|
|
||||||||
1 From the Department of Ophthalmology, Manchester Royal Eye Hospital, Manchester, United Kingdom; 2 Department of Optometry and Vision Science, University of Wales, Cardiff, Wales; 3 Jules Gonin Eye Hospital and Division of Medical Genetics, University of Lausanne, Lausanne, Switzerland; 4 University Department of Medical Genetics and Regional Genetic Service, St. Marys Hospital, Manchester, United Kingdom; 5 Department of Pathology and 6 School of Biological Sciences, University of Manchester, Manchester, United Kingdom; 7 Department of Ophthalmology, Academic Medical Centre, University of Amsterdam, Amsterdam, The Netherlands; and 8 Centro de Oftalmologia Barraquer, Barcelona, Spain.
| Abstract |
|---|
|
|
|---|
METHODS. Keratoplasty tissue from each patient was examined by light and electron microscopy (LM and EM). DNA was obtained, and exons 4 and 12 of BIGH3 were analyzed by polymerase chain reaction and single-stranded conformation polymorphism/heteroduplex analysis. Abnormally migrating products were analyzed by direct sequencing.
RESULTS. In two families with type I CDB (CDBI), the R124L mutation was defined. There were light and ultrastructural features of superficial granular dystrophy and atypical banding of the "rod-shaped bodies" ultrastructurally. Patients from three families with "honeycomb" dystrophy were found to carry the R555Q mutation and had characteristic features of Bowmans dystrophy type II (CDBII).
CONCLUSIONS. There is a strong genotype:phenotype correlation among CBDI (R124L) and CDBII (R555Q). LM and EM findings suggest that epithelial abnormalities may underlie the pathology of both conditions. The findings clarify the confusion over classification of the Bowmans layer dystrophies.
| Introduction |
|---|
|
|
|---|
The identification of mutations in the BIGH3 (or
transforming growth factor, ß-induced [TGFBI]) gene on
chromosome 5q31 in a variety of the corneal dystrophies has facilitated
a reevaluation of their clinical classifications.5
Mutations of two "hotspots" at amino acids 124 and 555 of the
encoded protein, keratoepithelin, have been shown in several
BIGH3-related dystrophies. Among the anterior dystrophies,
including Reis-Bücklers and Thiel-Behnke dystrophies, three
mutations have been described: R555Q in "honeycomb"
dystrophy,5
R124L in a geographic form, and in a single
Sardinian family a trinucleotide deletion,
F540.6
7
8
We
have undertaken a detailed analysis of families with Bowmans layer
dystrophies. Our results suggest a strong genotype:phenotype
correlation, which allows a reevaluation of the clinical nomenclature.
| Materials and Methods |
|---|
|
|
|---|
Light and Electron Microscopy
Tissue for LM was fixed in 10% formalin in 0.1 M phosphate buffer
and for EM in 2.5% glutaraldehyde in 0.1 M phosphate buffer. A segment
of cornea from one patient (family 1, individual 1.3) was fixed in
2.5% glutaraldehyde containing 0.05% cuprolinic blue. For LM, tissue
was dehydrated through a graded series of ethanol, cleared with xylene,
embedded in paraffin wax, cut into 5-µm sections, and mounted onto
5-aminopropyltriethoxysilane-coated glass slides. All corneas were
stained with hematoxylin and eosin (H&E), Massons trichrome, Congo
red, and periodic acid-Schiff (PAS) stains.
For EM, tissue was dissected (segments < 1 mm3), immersed in 2.5% glutaraldehyde in 0.1 M phosphate buffer, and postfixed in 1% osmium tetroxide in 0.05 M phosphate buffer for 1 hour. The tissue was dehydrated through a graded ethanol series and embedded in Spurr (family 1) or epon (patient 2.1) resin. Tissue from family 1 was also fixed in 2.5% glutaraldehyde containing 0.05% cuprolinic blue (BDH Ltd, Poole, Dorset, UK) in a critical electrolyte concentration mode.9
Resin-embedded blocks were cut into semi-thin (1 µm) and thin (65 nm) sections. Semi-thin sections were stained with toluidine blue, and thin sections with uranyl acetate and lead citrate.10 Thin sections were placed onto nickel grids, and transmission EM was carried out using a JEOL 1010 instrument (London, UK).
Nucleic Acid Preparation and PCR Analysis
Genomic DNA was extracted by conventional methods from blood
lymphocytes.
Exons 4 and 12 of BIGH3 were amplified as previously reported.5 Genomic DNA (4 ng) was suspended in a 5 µl reaction containing 10 pmol of both primers, 40 µM each dCTP, dGTP, dTTP, 5 µM dATP, and 1x PCR buffer (containing 10 mM Tris-HCl, pH 8.3, 50 mM KCl, 11.5 mM MgCl2, and 0.1% gelatin), overlaid with mineral oil. The samples were heated to 96°C (10 minutes) and cooled to 51°C, and 0.15 units Taq polymerase was added. The samples were processed as follows: 92°C (30 seconds), 51°C (30 seconds), 72°C (30 seconds) x35 cycles, and 72°C (10 minutes). For SSCP, an equal volume of formamide stop solution was added to the amplified products. Gels were run at 350 V (4°C) and silver stained according to standard protocols. Direct sequencing was performed by dye terminator cycle sequencing (Perkin Elmer, Applied Biosystems Ltd., Foster City, CA) using a fluorescent sequencer (ABI 373).
| Results |
|---|
|
|
|---|
|
|
CDBII.
Family 3.
The proband, patient 3.1, presented in 1984 aged 27 years and underwent
left penetrating keratoplasty (PK) in 1986 (Fig. 1C)
. In 1993 visual
acuities were right 6/36 and left 6/18, and superficial honeycomb
changes were noted. She underwent right LK in 1996. Her son, 3.2,
presented in 1987 at the age of 5 years with early morning erosions.
Bilateral honeycomb changes were noted, and by 1999 visual acuities had
fallen to right 6/18 and left 6/12.
Family 4.
The proband, 4.1, presented at age 38 years with three of her
five children (4.24.4) in 1981 with acuities of 6/24 in each eye and
recurrent erosions since childhood. She had honeycomb reticular opacity
of both corneas and underwent left LK in 1992 and right LK in 1994.
Patient 4.2 presented at the age of 11 years with photophobia and
recurrent episodic soreness and honeycomb corneal opacities and has
undergone bilateral LK. Patient 4.3 presented at age 5 years with mild
corneal signs and recurrent sore eyes. By 1996 acuities were reduced,
troublesome erosions. He has undergone bilateral PTK, improving vision
to 6/9 with cessation of erosions. Patient 4.4 presented in his first
year with sore sticky eyes and by 1992 had "typical honeycomb"
changes in both corneas. He has now undergone bilateral PTK with
cessation of erosions. Patient 4.5, the brother of 4.1, presented at
age 43 years with bilateral superficial "cobweb" corneal opacities
and reported recurrent ocular irritation from infancy. By age 52 years,
his acuities were right 6/12 and left 6/18. Left PTK improved visual
acuity to 6/9 with cessation of erosions.
Family 5.
Patient 5.1 presented with a history of recurrent sore eyes since
infancy with recent deterioration of vision. Examination revealed
typical honeycomb anterior corneal dystrophy. Her father, two of three
siblings, two of five offspring, and four of seven grandchildren are
also affected. She has undergone multiple surgical procedures. Patient
5.2, her daughter, presented late with gross visual failure, a history
of recurrent erosions and widespread amorphous corneal opacification
underlying grossly irregular honeycomb subepithelial fibrosis.
Extensive PTK at this late stage did not significantly improve corneal
clarity.
Light and Transmission Electron Microscopy
CDBI.
LM revealed, in family 1, an irregular, partially dehiscent epithelium
with separation of cells in the wing cell layer and focal detachment
from the substratum. Condensed basal epithelial cells were seen, with
small intraepithelial deposits of material that was eosinophilic on H&E
staining and red with Masson trichrome. In patient 2.1, the epithelium
was intact, but intraepithelial deposits were also present. In all
patients, Bowmans layer was absent in areas (Figs. 2A
2B
). The granular dystrophy-type material was widely dispersed in
thin bands beneath the epithelium, above and below what remained of
Bowmans layer. Deposits did not demonstrate features of amyloid and
did not stain with PAS. In patient 2.1, who had undergone his first PK,
the deposits were also in middle and deep stroma.
|
|
CDBII.
LM revealed an irregular epithelium with focal areas of
vacuolation and partial dehiscence of basal cells from the underlying
substratum. There was a variably thick band of hyaline, avascular,
paucicellular connective tissue beneath the epithelium, largely
obliterating Bowmans layer (Fig. 2C)
. This material was eosinophilic
on H&E staining, did not stain red with the Masson trichrome stain, was
not Congophilic, and did not stain with PAS. The deposits were limited
to the subepithelial region, with an abrupt transition to anterior
stroma (e.g., Fig. 2C
), but in one case (family 3, patient 1), small
collections of this hyaline material were seen between cells in the
wing cell epithelial layer.
On EM in all families the epithelium appeared degenerate with vacuolated cells, aggregates of cytokeratin filaments and areas of poor attachment to substratum (Fig. 4A ). A broad band of subepithelial material was composed of irregular aggregates of curved ("curly") fibrils, 9 to 10 nm diameter. Intermingled with these fibrils at irregular intervals were groups of normal collagen fibers (Fig. 4B) . The abnormal material abutted against basal epithelial cells where basal lamina material was absent and extended between epithelial cells (Fig. 4C) . In such areas hemidesmosomes were absent or poorly formed. In other areas there were large numbers of hemidesmosomes overlying apparently intact basal laminar material. In one case (4.1) collections of curly fibers were seen totally enclosed by epithelial cells but within the extracellular compartment (Figs. 4C 4D) .
|
|
| Discussion |
|---|
|
|
|---|
In CDBI (families 1 and 2) there is a superficial, confluent form of granular dystrophy. The onset of corneal erosions was in infancy, and visual failure followed during childhood. Surgical intervention was necessary in all individuals, in some in the second decade of life, and was followed by recurrence within the graft. In CDBII, the onset of erosions was also within the first years of life, but visual deterioration was considerably slower than in CBDI. In our families, PTK successfully treated the erosions for prolonged periods, thereby delaying the need for surgical intervention.
In both CDBI and CDBII there were epithelial abnormalities on LM and EM.
In CDBI, epithelial dehiscence, lack of cell contact with abnormal hemidesmosome/basal lamina implies an abnormality of cell attachment. In recurrent dystrophy, deposits were between the epithelium and donor Bowmans membrane with intraepithelial involvement seen on LM and EM. Electron-dense, rod-shaped deposits were found within small vesicles inside and between epithelial cells, in some cases directly connected to deposits in Bowmans layer. Beneath the deposits, both Bowmans layer and the stroma (of the donor tissue) were normal. In CDBII, small collections of hyaline material were seen between epithelial cells on LM, and in one case collections of curly fibers were seen totally enclosed by epithelial cells.
These observations suggest that the epithelium is likely to be responsible for the production of the abnormal material, which aggregates as "rod-shaped" deposits or "hyaline material," supporting the work of others.12 13 This implies that the designation of these conditions as Bowmans layer dystrophies is incorrect because these are more likely to be dystrophies of epithelial origin. This is in keeping with similar findings for classical granular dystrophy.14
In CDBI the rod-shaped bodies are described as indistinguishable from those in granular dystrophy.11 15 16 However, we found a periodic, 9-nm cross-banding, which has not been described. CDBI and granular dystrophy both result from BIGH3 mutations (R124L and R555W, respectively). We suggest that both mutations cause aggregation of a mutant keratoepithelin protein, which is supported by the observation that antibodies to this protein bind to deposits in BIGH3-related dystrophies.17 18 19 However, the materials show distinct patterns of accumulation and, although they share ultrastructural features, this distinct banding pattern suggests that the material is not identical.
Our findings confirm that CDBII, or honeycomb dystrophy, is caused by the R555Q BIGH3 mutation. We have clinical and ultrastructural details of seven further families with this phenotype. This is, in our experience, the commoner condition, accounting in total for 9 of 11 Bowmans dystrophy families. Ultrastructural examination of patients with CBDII who underwent surgery demonstrated the presence of curly fibrils as described by others.4 20 21
Recent reports suggest that the R124L and R555W BIGH3 mutations do not
account for all forms of Bowmans layer dystrophies. Other
BIGH3-related mutations may cause similar phenotypes,
including a
F540 mutation8
described in a Sardinian
family. In addition there remain other families with an anterior
corneal dystrophy, whose exact phenotype is not clearly defined, which
does not map to 5q31. These include families, which, confusingly, have
also been designated CDBII or Thiel-Behnke dystrophy and show genetic
linkage to chromosome 10q24.22
The identification of mutations in the BIGH3 gene in a variety of anterior and stromal corneal dystrophies has enabled their morphologic manifestations to be correlated with the underlying molecular defect. Previous literature on Bowmans layer dystrophies has been confused in the application of eponymous titles to different phenotypes. The histopathologic and ultrastructural appearances that we report, together with their clinical correlations, should help to clarify this confusion and allow their reevaluation as disorders of epithelial origin.
| Acknowledgements |
|---|
| Footnotes |
|---|
Supported by the Wellcome Trust (51390/Z), by the Swiss National Science Foundation Grant 32-053750.98, and by Action Research (RP).
Submitted for publication November 12, 1999; revised March 14, 2000; accepted April 19, 2000.
Commercial relationships policy: N.
Corresponding author: Graeme C. M. Black, University Department of Medical Genetics and Regional Genetic Service, St. Marys Hospital, Hathersage Road, Manchester, M13 0JH, UK. gblack{at}man.ac.uk
| References |
|---|
|
|
|---|
F540) in a large cohort of Sardinian Reis-Bücklers corneal dystrophy patients Hum Mutat 12,215-216
This article has been cited by other articles:
![]() |
A. J. Aldave and B. Sonmez Elucidating the Molecular Genetic Basis of the Corneal Dystrophies: Are We There Yet? Arch Ophthalmol, February 1, 2007; 125(2): 177 - 186. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |