Ophthalmic Genetics, 29:53–59, 2008Copyright c© Informa Healthcare USA, Inc.ISSN: 1381-6810 (print) / 1744-5094 (online)DOI: 10.1080/13816810802008242

RESEARCH ARTICLE

The D144E Substitution in the VSX1 Gene: ANon-pathogenic Variant or a Disease Causing Mutation?

Pras EranDepartment of Ophthalmology, Assaf Harofeh Medical Center, Zerifin, Israel andSackler Faculty of Medicine, Tel Aviv University, Tel Aviv, Israel

Abu AlmogitSheba Medical Center, Danek Gartener Institute of Human Genetics, Tel Hashomer, Israel

Zadok DavidDepartment of Ophthalmology, Assaf Harofeh Medical Center, Zerifin, Israel andSackler Faculty of Medicine, Tel Aviv University, Tel Aviv, Israel

Haike Reznik WolfSheba Medical Center, Danek Gartener Institute of Human Genetics, Tel Hashomer, Israel

Garzozi HanaDepartment of Ophthalmology, Bnei Zion Hospital, Haifa, Israel andRappaport Faculty of Medicine, Technion Institute of Technology, Haifa, Israel

Barkana YanivDepartment of Ophthalmology, Assaf Harofeh Medical Center, Zerifin, Israel andSackler Faculty of Medicine, Tel Aviv University, Tel Aviv, Israel

Pras ElonSackler Faculty of Medicine, Tel Aviv University, Tel Aviv, Israel andSheba Medical Center, Danek Gartener Institute of Human Genetics, Tel Hashomer, Israel

Avni IsaacDepartment of Ophthalmology, Assaf Harofeh Medical Center, Zerifin, Israel andSackler Faculty of Medicine, Tel Aviv University, Tel Aviv, Israel

Purpose: To identify the genetic defect associated with keratoconus (KC) in an Ashkenazi Jewishfamily and to evaluate its nature and its phenotypic expression within carriers. Methods: A threegeneration Ashkenazi Jewish family with KC was ascertained. Diagnosis was based on clinical ex-amination and corneal topography. Segregation analysis was performed using micro-satellite poly-morphic markers in close proximity to 7 previously associated KC loci and genes. Mutation analysisof the VSX1 gene was performed by direct sequencing of PCR-amplified exons, and a BseR1 restric-tion assay. In selected cases, where the genotype was consistent with KC, additional effort to detectsubtle corneal changes was made by computerized Orbscan measurements. Results: We found co-segregation between the KC phenotype and a polymorphic marker close to the VSX1. Sequencingrevealed a previously described missense mutation (D144E). All of the mutation carriers manifestedpathologic corneal findings; some had overt KC while others had subtle corneal alterations identifi-able only by Orbscan. Conclusions: These findings support the pathogenic role of VSX1 gene in KC.The variable expression among the carriers, suggests the involvement of other factors in determiningthe final phenotype.

Keywords VSX1; keratoconus; complex inheritance

Received 15 August 2007; Accepted 21 February 2008.Address correspondence to Eran Pras, MD, Department of Ophthalmology, Assaf Harofeh Medical Center, Zerifin, 70300, Israel. E-mail:eranpras@gmail.com

53

Oph

thal

mic

Gen

et D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y C

olum

bia

Uni

vers

ity o

n 12

/08/

14Fo

r pe

rson

al u

se o

nly.

54 P. ERAN ET AL.

INTRODUCTIONKeratoconus (KC)[MIM 148300] is a non-inflammatory pro-

gressive thinning disorder of the center of the cornea, with acomplex underlying basis.1 Corneal thinning causes irregularityin its curvature inducing myopia, astigmatism and corneal pro-trusion, leading to a variable degree of visual impairment.2,3 KConset is often first manifested at puberty, and typically graduallyprogresses until the third or fourth decade of life when it arrests.When severe, visual impairment may require corneal transplan-tation and thus in developed countries KC has become a leadingindication for corneal transplantation.

The reported prevalence of overt KC is about 1 per 20003

but the fraction of unidentified sub-clinical cases within theoverall genetic susceptible population remains unclear. KC ishighly variable in the phenotypic expression and in the tim-ing of disease onset. Hence, detection of mild disease dependson the sensitivity of the diagnostic criteria and early or sub-clinical cases (forme fruste KC) may often be missed. MostKC cases are sporadic; however evidence for a strong un-derlying genetic background arises from epidemiologic stud-ies that document positive family history in 6–23.5% of thepatients, and a high concordance of KC among monozygotictwins.4

Various approaches for identifying KC susceptibility geneshave been applied, including linkage mapping of six putativeloci on chromosomes 16q,5 15q22.23-q24.2,6 3p1-q13,7 2p24,8

5q14.3-q21.1,9 and 20q12.10 In most of the familial cases pene-trance is far less then 100% and so it became evident that otherfactors are implicated in disease pathogenesis. Eye rubbing, forexample, has long been recognized as an important risk factor,11

and chronic mechanical micro-trauma has been proposed as anunderlying mechanism for the association between KC, atopickeratoconjunctivitis12 and contact lens wear.13

One of the genes that has recently been implicated in KC is thevisual system homeobox gene (VSX1) located on chromosome20p11 [MIM 605020].14−17 This gene is highly expressed in thecornea18 and sequence analysis has identified several DNA vari-ants in KC patients (http://www.us.expasy.org/uniprot/Q9NZR4as provided in the public domain by the EMBL). However theophthalmic research community debates whether these changescan be regarded as disease-related mutations or non pathogenicsequence variants. This debate arises mainly due to the detectionof these variants in unaffected family members with a normalanterior surface corneal topography (ASCT).

In this study we found co-segregation between the diseaseand a mutation in the VSX1 gene in a multi generation KCfamily. Genetic analysis of this family in conjunction with ad-vanced clinical assessment which included corneal thicknessand posterior surface mapping granted us with a unique op-portunity to demonstrate corneal manifestations in all mutationcarriers. These vary from severe KC in one end, via “formefruste KC,” to apparently asymptomatic carriers at the otherend.

METHODS

Clinical Ascertainment and DNA SpecimensA three generation Ashkenazi Jewish family with kerato-

conus (Figure 1A) was recruited at the Assaf Harofeh MedicalCenter, Zerrifin, Israel, as part of a broader effort to identify dis-ease loci causing inherited visual diseases. The study protocolwas approved by the Seba Medical Center Helsinky committee,and the participants signed an informed consent form.

All family members underwent a comprehensive ocular andsystemic evaluation comprised of slit lamp biomicroscopy, di-lated retinoscopy, and computer-assisted videokeratography bya trained ophthalmologist with special attention for the presenceof any previously reported ocular or systemic abnormality as-sociated with KC.2 The diagnosis of KC was based on previouscorneal transplant or conical protrusion of the cornea associ-ated with videokeratographic findings compatible with KC.19 Inselected cases (family members 5 and 9), where the genotypewas consistent with KC additional effort to detect subtle cornealchanges was made by using computerized Orbscan (Bausch &Lomb) measurements.20 Heparinized blood was obtained fromeach participant for genomic DNA isolation.

GenotypingWe used polymorphic microsatellite markers in close

proximity to7 candidate loci previously implicated in KCformation5−10,14 (Table 1), using polymerase chain reaction(PCR). Amplification of the polymorphic markers was carriedout in a 25 µl reaction containing 50 ng of DNA, 13.4 ng ofeach unlabeled primer, 1.5 mM dNTP’s, 0.08 µg labeled primerin 1.5 mM MgCl2, PCR buffer, with 1.2 U of Taq polymerase(Bio-Line, London). After an initial denaturation of 5 min at950, 30 cycles were performed (94◦ for 2 min, 57◦ for 3 min,

TABLE 1Microsatellite markers

Chromosome Marker

2p24 D2S3052p24 D2S23733p14-q13 D3S12853p14-q13 D3S12713p14-q13 D3S12785q15 D5S15285q15 D5S238715q22-15q24 D15S52015q22-15q24 D15S636E15q22-15q24 D15S21116q22.2 D16S262416q22.3 D16S276116q23.1 D16S309020p11.21 D20S104320q12-13 D20S119

Oph

thal

mic

Gen

et D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y C

olum

bia

Uni

vers

ity o

n 12

/08/

14Fo

r pe

rson

al u

se o

nly.

D144E SUBSTITUTION IN THE VSX1 GENE 55

and 72◦ for 1 min), followed by a final extension of 7 minat 72◦. PCR products were electrophoresed on an automatedABI Prism 3100 Genetic Analyzer (Perkin Elmer). The alleleswere analyzed using the ‘Genscan-3.1 Genotyper-2.1, software(ABI).

Sequencing and Restriction Assay of the VSX1 GeneGenomic DNA was amplified by specific primer pairs for

each of the five VSX1 coding exons as well as exon-intronboundaries.21 PCR amplicons of two affected family members(individuals 3,6) were amplified using the conditions describedabove and sequenced in both directions using a commercial se-quencing service (Hy Laboratories, Ltd. (Hylabs) Park Tamar,Rehovot, Israel). The VSX1 exon-2 mutation was confirmed bytesting for the presence of a BseR1 restriction site in the mu-tated allele. Two hundred fifty three nucleotides of the ampli-cone encompassing exon-2 were amplified as described aboveand digested with BseR1, to produce two fragments of 105 and148 bp in the mutated allele, but only a 253 bp amplicone inthe wild type allele. One hundred and four unrelated AshkenaziJewish control DNA samples were also analyzed with the BseR1restriction assay.

RESULTS

Initial Patient AssessmentClinical information on 10 family members who took part

in the study is summarized in Table 2. A general physical ex-

amination did not reveal any additional abnormal finding. Oph-thalmic examinations disclosed iris heterochromia due to con-genital Horner syndrome in family member no. 5, vernal kera-toconjunctivitis in family member no. 10, and mild myopia infamily members 2 and 8. Four family members (individuals 1,3, 6, and 10) were initially classified as affected on the basis ofprior corneal transplant surgery (individuals 1 and 6) or typicalvideokeratographic patterns (individuals 3 and 10).

Interestingly, two different protrusion patterns were docu-mented (Figure 2, upper measurements). Fifty six years old pa-tient no. 3 showed inferior steep (I-S) corneal shape exceeding50 diopters of curvature in his non-operated left eye, while his18 years niece (individual no.10) demonstrated an asymmet-ric bowtie pattern with skewing of the radial axes (AB/SRAX)whose central corneal curvatures exceeded 56.0 diopters.

Molecular Studies and the Identification of the D144EMutation in VSX1

Using candidate gene approach, we genotyped all membersof the family with polymorphic PCR markers in close proximityto known KC loci. Taking into consideration the complexityin KC mode of inheritance, we hypothesized that all affectedfamily members should share a common allele of a polymorphicmarker close to the disease susceptibility gene. We were able toexclude 6 previously reported KC loci5−10 due to lack of allelesharing in the patients (data not shown). Cosegregation with theKC trait was noticed for the marker D20S1043 which is located

TABLE 2Clinical data of family members

Age atID # last exam Eye exam VK pattern Affectation status Genotype

1 84 S/P BE PKP Corneal grafts Keratoconus D144E/+2 52 Mild Myopia Normal Normal +/+3 56 S/P RE PKP

LE inferior steep coreal ectasia,iron deposit

RE: corneal graftLE: KeratoconusI-S pattern (Fig. 2)

Keratoconus D144E/+

4 25 Normal Normal Normal +/+5 21 Rt > Lt high WTR astigmatism iris

heterochromiaRegular WTR

astigmatism (Fig. 2)High WTR regular astigmatism

Irregular posterior corneal surfaceD144E/+

6 52 S/P BE PKP Corneal grafts Keratoconus D144E/+7 47 Normal Normal Normal +/+8 12 Mild myopia and WTR astigmatism Regular mild WTR

astigmatismMild WTR astigmatism +/+

9 19 Normal Normal (Fig. 2) Thin cornea D144E/+10 18 Vernal conjunctivities nipple shape

corneal protrusionKeratoconus-(Fig. 2)

AB/SRAX patternKeratoconus D144E/+

+—normal allele, WTR—‘with the rule’ astigmatism, RE—right eye, LE—left eye, BE—both eyes, AB/SRAX—asymmetric bowtie patternwith skewing of the radial axes, VK—Videokeratography.

Oph

thal

mic

Gen

et D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y C

olum

bia

Uni

vers

ity o

n 12

/08/

14Fo

r pe

rson

al u

se o

nly.

56 P. ERAN ET AL.

FIG. 1. Molecular Study for VSX1. A- Family pedigree and genotyping for the marker D20S1043. Filled symbols reflect obviousKC, clear symbols reflect unaffected status, and “?” reflect subtle corneal changes identified by Orbscan measurements. B-DNA sequence electropherogram of a normal control patient (up), and KC patient (bottom); a C-to-G change in the patient (arrow),results with an amino acid substitution D144E, of the VSX1. C-BseR1 restriction digest of amplified DNA at the mutation site inthe family. The 253-bp DNA segment is cleaved into two segments (148 and 105 bp) in the mutated allele.

in close proximity to the VSX1 gene (Figure 1A, allele-1). Thisallele is present in all four affected family members and in twoasymptomatic family members with a normal anterior topogra-phy (ASCT) (individuals 5 and 9). Sequencing the VSX1 codingregion gene disclosed a heterozygous C→G substitution atposition 696 of the cDNA, resulting in a change of a highly con-served aspartic acid at position 144, to glutamic acid (D144E)(Figure 1B). This change introduces a BseR1 restriction site inthe mutated allele. In concordance with the polymorphic markerresults this restriction site was present in the 4 affected KCfamily members and in the two apparently asymptomatic sibs(Figure 2C). Using the BseR1 restriction assay we haveidentified one carrier of the D144E mutation in 104 Ashkenazicontrols.

Advanced Patient AssessmentThe two asymptomatic carriers of the D144E mutation, who

demonstrated regular corneal videokeratographies (Figure 2lower measurements), were further evaluated by Orbscan mea-surements (Figure 3) in an effort to detect subtle corneal changes.Both individuals demonstrated subtle abnormalities unidenti-fiable by their anterior surface topography. Individual no. 5(Figure 3) whose topography showed regular astigmatism ex-hibited in his Orbscan standard pachymetric map with cen-tral corneal thickness exceeding 550 mic (RE-562 mic, LE-556mic), however posterior corneal surface of his right eye demon-strated an abnormal “posterior float” approximating 50 mic. Hiscousin (Figure 3, no. 9), demonstrated both anterior and poste-rior corneal surface maps within normal range but a relativelythin pachymetry with central corneal thickness below 500 mic(RE-453 mic, LE-462 mic).

DISCUSSIONDespite many decades of research, little is known about

the precise molecular defects and abnormal biochemical path-ways that cause keratoconus. Six putative KC loci have beenmapped,5−10 but in none was the disease gene cloned. In the

present study we screened a large family with polymorphicmarkers. Segregation was found with a marker close to the VSX1gene and sequencing lead to the identification of the D144E mu-tation. Despite seven different VSX1 sequence variations pre-viously identified in KC families, the role of this gene in thepathogenesis of KC remains controversial.14−17 D144E is con-sidered by some researchers pathogenic,15 while others regardit as a non pathogenic change.17 The discrepancy arises mainlyfrom the finding of asymptomatic carriers of this mutation. Twolines of evidence make us favor the view that D144E does havea pathogenic role:

1. Aspartic acid at position 144 of VSX1 is highly conservedthroughout evolution.

2. The reported frequencies of D144E substitution are muchhigher in KC patients and their relatives than it is in controlpopulation.14−17

Interestingly, no functional studies have been performed forD144E. Such studies have been done for another sequence varia-tion, R166W, and these have demonstrates impairment of VSX1-mediated activity.22

To date, attempts to diagnose latent KC in asymptomatic car-riers have been limited to ASCT evaluation with videokertogra-phy. Indeed, in the family presented in this study, ASCT in thetwo asymptomatic carriers did not show any findings compati-ble with KC. Only careful evaluation of these carriers by Orb-scan revealed a suspiciously thin pachymetric map in one fam-ily member (individual no. 9), and an irregular posterior cornealsurface (individual no. 5) in the other. These findings most prob-ably represent very mild corneal changes, at the very end of theKC spectrum, and support the notion that current KC diagnosticcriteria which are based on clinical findings and videokeratog-raphy indices alone19 should be revised to include pachymetricand posterior corneal surface maps. Such measures will improvediagnostic sensitivity, and the detection of latent KC.

Videokeratographic studies available from mutation carri-ers demonstrate a continuum from apparently normal to clearly

Oph

thal

mic

Gen

et D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y C

olum

bia

Uni

vers

ity o

n 12

/08/

14Fo

r pe

rson

al u

se o

nly.

D144E SUBSTITUTION IN THE VSX1 GENE 57

FIG. 2. Videokeratographic measurements of D144E bearing family members. Upper measurements: Two different keratoconiccorneal patters showing; inferior steep (ind. 3) and AB\SRAX (ind. 10). Lower measurements: Normal range anterior surfaceshapes; regular dome (ind. 9) and with-the-rule astigmatism (ind. 5).

pathologic corneas; ordinary corneal shapes of regular with-the-rule astigmatism (individual no. 5) and a normal dome (individ-ual no. 9) in asymptomatic carriers, to inferior steep (individualno. 3) and irregular astigmatism with skewing of the radial axes(individual no. 10). The wide clinical spectrum among carriers ofthe same mutation may reflect progression with age, modifyinggenes, and environmental factors. Interestingly, atopic kerato-

conjunctivitis, one of the few suspected risk factors for KC, waspresent in the family member no. 10 who was severely affected.

Another acknowledged risk factor for disease accelerationis surgical trauma. Patients with KC are considered to havea bio-mechanically unstable cornea prone to develop forwardcorneal bowing post surgery (e.g., corneal ectsia). Therefore theyare regarded as inappropriate candidates for LASIK surgery.23

Oph

thal

mic

Gen

et D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y C

olum

bia

Uni

vers

ity o

n 12

/08/

14Fo

r pe

rson

al u

se o

nly.

58 P. ERAN ET AL.

FIG. 3. Orbscan measurements of asymptomatic carriers. Ind. 9 demonstrates thin corneal, with pachymetric measurements below470 microns Ind. 5 demonstrates in his right eye “posterior-float” measurements approximating 50 mic.

One of the greatest challenges in the field of refractive surgeryis to be able to identify prone individuals in an attempt to re-duce the incidence of this complication. Anterior surface cornealtopography (ASCT) videokeratography has been considered asa possible tool for the identification of such patients at risk.19

Our study shows that ASCT may not be sensitive enough andthat Orbscan may be superior for the detection of early, sub-tle changes. The identification of more genetic variants as-sociated with KC may provide better selection for refractivesurgery candidates, and thus may reduce post surgical ectaticcomplications.

REFERENCES1. Loughnan MS, Snibson GR. Understanding keratoconus. Clin Ex-

periment Ophthalmol. 2001;29:339.2. Krachmer JH, Feder RS, Belin MW. Keratoconus and related

noninflammatory corneal thinning disorders. Surv Ophthalmol.1984;28:293–322.

3. Rabinowitz YS. Keratoconus Surv Ophthalmol. 1998;42:297–319.4. Edwards M, McGhee CN, Dean S. The genetics of keratoconus.

Clin Experiment Ophthalmol. 2001;29:345–351.

5. Tyynismaa H, Sistonen P, Tuupanen S, et al. A locus for autoso-mal dominant keratoconus: linkage to 16q22.3-q23.1 in Finnishfamilies. Invest Ophthalmol Vis Sci. 2002;43:3160–164.

6. Hughes AE, Dash DP, Jackson AJ, et al. Familial keratoconus withcataract: linkage to the long arm of chromosome 15 and exclu-sion of candidate genes. Invest Ophthalmol Vis Sci. 2003;44:5063–5066.

7. Brancati F, Valente EM, Sarkozy A, et al. A locus for autosomaldominant keratoconus maps to human chromosome 3p14-q13. JMed Genet. 2004;41:188–192.

8. Hutchings H, Ginisty H, Le Gallo M, et al. Identification of anew locus for isolated familial keratoconus at 2p24. J Med Genet.2005;42:88–94.

9. Tang YG, Rabinowitz YS, Taylor KD, et al. Genomewide linkagescan in a multigeneration Caucasian pedigree identifies a novellocus for keratoconus on chromosome 5q14.3-q21.1. Genet Med.2005;7:397–405.

10. Fullerton J, Paprocki P, Foote S, et al. Identity-by-descent approachto gene localization in eight individuals affected by keratoconusfrom north-west Tasmania, Australia. Hum Genet. 2002;110:462–470.

11. Krachmer JH. Eye rubbing can cause keratoconus. Cornea.2004;23:539–540.

Oph

thal

mic

Gen

et D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y C

olum

bia

Uni

vers

ity o

n 12

/08/

14Fo

r pe

rson

al u

se o

nly.

D144E SUBSTITUTION IN THE VSX1 GENE 59

12. Bawazeer AM, Hodge WG, Lorimer B. Atopy and keratoconus: amultivariate analysis. Br J Ophthalmol. 2000;84:834–836.

13. Phillips CI. Contact lenses and corneal deformation: cause, corre-late or co-incidence? Acta Ophthalmol. 1990;68:661–668.

14. Heon E, Greenberg A, Kopp KK, et al. VSX1: a gene for poste-rior polymorphous dystrophy and keratoconus. Hum Mol Genet.2002;11:1029–1036.

15. Bisceglia L, Ciaschetti M, De Bonis P, et al. VSX1 mutationalanalysis in a series of Italian patients affected by keratoconus: de-tection of a novel mutation. Invest Ophthalmol Vis Sci. 2005;46:39–45.

16. Mintz-Hittner HA, Semina EV, Frishman LJ, et al. VSX1 (RINX)mutation with craniofacial anomalies, empty sella, corneal en-dothelial changes, and abnormal retinal and auditory bipolar cells.Ophthalmology. 2004;111:828–836.

17. Aldave AJ, Yellore VS, Salem AK, et al. No VSX1 gene mu-tations associated with keratoconus. Invest Ophthalmol Vis Sci.2006;47:2820–2822.

18. Semina EV, Mintz-Hittner HA, Murray JC. Isolation and charac-terization of a novel human paired-like homeodomain-containingtranscription factor gene, VSX1, expressed in ocular tissues. Ge-nomics. 2000;63:289–293.

19. Rabinowitz YS. Videokeratographic indices to aid in screening forkeratoconus. J Refract Surg. 1995;11:371–379.

20. Auffarth GU, Wang L, Volcker HE. Keratoconus evaluation us-ing the Orbscan Topography System J Cataract Refract Surg.2000;26:222–228.

21. Rozen S, Skaletsky H. Primer-3 on the WWW for general usersand for biologist programmers. Methods Mol Biol. 2000;132:365–386.

22. Dorval KM, Bobechko BP, Ahmad KF, et al. Transcriptional activ-ity of the paired-like homeodomain proteins CHX10 and VSX1. JBiol Chem. 2005; 280:10100–10108.

23. Randleman JB, Russle B, Ward MA, et al. Risk factors and prog-nosis for corneal ectasia after LASIK. Ophthalmology. 2003;110:267–275.

Oph

thal

mic

Gen

et D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y C

olum

bia

Uni

vers

ity o

n 12

/08/

14Fo

r pe

rson

al u

se o

nly.