5.72
0.01
0.02
0.03
0.04
1.20 4.52 0.99 2.74 6.96 6.69 7.39 4.57 2.77 3.32 6.09 3.42
1.78 4.54 1.00* 2.69 6.85 6.58 7.27 4.49 2.72 3.26 5.99 3.64
2.09 4.55 * 0.99 2.64 6.74 6.47 7.15 4.41 2.67 3.20 5.89 3.73
2.30
4.54 0.99 2.59 6.62 6.36 7.02 4.33 2.62 3.15 5.79 3.76
0.05
2.44 4.53 0.98 2.54 6.51 6.25 6.90 4.25 2.57 3.09 5.69 3.77*
0.10
0.20
0.30
0.40
2. 73 4.32 0.92 2.29 5.91 5.67 6.26 3.84 2.93 2.79 5.16 3.62
2.50 3.55 0.70 1.72 4.60 4.42 4.88 2.95 1.70 2.15 4.01 2.93
1.84 2.49 0.40 1.11 3.15 3.02 3.32 1.94 1.05 1.45 2.72 2.01
0.96 1.24 0.12 0.49 1.56 1.51 1.6L 0.87 0.39 0.73 1.33 0.99
LOD = logarithm of odds ratio.
* Zm"
value.
disorders were independently segregating in this familyIGD plus nonocular features (IGDS) and a second genetic anomaly (isolated maxillary hypoplasia without any associated dental or ocular involvement)-or whether this family presented with a single unknown disorder of variable expressivity and penetrance. 27 In favor of the former hypothesis, the branch of the family involving individual III-6 demonstrates maxillary hypoplasia and inguinal hernia segregating independently of an eye defect (Fig 1). Linkage analysis of the isolated maxillary hypoplasia phenotype (individuals with IGDS had unknown phenotype, except obligate carriers 11-2 and III-2) formally excluded a 20-cM interval encompassing the ARS critical region at 4q25 from containing the locus underlying the isolated maxillary hypoplasia phenotype (data not shown). Haplotype analysis of individuals with isolated maxillary hypoplasia with additional markers also excluded the 4q25 interval as linked to the locus responsible for the isolated maxillary hypoplasia (data not shown). A locus elsewhere in the genome must therefore underlie the maxillary hypoplasia in these individuals. Genetic linkage analysis of the IGDS kindred therefore was conducted under the hypothesis that two independent disorders were segregated in this family. Complete linkage of the IGDS phenotype was found to D4S407 and D4S1616, two markers in the 4q25 region (LOD scores of 7.83 and 4.82, respectively; e = 0). In this linkage analysis, individuals with maxillary hypoplasia but without eye defects (individuals III-6, IV-2, IV-9, IV-lO, IV18, and IV-21) were treated as being unaffected with respect to the IGDS phenotype.
Linkage Analysis of the lridogoniodysgenesis Syndrome Family with Additional 4q25 Loci Linkage studies were conducted on the IGDS kindred with additional markers in the 4q25 region. The linkage results are summarized in Table 3. Recombination was
1912
not observed between the affected phenotypes in this family and loci D4S406, D4S407, D4S1616, D4S193 , D4S1611, D4S1613, D4S1612, and D4S427. Two recombinants between the phenotype in the IGDS family and D4S 1564 were observed, indicating that D4S 1564 must flank proximal to the ARS locus. The distal limit of the critical region is delineated by a recombinant with D4S430, located approximately 14 cM from D4S1564.34 These results are consistent with the previous localization of a locus responsible for ARS to 4q25, and with the possibility that IGDS and ARS are allelic. A schematic diagram of the critical region for ARS and IGDS is shown in Figure 3. These results indicate that mutation of a gene within the 14-cM interval flanked by D4S 1564 and D4S430 is responsible for the phenotype present in the IGDS family.
Discussion We have conducted genetic linkage analysis on two kindreds with iris stromal hypoplasia and iridocorneal angle anomalies to test whether IGD is linked to the same region as the ARS locus at 4q25. A kindred with IGDA (no associated non ocular features) was unlinked to the 4q25 region. Iridogoniodysgenesis anomaly is therefore genetically distinct from ARS. This finding of genetic heterogeneity is consistent with the report of the absence linkage in a family with ARA to markers in the 4q25 region. IS However, a second family with IGD plus nonocular features (IGDS) was significantly linked to loci in the 4q25 region. Findings in this family included maxillary hypoplasia, dental anomalies, and umbilical hernias. The ocular features, aside from one eye, however, included only iris stromal defects without the variability in size or location of Schwalbe line, without adhesions between the iris stroma and Schwalbe line and without the abnormal shapes and locations of the pupil typical of ARS. The
Walter et al . Genetic Analysis of lridogoniodysgenesis
16 15
P
14 13 12 11 11 12
S1570
Figure 3. Critical region for Axenfeld-Rieger syndrome/iridogoniodysgenesis syndrome (IGDS) on chromosome 4q25. Shaded region = minimum critical region for IGDS as determined by examinations of recombinant chromosomes in the IGDS family. Genetic distances (in cM) between markers are indicated to the left (obtained from the public database ftp.mtldclin.edu/pub/maps). Vertical lines = the published physical localizations of markers on human chromosome 4. Parentheses = the physical localization of the epidermal growth factor gene (EGF) which was not informative in genetic linkage analysis of the IGDS family.
(EGF)
13 21
q
22 23 24 25 26 27 28 31 32 33
34
35
FABP2
fact that one eye differed raised the question of a possible relation between the IGOS phenotype and that of ARS, despite the absence of the characteristic/diagnostic ocular features of ARS in the remaining 21 eyes. Our linkage results for these families with various presentations of "iris hypoplasia" are especially intriguing because a recent report has suggested that autosomal-dominant iris hypoplasia maps to the 4q25 region?8 Closer examination of the family in this report, however, indicates that at least one member of this family also had nonocular features similar to those found in ARS. It is likely, therefore, that this family is very similar to the family investigated here with IGOS. Taken together, these results are consistent with the suggestion of wide clinical variability in the presentation of the ocular features associated with defects of the ARS locus at 4q25, but that the most consistent findings of the 4q25 ARS locus are the nonocular features found in addition to these ocular findings (Table 4). In particular, a prominent Schwalbe line with or without adhesions between the iris and this prominent line does not necessarily appear as a
clinical feature in some families with ocular-disorder phenotypes mapping to the ARS locus at 4q25. We suggest, therefore, that a re-evaluation of the diagnostic criteria of ARS may be necessary. It is apparent that the four autosomal-dominant iris hypoplasia phenotypes (ARA, ARS, IGOA, and IGOS) have certain features in common, other than their patterns of inheritance (Table 4). The ARS gene defect results in ocular features of iris stromal hypoplasia, a prominent and displaced Schwalbe line plus adhesions between the iris and Schwalbe line, in association with nonocular features and maps to 4q25. l5 The ARA gene, responsible for the identical ocular features, but without dental or umbilical features , however, does not map to 4q25. l8 Axenfeld-Rieger syndrome and ARA are, therefore, genetically distinct disorders. The situation concerning the two IGO phenotypes is similar. Iridogoniodysgenesis syndrome also maps to the 4q25 region (and therefore is possibly allelic to ARS 28 ) , but IGOA maps elsewhere in the genome (Table 1). It remains to be seen if IGOA and ARA are genetically distinct.
1913
Ophthalmology Table
Volume 103, Number 11, November 1996
4. Clinical and Genetic Summary of the Four Iris Hypoplasia Phenotypes Genetics Symptoms
Disorder
ARA
ARS
IGDA
IGDS
Genetic Localization
Ocular Features
Nonocular Features
Inheritance Pattern
Iris stromal hypoplasia Prominent Schwalbe line Adhesions between iris and Schwalbe line Glaucoma As above
H ypertelorism * Maxillary hypoplasia*
Autosomal dominant Fully penetrant Variable expressivity
Hypertelorism Maxillary hypoplasia Dental anomalies Umbilical hernia Hypospadius {males} None
As above
4q25
As above
?t
21-26, this
4q25
27, 28, this
Iris stromal hypoplasia Goniodysgenesis with "woolly" excess tissue in angle Anomalous angle vascularity Glaucoma Iris stromal hypoplasia Goniodsygenesis with open iridocorneal angle Glaucoma
Maxillary hypoplasia Dental anomalies Umbilical hernia Hypospadius {males}
As above
?t
References
9, 10, 18
1, 3, 4, 6-8, 16
report
report
ARA = Axenfeld-Rieger anomaly; ARS = Axenfeld-Rieger syndrome; IGDA = iridogoniodysgenesis anomaly; IGDS = iridogoniodysgenesis syndrome.
* Shields9,10 reported no associated nonocular findings associated with ARA; Leguis et aI, is however, reported ARA patients with the nonocular features indicated here.
t Location of the gene(s} for ARA and IGDA are not known; however, the 4q25 region has been excluded from containing the gene underlying either phenotype.
There are some differences in the ocular features between the two forms of IGD, unlike the situation for the ARAIARS phenotypes. Neither form of IGD shows the prominence of Schwalbe line with iris adhesions characteristic of ARA and ARS. Both forms show marked hypoplasia ofthe iris stroma, but IGDA shows angle abnormalities not seen in families with IGDS (Fig 2). The majority of those affected with IGDA have a variable amount of excess' 'woolly" tissue in the iridocorneal angle that obscures the posterior trabecular meshwork scleral spur and iris root. Anomalous vessels are also present in the angle. 23 .24 ,26 These iridocorneal angle features are not present in the syndromic form of IGD.D 28 It is now thought that most congenital anterior segment anomalies are the result of aberrant migration or terminal induction of the neural crest cells. 35 ,36 During the sixth week of embryonic life, three waves of neural crest cells migrate centrally and differentiate to form the corneal endothelium, keratocytes, and iris stroma, Disorders of the anterior segment have been grouped into four catego-
1914
ries 37 .38 based on defects affecting the formation, migration, proliferation, and differentiation of these neural crest cells. Axenfeld-Rieger anomaly, ARS, IGDA, and IGDS all should be placed in the category of defects likely resulting from abnormal neural crest cell migration or developmental arrest. Molecular description of the gene defects in these diseases would establish the true relation between these classifications. However, with the exception of the molecular defects causing aniridia, some cases of Peter anomaly, and autosomal-dominant keratitis, all recently shown to lie in the PAX-6 gene,30,39~41 molecular characterization of the genetic defects underlying such anterior segment developmental disorders have not yet been identified.
References 1. Axenfeld T, Embryotoxon corneae posterius. Ber Deutsch Ophthalmol Ges 1920;42:381-2.
Walter et al . Genetic Analysis of Iridogoniodysgenesis 2. Rieger H. Verlagerung und Schitzform der Pupille mit Hypoplasie des Irisvordblattes. Z Augenheilk 1934;84:98-9. 3. Rossano R. Absence presque complete du feuillet mesodermique de I'iris dans deux generations. Hypertension oculaire et polycorie dans un cas. Bull Soc Ophtalmol (Paris) 1934;1:3-12. 4. Rieger H. Beitraege zur Kenntnis seltener Missbildungen der Iris: Uber Hypoplasie des Irisvorderblattes mit Verlagerung und Entrundung der Pupille. Albrecht von Graefes Arch Klin Exp Ophthalmol 1935; 133:602-35. 5. Mathis H. Dysgenesis mesodermalis corneae et iridis (Gesellschaftsberichte Aussprache). Z Augenhelik 1935; 86 :333. 6. Rieger H. Erbfragen in der Augenheilkunde. Albrecht von Graefes Arch Klin Exp OphthalmoI1941;143:277-99. 7. Alkemade PPH. Dysgenesis Mesodermalis of the Iris and the Cornea: A Study of Rieger's Syndrome and Peter's Anomaly. Assen, The Netherlands: Van Gorcum; 1969. 8. Jorgenson RJ, Levin LS, Cross HE, et al. The Rieger syndrome. Am J Med Genet 1978;2:307-18. 9. Shields MB. Axenfeld-Rieger syndrome: a theory of mechanism and distinctions from the iridocorneal endothelial syndrome. Trans Am Ophthalmol Soc 1983;81:736-84. 10. Shields MB, Buckley E, Klintworth GK, Thresher R. Axenfeld-Rieger syndrome. A spectrum of developmental disorders. Surv Ophthalmol 1985;29:387-409. 11. Nielsen F, Tranebjaerg L. A case of partial monosomy 21q22.2 associated with Rieger's syndrome. J Med Genet 1984;21:218-21. 12. Kulharya AS, Maberry M, Kukolich MK, et al. Interstitial deletions 4q21.1q25 and 4q25q27: phenotype variability and relation to Rieger anomaly. Am J Med Genet 1995;55:165-70. 13. Motegi T, Nakamura K, Terakawa T, et al. Deletion of a single chromosome band 4q26 in a malformed girl: exclusion of Rieger syndrome associated gene(s) from the 4q26 segment. J Med Genet 1988;25:628-33. 14. Makita T, Masuno M, Irnaizumi K, et al. Rieger syndrome with de novo reciprocal translocation t(1 ;4)(q23.1; q25). Am J Med Genet 1995 ;57:19-21. 15. Murray JC, Bennett SR, Kwitek AE, et al. Linkage of Rieger syndrome to the region of the epidermal growth factor gene on chromosome 4. Nature Genet 1992;2:469. 16. Alward LM, Murray Je. Axenfeld-Rieger syndrome. In: Wiggs J, ed. Molecular Genetics of Ocular Disease. New York: Wiley-Liss, 1995;31-50. 17. Ferrell RE, Hittner HM, Kretzer FL, Antoszyk JH. Anterior segment mesenchymal dysgenesis: probable linkage to the MNS blood group on chromosome 4. Am J Hum Genet 1982;34:245-9. 18. Legius E, de Die Smulders CEM, Verbraak F, et al. Genetic heterogeneity in Rieger eye malformation. J Med Genet 1994;31:340-1. 19. Pearce WG, Kerr CB . Inherited variation in Rieger's malformation. Br J Ophthalmol 1965;49:530-7. 20. Geyer 0 , Loewenstein A, Garty BZ, Lazar M. Different manifestation of Rieger syndrome in monozygotic twins. J Pediatr Ophthalmol Strabismus 1994;31:57-8. 21. Jerndal T. Dominant goniodysgenesis with late congenital
22. 23. 24. 25. 26. 27.
28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41.
glaucoma: a re-examination of Berg' s pedigree. Am J OphthaI mol 1972;74:28-33. Berg F. Erbliches jugendliches Glaukom. Acta Ophthalmol 1932; 10:568-87. Weatherill JR, Hart CT. Familial hypoplasia of the iris stroma associated with glaucoma. Br J Ophthalmol 1969; 53:433-8. Martin JP, Zorab Ee. Familial glaucoma in nine generations of a South Hampshire family. Br J Ophthalmol 1974;58: 536-42. Pearce WG, Wyatt HT, Boyd TAS, et al. Autosomal dominant iridogoniodysgenesis: a genetic and clinical study. Birth Defects 1982; 18:561-9. Pearce WG, Wyatt HT, Boyd TAS, et al. Autosomal dominant iridogoniodysgenesis: genetic features. Can J Ophthalmol 1983; 18:7-10. Chisholm lA, Chudley AE. Autosomal dominant iridogoniodysgenesis with associated somatic anomalies: four-generation family with Rieger's syndrome. Br J Ophthalmol 1983;67:529-34. Heon E, Sheth BP, Kalenak JW, et al. Linkage of autosomal dominant iris hypoplasia to the region of the Rieger syndrome locus (4q25). Hum Mol Genet 1995;4:1425-39. Mattox C, Walton DS. Hereditary primary childhood glaucomas. International Ophthalmol Clinics 1993;33:121 -34. Mirzayans F, Pearce WG, MacDonald 1M, Walter MA. Mutation of the PAX6 gene in patients with autosomal dominant keratitis. Am J Hum Genet 1995;57:539-28. Nichols BE, Stone EM, Sheffield VCA. User-friendly Macintosh interface for DOS-based linkage analysis. CABIOS 1993; 9:757 -9. Ott J. Analysis of human genetic linkage, 2nd ed. Baltimore: Johns Hopkins University Press, 1991; 198-216. Morton NE. Sequential test for the detection of linkage. Am J Hum Genet 1955;8:80-96. Gyapay G, Morissette J, Vignal A, et al. The 1993-94 Genethon human genetic linkage map. Nature Genet 1994;7:246-339. Kupfer C, Kaiser-Kupfer MI. New hypothesis of developmental anomalies of the anterior chamber associated with glaucoma. Trans Ophthalmol Soc UK 1978;98:213-5. Kaiser-Kupfer MI. Neural crest origin of trabecular meshwork and other structures of the anterior chamber. Am J Ophthalmol 1989; 107:671-3. Bahn CF, Falls HF, Varley GA, et al. Classification of corneal endothelial disorders based upon neural crest origin. Ophthalmology 1984;91:558-63. Dowling JL, Albert DM, Nelson LB, Walton DS. Primary glaucoma associated with iridotrabecular dysgenesis and ectropion uveae. Am J Ophthalmol 1985;92:912-21. Jordan T, Hanson I , Zaletayev D, et al. The human PAX6 gene is mutated in two patients with aniridia. Nature Genet 1992; 1:328-32. Glaser T, Walton DS, Maas RL. Genomic structure, evolutionary conservation and aniridia mutations in the human pax6 gene. Nature Genet 1992;2:232-9. Hanson 1M, Fletcher JM, Jordan T, et al. Mutations at the P AX6 locus are found in heterogeneous anterior segment malformations including Peters anomaly. Nature Genet 1994;6:168-73.
1915
e
Originally received: December 21, 1995. Revision accepted: July 18, 1996. From the Department of Ophthalmology, University of Alberta, Edmonton, Alberta, Canada. Part of the linkage analysis of the IGDS pedigree was presented at the American Society of Human Genetics Meeting, Minneapolis, October 1995. Supported by grants from the Alberta Heritage Fund for Medical Research (AHFMR grant no. 9400216) (Edmonton, AB, Canada), the Medical Research Council of Canada (MRC grant no. MT12916) (Ottawa, ON, Canada), and an AHFMR postdoctoral fellowship (Dr. Mears). Dr. Walker is an MRC and AHFMR scholar. Reprint requests to Michael A. Walter, PhD, Ocular Genetics Laboratory, 671 Heritage Medical Research Center, University of Alberta, Edmonton, Alberta, Canada.
Axenfeld,l in 1920, first described a patient with a prominent annular white line near the limbus at the level of Descemet membrane (embryotoxon posterior). From the iris stroma, a number of delicate fibrillae crossed the anterior chamber to this line. Rieger,2 in 1934, subsequently described two patients with the findings noted by Axenfeld, but also noted marked iris stromal atrophy and congenital pupillary abnormalities (ectopia, dyscoria). In the same year (1934), Rossano 3 described similar changes in a father and son who also showed hypertelorism with a broad flat nose. Rieger,4 in 1935, considered the embryotoxon posterior and iris hypoplasia to be features of the same disorder that he termed dysgenesis mesodermalis corneae et iris. Some of these patients had associated
1907
Ophthalmology
Volume 103, Number 11, November 1996
nonocular developmental defects, especially of the teeth, facial bones, and the periumbilical skin. 5- s The similarity of anterior segment angle defects described by Axenfeld and Rieger has led to the suggestion that these descriptions are part of a spectrum of developmental disorders. 4 ,9,10 Patients with Axenfeld-Rieger anomaly (ARA) present with the characteristic ocular features alone; the phenotype in those with both ocular findings and nonocular features (including maxillary hypoplasia, hypodontia, and failure of involution of the periumbilical skin) is classified as Axenfeld-Rieger syndrome (ARS). Glaucoma occurs in approximately half of the patients with ARA and ARS.lO Various chromosomal aberrations have been found in association with ARS, including involvements of chromosomes 4, 6, 9, 13, 18, and 21 (reviewed by Nielsen and Tranebjaerg ll ). The most consistent evidence points to an ARS locus mapping to 4q25. Several patients with deletions, including 4q25-27, have been described (reviewed by Kulharya and colleagues 12). Motegi and coworkers 13 suggested that ARS is not determined by an abnormality in 4q26. They described a patient with an interstitial deletion involving 4q26 who did not have the features of ARS, although the patient did have coloboma and microphthalmos. Taken together, these studies indicated that a locus in either 4q25 or 4q27 underlies ARS. A recent report of a patient with a translocation t(1; 4) (q23.I; q25) and ARS further narrowed the critical region to 4q25. 14 Genetic linkage analysis confirmed that a locus for ARS maps to 4q25. 15 Tight linkage to the epidermal growth factor (EGF) gene initially supported its role as a candidate gene. Subsequent analysis of the EGF gene in patients with ARS, however, has failed to find mutations that segregate with ARS.16 As the clinically related anterior segmental ocular dysgenesis syndrome (MIM# 107250) also has been mapped to 4q by linkage to the MNS blood group,17 it is possible that anterior segmental ocular dysgenesis syndrome is a variant of ARS. It is clear, however, that ARS and ARA are genetically heterogeneous. Legius and colleagues lS described a family with ARA in three successive generations in which linkage to the 4q25 region was excluded. Although the patients had the ocular malformation of this disorder and maxillary hypoplasia and hypertelorism, they did not show the dental or umbilical anomalies found in typical ARS. The known variability in expression of ARS 19 ,20 has made precise subdivision of the Axenfeld-Rieger-type eye malformations for the purposes of genetic linkage analysis very difficult. Iridogoniodysgenesis (IGD) is a related ocular abnormality characterized by abnormalities in differentiation of the iridocorneal angle tissue ("goniodysgenesis") and maldevelopment of the anterior stromal layer of the iris associated with increased intraocular pressure resulting in juvenile glaucoma. In one IGD family with 23 affected members, secondary glaucoma had developed in 19 by 30 years of age, whereas the remaining four members without glaucoma were younger than 7 years of age at time of examination. 21 Glaucoma onset occurred at less than 11 years in five individuals, between 11 and 20 years 1908
in seven individuals, and between 21 and 30 years in the remaining seven individuals. Iridogoniodysgenesis was first recognized in 1932 by Berg 22 as an autosomally inherited dominant disorder. lerndal 21 later re-examined and expanded Berg's pedigree and confirmed the iris and iridocorneal angle defects characteristic of IGD. Weatherill and Hart,23 in examining a different family, found iris hypoplasia and iridocorneal angle defects in 30 affected individuals. These latter two studies established the slitlamp and gonioscopic features of the iris and anterior chamber angle abnormalities upon which rests the current understanding of IGD. The slit-lamp appearance of the iris stroma is similar in IGD without somatic features (IGD anomaly [IGDAD and in IGD with ARS-like nonocular features (IGD syndrome [IGDSD. 23 - 27 However, the excess angle tissue noted on gonioscopy in the majority of patients with IGDA has not been found in those affected with IGDS. 2s ,27 Iridogoniodysgenesis has been observed, usually in isolated cases, in conjunction with other ocular and nonocular disorders. Iridogoniodysgenesis, with variable amounts of iris stromal hypoplasia, iris processes, and broad peripheral anterior synechiae in the angle, has been observed in association with Prader-Willi syndrome (interstitial deletion of the proximal long arm of chromosome 15), and sporadically with unilateral congenital ectropion uveae, due to iris pigment epithelium proliferation. 29 However, IGD is predominantly observed as an autosomal-dominant, inherited trait. The genetic defect(s) that underlie IGD appear to be the result of aberrant migration of the neural crest cells. Neither IGDA nor IGDS displays the prominent and centrally displaced Schwalbe line with iris adhesions that results in the abnormal shapes and locations of the pupil characteristic of ARA or ARS. Axenfeld-Rieger anomaly, ARS, IGDA, and IGDS, however, all have a goniodysgenesis responsible for the intraocular pressure regulation disturbance that leads to glaucoma. Recently, a family described as having autosomal-dominant iris hypoplasia was found to be linked to the region of the ARS locus at 4q25. 2S This family had ocular features very similar to those of IGD; therefore, we have investigated whether IGD also is linked to the 4q25 region. Although we confirmed linkage of the phenotype in a family with IGDS to the ARS region at 4q25, IGDA was found to be unlinked to this region.
Subjects and Methods Subjects The pedigrees of the two families analyzed in this study are shown in Figure 1. The IGDA family is originally from the Maritime region of Canada, and the IGDS family is from northeastern Alberta. The clinical features of the IGDA family have been reported previousll5,26 and include autosomal-dominant inheritance of IGDA through eight generations. Ocular features include marked iris stromal hypoplasia and iridocorneal angle malformations
IGDS FAMILY
was treated as having an unknown phenotype in linkage analysis.
Figure 1. Pedigrees of two families with iridogoniodysgenes is (IGO). Asterisks = individuals whose DNA was available for DNA studies. Plus symbol = individuals in the lGO syndrome fami ly with isolated maxillary hypoplasia and inguinal hernias in males. Arrow = individual III-6 referred to the text. Question mark = individual lY-29 in the IGO anomaly family whose phenotype was equivocal and who
V
IV
III
II
VIII
VII
VI
£..
V
~.
ro
~
-<
~
o· 0...
5:. o ~
o ......., ......
~
~
>'[
F
rt
(1)
~
oro
-l
~
IV
~ (t
IGDAFAMILY
III
II
Ophthalmology
Volume 103 , Number 11 , November 1996
Figure 2. Survey and Koeppe photographs of patients with iridogoniodysgenesis anomaly (lGDA) and iridogoniodysgenesis syndrome (lGDS). A, left eye ofVI-27 (lGDA family) shows marked iris stroma hypoplasia with exposure of the sphincter muscle. B, gonioscopy of the left eye of VI-27 (lGDA family) shows excess trabecular tissue, anomalous angle vascularity and white membrane-like tissue over the iris root. Anomalous vessels appear atthe 11-, 1-, 1:30-,2-, and 3-o'clock positions. Abnormal "yelLowish" tissue can be observed at the ll -o'clock position. (Reprinted with permission from the Canadian}oumal of Ophthalmology.) C, lefr eye of IV6 (lGDS family) shows marked iris stroma hypoplasia with exposure of the sphincter muscle. D, gonioscopy of the left eye of IV -6 (IGDS family) shows open-angle without visible s tructural abnormalities. but with white membrane-like tissue over the iris root. E, photograph of the teeth of IV6 (IGDS family) shows conical and misshaped teeth.
with excess "woolly" tissue in the angle and anomalous angle vascularity (Figs 2A and 2B). Nine of ten affected individuals in this family who were younger than 30 years of age at their initial presentation had abnormal intraocular pressures (>22 mmHg). The remaining affected individual was younger than 1 year of age at the time of examination and therefore may present with glaucoma at a later time. The IGDS family also has been clinically described previousll7 and presented with autosomaldominant inheritance of IGD and somatic abnormalities. Ocular findings included iris stromal hypoplasia (Fig 2C)
1910
and associated juvenile glaucoma. The angle is open without excess tissue or anomalous vascularity (Fig 2D). One eye within the IGDS pedigree had a distinct iris hypoplasia in which the pupil was downwardly displaced by a localized adhesion of the iris to the posterior periphery of the cornea, not unlike that found in ARS. The remaining 21 eyes with ocular defects in this family lacked corneal involvement, which is more characteristic of IGDA. The somatic abnormalities in the IGDS family include maxillary hypoplasia with dental anomalies (Fig 2E), inguinal hernia, redundant perumbilical skin, and, in
Walter et al . Genetic Analysis of lridogoniodysgenesis Table l. Iridogoniodysgenesis Anomaly versus Polymorphic Markers in the 4q25 Region LOD Score at Recombination Fractions «(})
0.00
D4S2623 D4S1616 D4S1611
-00 _00 _00
0.05
0.10
0.15
0.20
0.30
0.40
0.50
-6.03 -7.57 -2.33
-3.32 -4.72 -0.88
-1.98 -3.20 -0.24
-1.20 -2.22 0.08
-0.45 -1.04 0.25
-0.19 -0.39 0.15
0.00 0.00 0.00
LOD = logarithm of odds ratio.
males, hypospadias. Dental anomalies more severe than shown in Figure 2E occur in this family, but in general have been corrected by dental plates before ophthalmologic examination. The karyotype of the IGDS family is normal. 27 The study and collection of blood samples from all individuals included in this report were approved by the Research Ethics Board of the Faculty of Medicine of the University of Alberta. Genotypic and Molecular Studies Blood samples were collected in EDTA-containing tubes, and DNA was prepared from isolated leukocytes by standard organic solvent extraction procedures. Oligonucleotide primers for all markers used in this study were obtained from Research Genetics (Huntsville, AL). Polymerase chain reaction amplification for linkage analyses was done with 35S_dATP incorporated directly into the peR product as described. 30 Polymerase chain reaction primers were annealed at 55°C. Polymerase chain reaction products were separated on 6% denaturing polyacrylamide gels. After electrophoresis, gels were dried and used for autoradiography.
Results Linkage Analysis of lridogoniodysgenesis Anomaly
Linkage Analysis Linkage analysis was conducted with a DOS-compatible Power Macintosh 6100/66 computer (Apple Computers,
Table 2. Statistical Analysis of Linkage Heterogeneity of lridogoniodysgenesis Anomaly and Syndrome*
df Linkage heterogeneity versus linkage homogeneity Linkage homogeneity versus no linkage Linkage heterogeneity versus no linkage
Inc, Cupertino, CA). The Hypercard-based program LINKAGE INTERFACE3l was used to store and export information for linkage analysis with the DOS-computer program LINKAGE. 32 LOD score values were calculated with the MLINK option of the LINKAGE program. The allele frequencies were assumed to be equal for the markers used in this study. Varying the allele frequency, from 0.01 to 0.50, of the D4S407 allele found to be linked to the IGDS phenotype did not affect the linkage results. This indicated that the assumption of equal allele frequency did not bias our results. Although there is no evidence of reduced penetrance for either ARS or IGD, the phenotypes in these families were analyzed as autosomal-dominant traits with incomplete penetrance (95%) and gene frequencies of 0.0001. Varying the estimations of the disease allele frequencies from 0.001 to 0.00001 did not significantly alter the LOD scores. Significance of linkage was evaluated with standard criteria (Zmax > 3). Exclusion of linkage was evaluated with the criteria suggested by Morton 33 (Zmax < -2). Linkage heterogeneity was evaluated with the HOMOG program?2
2
Chi-square
P
14.06
<0.001
4.57
<0.05
18.62
<0.0001
* Statistical analyses of the LOD scores of D4S1616 within the iridogoniodysgenesis anomaly and syndrome families were conducted by the HOMOG program. 32
Investigation of the family with the phenotype of IGDA (no associated nonocular features) with the 4q25 markers D4S2623, D4S1616, and D4S1611 excluded the 4q25 region from containing the locus underlying the phenotype present in this family (Table 1). The latter two markers are the closest markers to ARS 28 ; therefore, this result effectively excludes IGDA from being allelic with ARS. Analysis of the LOD scores of both the IGDA and IGDS family (linkage results discussed below) for the marker D4S1616 with the HOMOG program resulted in significant evidence of linkage heterogeneity (Table 2). Further analysis with additional markers has excluded the entire length of chromosome 4 from containing the IGDA locus (A. J. Mears and M. A. Walter, unpublished data). These results indicate that IGDA is genetically and clinically distinct from ARS. Segregation and Linkage Analysis of a Family with lridogoniodysgenesis Syndrome As is apparent from examination of the IGDS pedigree (Fig 1), it was not initially clear whether two distinct
1911
Ophthalmology
Volume 103 , Number 11, November 1996
Table 3. Iridogoniodysgenesis Syndrome versus Chromosome 4q25 Markers LOD Score at Recombination Fractions (9)
0.00
D4S1570 D4S1564 D4S1571 D4S2623 D4S406 D4S193 D4S407 D4S1616 D4S1613 D4S828 D4S427 D4S430
-14.98 4.47 0.99 2.79* 7.07* 6.80* 7.51 * 4.65* 2.82* 3.37* 6.19* -5.72
0.01
0.02
0.03
0.04
1.20 4.52 0.99 2.74 6.96 6.69 7.39 4.57 2.77 3.32 6.09 3.42
1.78 4.54 1.00* 2.69 6.85 6.58 7.27 4.49 2.72 3.26 5.99 3.64
2.09 4.55 * 0.99 2.64 6.74 6.47 7.15 4.41 2.67 3.20 5.89 3.73
2.30
4.54 0.99 2.59 6.62 6.36 7.02 4.33 2.62 3.15 5.79 3.76
0.05
2.44 4.53 0.98 2.54 6.51 6.25 6.90 4.25 2.57 3.09 5.69 3.77*
0.10
0.20
0.30
0.40
2. 73 4.32 0.92 2.29 5.91 5.67 6.26 3.84 2.93 2.79 5.16 3.62
2.50 3.55 0.70 1.72 4.60 4.42 4.88 2.95 1.70 2.15 4.01 2.93
1.84 2.49 0.40 1.11 3.15 3.02 3.32 1.94 1.05 1.45 2.72 2.01
0.96 1.24 0.12 0.49 1.56 1.51 1.6L 0.87 0.39 0.73 1.33 0.99
LOD = logarithm of odds ratio.
* Zm"
value.
disorders were independently segregating in this familyIGD plus nonocular features (IGDS) and a second genetic anomaly (isolated maxillary hypoplasia without any associated dental or ocular involvement)-or whether this family presented with a single unknown disorder of variable expressivity and penetrance. 27 In favor of the former hypothesis, the branch of the family involving individual III-6 demonstrates maxillary hypoplasia and inguinal hernia segregating independently of an eye defect (Fig 1). Linkage analysis of the isolated maxillary hypoplasia phenotype (individuals with IGDS had unknown phenotype, except obligate carriers 11-2 and III-2) formally excluded a 20-cM interval encompassing the ARS critical region at 4q25 from containing the locus underlying the isolated maxillary hypoplasia phenotype (data not shown). Haplotype analysis of individuals with isolated maxillary hypoplasia with additional markers also excluded the 4q25 interval as linked to the locus responsible for the isolated maxillary hypoplasia (data not shown). A locus elsewhere in the genome must therefore underlie the maxillary hypoplasia in these individuals. Genetic linkage analysis of the IGDS kindred therefore was conducted under the hypothesis that two independent disorders were segregated in this family. Complete linkage of the IGDS phenotype was found to D4S407 and D4S1616, two markers in the 4q25 region (LOD scores of 7.83 and 4.82, respectively; e = 0). In this linkage analysis, individuals with maxillary hypoplasia but without eye defects (individuals III-6, IV-2, IV-9, IV-lO, IV18, and IV-21) were treated as being unaffected with respect to the IGDS phenotype.
Linkage Analysis of the lridogoniodysgenesis Syndrome Family with Additional 4q25 Loci Linkage studies were conducted on the IGDS kindred with additional markers in the 4q25 region. The linkage results are summarized in Table 3. Recombination was
1912
not observed between the affected phenotypes in this family and loci D4S406, D4S407, D4S1616, D4S193 , D4S1611, D4S1613, D4S1612, and D4S427. Two recombinants between the phenotype in the IGDS family and D4S 1564 were observed, indicating that D4S 1564 must flank proximal to the ARS locus. The distal limit of the critical region is delineated by a recombinant with D4S430, located approximately 14 cM from D4S1564.34 These results are consistent with the previous localization of a locus responsible for ARS to 4q25, and with the possibility that IGDS and ARS are allelic. A schematic diagram of the critical region for ARS and IGDS is shown in Figure 3. These results indicate that mutation of a gene within the 14-cM interval flanked by D4S 1564 and D4S430 is responsible for the phenotype present in the IGDS family.
Discussion We have conducted genetic linkage analysis on two kindreds with iris stromal hypoplasia and iridocorneal angle anomalies to test whether IGD is linked to the same region as the ARS locus at 4q25. A kindred with IGDA (no associated non ocular features) was unlinked to the 4q25 region. Iridogoniodysgenesis anomaly is therefore genetically distinct from ARS. This finding of genetic heterogeneity is consistent with the report of the absence linkage in a family with ARA to markers in the 4q25 region. IS However, a second family with IGD plus nonocular features (IGDS) was significantly linked to loci in the 4q25 region. Findings in this family included maxillary hypoplasia, dental anomalies, and umbilical hernias. The ocular features, aside from one eye, however, included only iris stromal defects without the variability in size or location of Schwalbe line, without adhesions between the iris stroma and Schwalbe line and without the abnormal shapes and locations of the pupil typical of ARS. The
Walter et al . Genetic Analysis of lridogoniodysgenesis
16 15
P
14 13 12 11 11 12
S1570
Figure 3. Critical region for Axenfeld-Rieger syndrome/iridogoniodysgenesis syndrome (IGDS) on chromosome 4q25. Shaded region = minimum critical region for IGDS as determined by examinations of recombinant chromosomes in the IGDS family. Genetic distances (in cM) between markers are indicated to the left (obtained from the public database ftp.mtldclin.edu/pub/maps). Vertical lines = the published physical localizations of markers on human chromosome 4. Parentheses = the physical localization of the epidermal growth factor gene (EGF) which was not informative in genetic linkage analysis of the IGDS family.
(EGF)
13 21
q
22 23 24 25 26 27 28 31 32 33
34
35
FABP2
fact that one eye differed raised the question of a possible relation between the IGOS phenotype and that of ARS, despite the absence of the characteristic/diagnostic ocular features of ARS in the remaining 21 eyes. Our linkage results for these families with various presentations of "iris hypoplasia" are especially intriguing because a recent report has suggested that autosomal-dominant iris hypoplasia maps to the 4q25 region?8 Closer examination of the family in this report, however, indicates that at least one member of this family also had nonocular features similar to those found in ARS. It is likely, therefore, that this family is very similar to the family investigated here with IGOS. Taken together, these results are consistent with the suggestion of wide clinical variability in the presentation of the ocular features associated with defects of the ARS locus at 4q25, but that the most consistent findings of the 4q25 ARS locus are the nonocular features found in addition to these ocular findings (Table 4). In particular, a prominent Schwalbe line with or without adhesions between the iris and this prominent line does not necessarily appear as a
clinical feature in some families with ocular-disorder phenotypes mapping to the ARS locus at 4q25. We suggest, therefore, that a re-evaluation of the diagnostic criteria of ARS may be necessary. It is apparent that the four autosomal-dominant iris hypoplasia phenotypes (ARA, ARS, IGOA, and IGOS) have certain features in common, other than their patterns of inheritance (Table 4). The ARS gene defect results in ocular features of iris stromal hypoplasia, a prominent and displaced Schwalbe line plus adhesions between the iris and Schwalbe line, in association with nonocular features and maps to 4q25. l5 The ARA gene, responsible for the identical ocular features, but without dental or umbilical features , however, does not map to 4q25. l8 Axenfeld-Rieger syndrome and ARA are, therefore, genetically distinct disorders. The situation concerning the two IGO phenotypes is similar. Iridogoniodysgenesis syndrome also maps to the 4q25 region (and therefore is possibly allelic to ARS 28 ) , but IGOA maps elsewhere in the genome (Table 1). It remains to be seen if IGOA and ARA are genetically distinct.
1913
Ophthalmology Table
Volume 103, Number 11, November 1996
4. Clinical and Genetic Summary of the Four Iris Hypoplasia Phenotypes Genetics Symptoms
Disorder
ARA
ARS
IGDA
IGDS
Genetic Localization
Ocular Features
Nonocular Features
Inheritance Pattern
Iris stromal hypoplasia Prominent Schwalbe line Adhesions between iris and Schwalbe line Glaucoma As above
H ypertelorism * Maxillary hypoplasia*
Autosomal dominant Fully penetrant Variable expressivity
Hypertelorism Maxillary hypoplasia Dental anomalies Umbilical hernia Hypospadius {males} None
As above
4q25
As above
?t
21-26, this
4q25
27, 28, this
Iris stromal hypoplasia Goniodysgenesis with "woolly" excess tissue in angle Anomalous angle vascularity Glaucoma Iris stromal hypoplasia Goniodsygenesis with open iridocorneal angle Glaucoma
Maxillary hypoplasia Dental anomalies Umbilical hernia Hypospadius {males}
As above
?t
References
9, 10, 18
1, 3, 4, 6-8, 16
report
report
ARA = Axenfeld-Rieger anomaly; ARS = Axenfeld-Rieger syndrome; IGDA = iridogoniodysgenesis anomaly; IGDS = iridogoniodysgenesis syndrome.
* Shields9,10 reported no associated nonocular findings associated with ARA; Leguis et aI, is however, reported ARA patients with the nonocular features indicated here.
t Location of the gene(s} for ARA and IGDA are not known; however, the 4q25 region has been excluded from containing the gene underlying either phenotype.
There are some differences in the ocular features between the two forms of IGD, unlike the situation for the ARAIARS phenotypes. Neither form of IGD shows the prominence of Schwalbe line with iris adhesions characteristic of ARA and ARS. Both forms show marked hypoplasia ofthe iris stroma, but IGDA shows angle abnormalities not seen in families with IGDS (Fig 2). The majority of those affected with IGDA have a variable amount of excess' 'woolly" tissue in the iridocorneal angle that obscures the posterior trabecular meshwork scleral spur and iris root. Anomalous vessels are also present in the angle. 23 .24 ,26 These iridocorneal angle features are not present in the syndromic form of IGD.D 28 It is now thought that most congenital anterior segment anomalies are the result of aberrant migration or terminal induction of the neural crest cells. 35 ,36 During the sixth week of embryonic life, three waves of neural crest cells migrate centrally and differentiate to form the corneal endothelium, keratocytes, and iris stroma, Disorders of the anterior segment have been grouped into four catego-
1914
ries 37 .38 based on defects affecting the formation, migration, proliferation, and differentiation of these neural crest cells. Axenfeld-Rieger anomaly, ARS, IGDA, and IGDS all should be placed in the category of defects likely resulting from abnormal neural crest cell migration or developmental arrest. Molecular description of the gene defects in these diseases would establish the true relation between these classifications. However, with the exception of the molecular defects causing aniridia, some cases of Peter anomaly, and autosomal-dominant keratitis, all recently shown to lie in the PAX-6 gene,30,39~41 molecular characterization of the genetic defects underlying such anterior segment developmental disorders have not yet been identified.
References 1. Axenfeld T, Embryotoxon corneae posterius. Ber Deutsch Ophthalmol Ges 1920;42:381-2.
Walter et al . Genetic Analysis of Iridogoniodysgenesis 2. Rieger H. Verlagerung und Schitzform der Pupille mit Hypoplasie des Irisvordblattes. Z Augenheilk 1934;84:98-9. 3. Rossano R. Absence presque complete du feuillet mesodermique de I'iris dans deux generations. Hypertension oculaire et polycorie dans un cas. Bull Soc Ophtalmol (Paris) 1934;1:3-12. 4. Rieger H. Beitraege zur Kenntnis seltener Missbildungen der Iris: Uber Hypoplasie des Irisvorderblattes mit Verlagerung und Entrundung der Pupille. Albrecht von Graefes Arch Klin Exp Ophthalmol 1935; 133:602-35. 5. Mathis H. Dysgenesis mesodermalis corneae et iridis (Gesellschaftsberichte Aussprache). Z Augenhelik 1935; 86 :333. 6. Rieger H. Erbfragen in der Augenheilkunde. Albrecht von Graefes Arch Klin Exp OphthalmoI1941;143:277-99. 7. Alkemade PPH. Dysgenesis Mesodermalis of the Iris and the Cornea: A Study of Rieger's Syndrome and Peter's Anomaly. Assen, The Netherlands: Van Gorcum; 1969. 8. Jorgenson RJ, Levin LS, Cross HE, et al. The Rieger syndrome. Am J Med Genet 1978;2:307-18. 9. Shields MB. Axenfeld-Rieger syndrome: a theory of mechanism and distinctions from the iridocorneal endothelial syndrome. Trans Am Ophthalmol Soc 1983;81:736-84. 10. Shields MB, Buckley E, Klintworth GK, Thresher R. Axenfeld-Rieger syndrome. A spectrum of developmental disorders. Surv Ophthalmol 1985;29:387-409. 11. Nielsen F, Tranebjaerg L. A case of partial monosomy 21q22.2 associated with Rieger's syndrome. J Med Genet 1984;21:218-21. 12. Kulharya AS, Maberry M, Kukolich MK, et al. Interstitial deletions 4q21.1q25 and 4q25q27: phenotype variability and relation to Rieger anomaly. Am J Med Genet 1995;55:165-70. 13. Motegi T, Nakamura K, Terakawa T, et al. Deletion of a single chromosome band 4q26 in a malformed girl: exclusion of Rieger syndrome associated gene(s) from the 4q26 segment. J Med Genet 1988;25:628-33. 14. Makita T, Masuno M, Irnaizumi K, et al. Rieger syndrome with de novo reciprocal translocation t(1 ;4)(q23.1; q25). Am J Med Genet 1995 ;57:19-21. 15. Murray JC, Bennett SR, Kwitek AE, et al. Linkage of Rieger syndrome to the region of the epidermal growth factor gene on chromosome 4. Nature Genet 1992;2:469. 16. Alward LM, Murray Je. Axenfeld-Rieger syndrome. In: Wiggs J, ed. Molecular Genetics of Ocular Disease. New York: Wiley-Liss, 1995;31-50. 17. Ferrell RE, Hittner HM, Kretzer FL, Antoszyk JH. Anterior segment mesenchymal dysgenesis: probable linkage to the MNS blood group on chromosome 4. Am J Hum Genet 1982;34:245-9. 18. Legius E, de Die Smulders CEM, Verbraak F, et al. Genetic heterogeneity in Rieger eye malformation. J Med Genet 1994;31:340-1. 19. Pearce WG, Kerr CB . Inherited variation in Rieger's malformation. Br J Ophthalmol 1965;49:530-7. 20. Geyer 0 , Loewenstein A, Garty BZ, Lazar M. Different manifestation of Rieger syndrome in monozygotic twins. J Pediatr Ophthalmol Strabismus 1994;31:57-8. 21. Jerndal T. Dominant goniodysgenesis with late congenital
22. 23. 24. 25. 26. 27.
28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41.
glaucoma: a re-examination of Berg' s pedigree. Am J OphthaI mol 1972;74:28-33. Berg F. Erbliches jugendliches Glaukom. Acta Ophthalmol 1932; 10:568-87. Weatherill JR, Hart CT. Familial hypoplasia of the iris stroma associated with glaucoma. Br J Ophthalmol 1969; 53:433-8. Martin JP, Zorab Ee. Familial glaucoma in nine generations of a South Hampshire family. Br J Ophthalmol 1974;58: 536-42. Pearce WG, Wyatt HT, Boyd TAS, et al. Autosomal dominant iridogoniodysgenesis: a genetic and clinical study. Birth Defects 1982; 18:561-9. Pearce WG, Wyatt HT, Boyd TAS, et al. Autosomal dominant iridogoniodysgenesis: genetic features. Can J Ophthalmol 1983; 18:7-10. Chisholm lA, Chudley AE. Autosomal dominant iridogoniodysgenesis with associated somatic anomalies: four-generation family with Rieger's syndrome. Br J Ophthalmol 1983;67:529-34. Heon E, Sheth BP, Kalenak JW, et al. Linkage of autosomal dominant iris hypoplasia to the region of the Rieger syndrome locus (4q25). Hum Mol Genet 1995;4:1425-39. Mattox C, Walton DS. Hereditary primary childhood glaucomas. International Ophthalmol Clinics 1993;33:121 -34. Mirzayans F, Pearce WG, MacDonald 1M, Walter MA. Mutation of the PAX6 gene in patients with autosomal dominant keratitis. Am J Hum Genet 1995;57:539-28. Nichols BE, Stone EM, Sheffield VCA. User-friendly Macintosh interface for DOS-based linkage analysis. CABIOS 1993; 9:757 -9. Ott J. Analysis of human genetic linkage, 2nd ed. Baltimore: Johns Hopkins University Press, 1991; 198-216. Morton NE. Sequential test for the detection of linkage. Am J Hum Genet 1955;8:80-96. Gyapay G, Morissette J, Vignal A, et al. The 1993-94 Genethon human genetic linkage map. Nature Genet 1994;7:246-339. Kupfer C, Kaiser-Kupfer MI. New hypothesis of developmental anomalies of the anterior chamber associated with glaucoma. Trans Ophthalmol Soc UK 1978;98:213-5. Kaiser-Kupfer MI. Neural crest origin of trabecular meshwork and other structures of the anterior chamber. Am J Ophthalmol 1989; 107:671-3. Bahn CF, Falls HF, Varley GA, et al. Classification of corneal endothelial disorders based upon neural crest origin. Ophthalmology 1984;91:558-63. Dowling JL, Albert DM, Nelson LB, Walton DS. Primary glaucoma associated with iridotrabecular dysgenesis and ectropion uveae. Am J Ophthalmol 1985;92:912-21. Jordan T, Hanson I , Zaletayev D, et al. The human PAX6 gene is mutated in two patients with aniridia. Nature Genet 1992; 1:328-32. Glaser T, Walton DS, Maas RL. Genomic structure, evolutionary conservation and aniridia mutations in the human pax6 gene. Nature Genet 1992;2:232-9. Hanson 1M, Fletcher JM, Jordan T, et al. Mutations at the P AX6 locus are found in heterogeneous anterior segment malformations including Peters anomaly. Nature Genet 1994;6:168-73.
1915