Human Molecular Genetics, 2008, Vol. 17, No. 5 doi:10.1093/hmg/ddm342 Advance Access published on November 23, 2007

710–716

Ancestral LOXL1 variants are associated with pseudoexfoliation in Caucasian Australians but with markedly lower penetrance than in Nordic people Alex W. Hewitt1,2,{, Shiwani Sharma1,{, Kathryn P. Burdon1, Jie Jin Wang3, Paul N. Baird2, David P. Dimasi1, David A. Mackey2,4, Paul Mitchell3 and Jamie E. Craig1, 1

Flinders Medical Centre, Department of Ophthalmology, Flinders University, Adelaide, Australia, 2Centre for Eye Research Australia, University of Melbourne, Royal Victorian Eye and Ear Hospital, Melbourne, Australia, 3Centre for Vision Research, Department of Ophthalmology and Westmead Millennium Institute, University of Sydney, Westmead, Australia and 4Department of Ophthalmology, Royal Hobart Hospital, University of Tasmania, Hobart, Australia Received October 23, 2007; Revised and Accepted November 21, 2007

Pseudoexfoliation syndrome is a generalized disorder of the extracellular matrix, characterized by the pathological accumulation of abnormal fibrillar material in the anterior segment of the eye predisposing to glaucomatous optic neuropathy. We investigated the role of lysyl oxidase-like 1 (LOXL1) sequence variation in a Caucasian Australian population-based cohort of 2508 individuals, 86 (3.4%) of whom were diagnosed with pseudoexfoliation syndrome. Two non-synonymous variants in exon 1 of LOXL1 (Arg141Leu;Gly153Asp) were found to be strongly associated with pseudoexfoliation. Two copies of the high risk haplotype at these singlenucleotide polymorphisms conferred a risk of 7.20 (95%CI: 3.04–20.75) compared with no copies of the high risk haplotype. Each of the disease-associated alleles is by far commoner in the normal population, and examination of cross-species homology reveals that the two disease-associated coding variants belong to the ancestral version of the gene. LOXL1 was found to be expressed by reverse transcription–polymerase chain reaction in all ocular tissues examined except retina. The presence of LOXL1 protein in ocular tissues of interest was demonstrated by western blotting. Specific bands of 130 and 80 kDa, representing polymerized protein forms, were detected in the cornea, iris, ciliary body, lens capsule and optic nerve. The 42 kDa mature form of LOXL1 was detected in the iris and ciliary body. Our Caucasian population has a 9-fold lower lifetime incidence of pseudoexfoliation syndrome compared with Nordic populations despite having similar allelic architecture at the LOXL1 locus. This strongly suggests that as yet unidentified genetic or environmental factors independent of LOXL1 strongly influence the phenotypic expression of the syndrome.

BACKGROUND The allelic architecture for many human diseases is being rapidly uncovered. Improved access to array-based technology, in conjunction with the Human Genome and International HapMap Projects, have permitted a comprehensive examination of the genetic underpinnings of many traits

important to human health (1 – 4). Results from the opening barrage of genome-wide investigations have begun to answer the question of the contribution provided by common genetic variants in the aetiology of common disease. Proponents of the common disease– common variants hypothesis,



To whom correspondence should be addressed at: Department of Ophthalmology, Flinders University, Flinders Drive, Bedford Park, Adelaide 5042, South Australia. Tel: þ61 8 8204 4624; Fax: þ61 8 8277 0899; Email: [email protected] †These authors contributed equally to this work.

# The Author 2007. Published by Oxford University Press. All rights reserved. For Permissions, please email: [email protected]

Human Molecular Genetics, 2008, Vol. 17, No. 5

which predict that genes responsible for the majority of common diseases have relatively simple allelic spectra, have argued that if the loci contributing to common disease have a moderately sized set of disease alleles, the allelic spectra should be simple (5,6). Conversely, given that variants predisposing to common complex diseases generally appear to have lower penetrance and weaker selection pressures than rarer Mendelian disease-associated alleles, some have argued that they are likely to be underpinned by considerably greater degrees of allelic heterogeneity (7,8). Herein, we describe a locus-specific contribution made to a disease that is relatively uncommon in our study population, but find a remarkably similar allelic architecture to that of the Nordic population in which the disease is extremely common. Pseudoexfoliation syndrome (OMIM:177650) is a generalized disorder of the extracellular matrix characterized clinically by the pathological accumulation of abnormal fibrillar material in the anterior segment of the eye (9). In addition to being a significant risk factor for glaucomatous optic neuropathy, pseudoexfoliation syndrome has also been associated with lens zonule weakness, cataract formation, and systemic vascular complications (9 – 12). The prevalence of pseudoexfoliation syndrome varies markedly between populations, with the highest rates (up to 40%) being documented in people residing in Nordic countries, whilst Anglo-Celtic Caucasians have a remarkably lower prevalence (13 – 15). Recently Thorleifsson et al. (16) performed a genome-wide scan of Icelandic patients with glaucoma and pseudoexfoliation syndrome. They identified two non-synonymous singlenucleotide polymorphisms (SNPs) in exon 1 of the lysyl oxidase-like 1 (LOXL1) gene which together conferred a population-attributable risk in Icelandic and Swedish individuals of 99% (16). Interestingly, the highest risk diplotype was identified in approximately a quarter of the general population who had not been clinically examined; a not unexpected finding given the high prevalence of pseudoexfoliation in their population (16). We investigated the role of LOXL1 sequence variation in a large Caucasian population recruited through the Blue Mountains Eye Study (BMES), a population-based study of 4838 individuals aged 49 years or older, who had a comprehensive baseline eye examination, including a specific examination for pseudoexfoliation syndrome. The two previously identified coding variants in LOXL1 were found to be strongly associated with pseudoexfoliation syndrome. However, despite our study population’s remarkably lower lifetime incidence of the disease, the underlying allelic architecture at LOXL1 was almost identical to that described by Thorleifsson et al. (16) in Iceland and Sweden. We also investigated LOXL1 expression directly in all ocular tissues for the first time by reverse transcription – polymerase chain reaction (RT – PCR) and western blotting, demonstrating expression in all tissues examined aside from the retina.

METHODS Subject recruitment Subjects participated in the BMES, as previously described (17,18). In brief, the BMES is a population-based cohort

711

study investigating the etiology of common ocular diseases among suburban residents aged 49 years or older, living in the Blue Mountains region, west of Sydney, Australia. The population in this area is stable and ethnically homogeneous of predominantly Anglo-Celtic descent. Subjects were recruited during one of four surveys between 1992 and 2004. The baseline BMES survey was conducted between 1992 and 1993, recruiting a total of 3654 participants (82.4% of 4433 eligible persons identified in a private census). Of these people, 2564 (70.2%) were re-examined during the 5- and 10-year follow-up studies. An ancillary study conducted between 1998 and 2000 examined an additional 1174 people who had either reached the eligible age (49þ years) for participation or had relocated into the study area (85.2% of 1378 newly eligible persons identified in a second private census). DNA samples were obtained during the 5-year follow-up and ancillary surveys. Approval for this study was obtained from the relevant Human Research Ethics Committees of the Westmead Millennium Institute at the University of Sydney, as well as from Flinders Medical Centre and Flinders University. Pseudoexfoliation syndrome was diagnosed at slitlamp examination by ophthalmologists as part of a comprehensive ocular examination, including pupil dilation. Given the inherent difficulties in detecting pseudoexfoliation after cataract surgery, participants who had undergone bilateral cataract surgery were considered unclassifiable. Analysis was performed comparing the diagnosed pseudoexfoliation cases against both the subpopulation in whom pseudoexfoliation had been clinically excluded and the total unselected population. Glaucoma diagnosis was strictly based on concordant findings of typical glaucomatous visual field defects on the Humphrey 30-2 test together with corresponding optic disc rim thinning, including an enlarged cup–disc ratio (0.7) or cup–disc ratio asymmetry (0.3) between the two eyes (17). Genotyping and data analysis Using the software program Tagger, SNPs across LOXL1 were selected on the basis of linkage disequilibrium patterns within European people from the Centre d’E´tude du Polymorphisme Humain from Utah (CEPH), as part of the International HapMap Project (3,19). Twelve tagging SNPs, which captured all alleles with an r 2 of 0.8, were selected and genotyped (Fig. 1). This included the forced selection of SNP rs3825942 which had been implicated as being associated with pseudoexfoliation by the work of Thorleifsson et al. (16). The additional previously identified disease-associated SNP (rs1048661), which was not included as part of the International HapMap Project, was also specifically genotyped in our population (16). Genotyping was performed with the use of Sequenom iPLEX GOLD chemistry on an Autoflex Mass Spectrometer at the Australian Genome Research Facility (Queensland, Australia). Map coordinates provided are those from National Center for Biotechnology Information Build 36 (August 2007). The SNP name designations were obtained from dbSNP and HapMap. Individuals with .5% missing genotypes were excluded from analysis. SNP genotyping in control samples was checked for compliance with the Hardy – Weinberg equilibrium by using Haploview 4.0 (20).

712

Human Molecular Genetics, 2008, Vol. 17, No. 5

Figure 1. Association analysis of the LOXL1 gene. (A) Displays the x2 allelic association between cases and all unselected controls at each SNP and their relative location. The dashed line represents the Bonferroni adjusted level for significance. Pairwise D0 and r 2 values are displayed in panel (B) and the OR as well as P-values for important haplotypes as well as diplotypes compared with all other haplotypes or diplotypes, are revealed in panel (C). The estimated frequency (f) in cases and controls is also shown. ( ) Individuals with the high risk diplotype were compared with those with no copies of the high risk haplotype.

Haplotypes and diplotypes for each individual were estimated using the expectation maximization algorithm in HAPLO.STATS. We performed association analysis using contingency tables of individual SNPs, haplotypes, and diplotypes using Haploview 4.0, SNPStats and SPSS (v14.0 SPSS Inc., Chicago, IL, USA) (20,21). Given that the results did not change considerably when only the pseudoexfoliation-free controls were used, data are presented for the total unselected control cohort. An online genetics power calculator was used to estimate the power of this study, after considering a variety

of effect sizes and allele frequencies (22). Assuming a multiplicative genetic model and a heterozygous odds ratio (OR) of 1.5, our study of unselected cases had a power of 90% to detect a disease-associated allele with a population frequency of at least 0.50. Homology modelling The LOXL1 protein sequence of humans and other animal species was obtained from the Ensembl database (accession

Human Molecular Genetics, 2008, Vol. 17, No. 5

713

Table 1. Demographic features of the study population. Pseudoexfoliation status was not classifiable if participants had previously undergone cataract surgery. The unselected control group comprises both the subjects who had no clinical signs of pseudoexfoliation and those whose clinical status was unclassifiable due to previously undergoing cataract surgery

Total number Female (%) Mean age (SD) years Number glaucoma (%) 

Cases (pseudoexfoliation)

Controls without pseudoexfoliation

Pseudoexfoliation not classifiable

Total unselected controls

86 54 (62.8) 76.4 (8.1) 11 (12.8)

2087 1152 (55.2) 68.6 (10.0) 79 (3.8)

335 221 (66) 77.3 (8.7) 30 (9.0)

2422 1373 (56.7) 69.9 (10.3) 109 (4.5)

P , 0.001 compared with case subjects.

codes available upon request). Protein alignments were performed using CLUSTALW, with a BLOSUM-62 proteinweighted matrix; a gap open penalty score of 10 and a gap extension penalty score of 0.05 (23).

Ocular expression and western blotting Post-mortem human eyes were obtained through the Eye Bank of South Australia following the ethical guidelines of the Clinical Research Committee of Flinders Medical Centre, Australia. Because human corneal tissue is reserved for use in transplantation, mouse cornea was used. Mouse corneal tissue was obtained following the ethical guidelines of the Animal Welfare Committee at Flinders University, Australia. Human lens epithelium cell line SRA 01/04 was provided kindly by Dr Venkat Reddy from the Kellogg Eye Institute, Michigan, USA. SRA 01/04 and HEK (Human Embryonic Kidney) 293A cells were maintained in Dulbecco’s modified Eagle’s medium (DMEM) (GIBCO, Invitrogen Pty Ltd, Victoria, Australia) supplemented with 10% foetal bovine serum and penicillin/streptomycin in a humidified atmosphere at 378C and 5% CO2. Total RNA from human and mouse ocular tissues and cultured mammalian cells was extracted using the RNeasy mini kit (Qiagen Pty Ltd, Victoria, Australia) in accordance with the manufacturer’s protocol. First strand cDNA synthesis was performed with Superscript III (Invitrogen Pty Ltd, Victoria, Australia) using random hexamers. Reverse transcription of human retina and lens RNA was performed using an oligo-dT primer. RT – PCR for LOXL1 was carried out with 50 CAGCGC GTGAAGAACCAGGGCACA (forward) and 50 ATGCATG AATTCTGAGGCCGGCAGGGAGGGATGC (reverse) primers, respectively located in exon 3 and 7, using the GC-RICH PCR system (Roche Diagnostics Pty Ltd, New South Wales, Australia) at 958C-3 min; 958C-30 s, 608C-30 s, 688C-30 s for 35 or 45 cycles; 688C-7 min. The reverse primer had an extension of 11 nucleotides at the 50 end to introduce a restriction enzyme site in the PCR product for later cloning. Ocular tissues were homogenized in 6M urea, 2% DTT, 2% CHAPS and 0.1% SDS buffer using the TissueLyser (Qiagen Pty Ltd, Victoria, Australia) for protein extraction. Protein concentration was estimated by the Bradford method (24). Thirty microgram of each protein extract was size fractionated by SDS – PAGE and transferred onto Hybond C Extra (GE Healthcare Bio-Sciences Pty Ltd, New South Wales, Australia). Western blot was hybridized with 1:500 dilution of the goat anti-human LOXL1 primary antibody (Santa Cruz

Biotechnology Inc., CA, USA), 1:500 dilution of the biotinylated rabbit anti-goat-Ig secondary antibody (Dako Corporation) and 1:1000 dilution of streptavidin-conjugatedhorseradish peroxidase (Dako Corporation). Antibody binding was detected with the ECL Western Blotting Analysis System (GE Healthcare Bio-Sciences Pty Ltd, New South Wales, Australia).

RESULTS Complete genotyping data of the LOXL1 SNPs were available for 2508 participants, of whom 86 (3.4%) had been diagnosed with pseudoexfoliation syndrome. The disease status of 335 (13.4%) participants was unclassifiable. The demographic features of our population-based study cohort are presented in Table 1. All genotyped SNPs were in Hardy–Weinberg equilibrium. Seven SNPs were found to be associated with pseudoexfoliation, after age-adjustment and Bonferroni correction (Fig. 1). However, no tagging SNP remained significant after additional adjustment for the two non-synonymous SNPs located in exon 1 of LOXL1 (rs1048661: Arg141Leu and rs3825942: Gly153Asp). The age-adjusted OR for developing pseudoexfoliation and having the G allele at SNP rs1048661 or the G allele at rs3825942 were 1.86 (95%CI: 1.27– 2.76) and 3.81 (95%CI: 1.88– 9.02), respectively (Table 2). The high risk haplotype for these SNPs (G,G) conferred an OR of 4.74 (95%CI: 2.32– 11.26) compared with the lower risk haplotype (G,A). The (T,A) haplotype expected to have the lowest risk at these SNPs was not observed in our population, consistent with the findings of Thorleifsson et al. (16). Two copies of the high risk (G,G) haplotype (the high risk diplotype) were identified in 54.7% of cases compared with 24.6% of our control population. The high risk diplotype conferred an OR of 7.20 (95%CI: 3.04 – 20.75) compared with no copies of the high risk haplotype (Fig. 1). We investigated the distribution of LOXL1 risk diplotypes in glaucoma cases without pseudoexfoliation syndrome, versus controls without glaucoma and found no significant difference (P ¼ 0.66). Pseudoexfoliative glaucoma was strongly associated with the high risk LOXL1 diplotype. However, the magnitude of this association was not greater than that with pseudoexfoliation syndrome alone, possibly reflecting the relatively small number of pseudoexfoliative glaucoma cases in this population-based study (data not shown). The two significant disease-associated non-synonymous coding SNPs in LOXL1 were found to be at a very high frequency in our control population, similar to the findings of

714

Human Molecular Genetics, 2008, Vol. 17, No. 5

Table 2. Prevalence of pseudoexfoliation syndrome and allele frequencies of the disease-associated non-synonymous LOXL1 coding SNPs in the Nordic population compared with our Anglo-Celtic population. Data from combined pseudoexfoliative glaucoma and glaucoma-free pseudoexfoliation syndrome subjects extrapolated from  Thorleifsson et al. (16). OR, odds ratios; CI, confidence interval; f, frequency. x2 comparison between Nordic and Anglo-Celtic allele frequen  P ¼ 0.45 cies:  P ¼ 0.22, P ¼ 0.13, P ¼ 0.0007, Disease prevalence over age 80 years

Nordic (40.0%) f cases

rs1048661 (G) rs3825942 (G)

0.83 0.99

f controls 0.64 0.86

Anglo-Celtic (4.6%) OR (95%CI) 2.53 (2.06– 3.12) 19.56 (8.93– 53.53)

P-value 2.0410219 2.3610223

f cases 0.78   0.95

f controls 

0.66    0.84

OR (95%CI)

P-value

1.86 (1.27– 2.76) 3.81 (1.88– 9.02)

8.491024 7.831025

Figure 2. Cross-species homology of the LOXL1 protein region containing the disease associated non-synonymous variants. The variants R141L and G153D correspond to SNPs rs1048661 and rs3825942, respectively. Note that the wild-type is the disease-associated allele.

Thorleifsson and coworkers (13 – 16), despite the substantially lower prevalence of pseudoexfoliation in our population. The allele frequency of these SNPs in our control population did not significantly differ from the allele frequency reported in the Nordic populations (Table 2). Both disease-associated SNPs were identified at a slightly higher allele frequency in Nordic cases compared with the Australian cases. However, this difference was statistically significant only for rs3825942 (Table 2). To investigate the origins of these disease-associated coding variants we examined their cross-species homology. Both disease-associated variants are well conserved across mammalian species (Fig. 2). It is noteworthy that the common ancestral wild-type allele at each SNP is the disease-associated allele. Ocular expression of LOXL1 RT – PCR analysis revealed LOXL1 expression in all the ocular tissues analysed except the human retina (Fig. 3). Expression, as indicated by the 585 bp amplicon, in the mouse cornea, human iris, ciliary body and lens (the tissues surrounding the aqueous humour containing ocular anterior chamber), is highly consistent with involvement in pseudoexfoliation syndrome. Specificity of the amplicon was confirmed by sequencing. Expression of LOXL1 in SRA 01/04 human lens epithelial cells correlates well with its expression in the postmortem human lens. LOXL1 was also expressed in HEK 293A fibroblast cells. The presence of the LOXL1 protein in ocular tissues of interest was determined by western blotting. Specific protein bands of 130 and 80 kDa were detected in the mouse

Figure 3. RT– PCR analysis of LOXL1 expression in ocular tissues (A) and cultured cell lines (B). Human LOXL1 specific primers were used for cDNA amplification from (A) mouse cornea, human iris, ciliary body, lens, retina and optic nerve, and (B) human lens epithelium cells SRA 01/04 and human embryonic kidney fibroblast cells HEK 293A. PCR products were analysed on a 1.2% agarose gel. RT, reverse transcription; kb, kilobase; bp, basepair; 2ve, negative.

cornea, human iris, ciliary body, lens capsule and optic nerve (Fig. 4). In the human iris and ciliary body, an additional protein band of 42 kDa was observed. The size of the latter band corresponds with one of the mature forms of LOXL1 that has previously been reported in mouse lung and aorta (25). The 130 kDa band present in all the ocular tissues analysed most likely resulted from polymerization of the LOXL1 precursor protein with predicted mass of 63 kDa. However, the precursor protein monomer itself was not detected in these tissues suggesting that the majority of the protein exists in a polymerized form. Similarly, the 80 kDa band likely arose from polymerization of the 42 kDa mature protein. In most of the ocular tissues tested, the 80 kDa protein band was relatively more intense than the 130 kDa band, suggesting a greater proportion of the protein is present in the mature form.

DISCUSSION Pseudoexfoliation syndrome is a relatively uncommon age-related disease characterized by a generalized fibrillar

Human Molecular Genetics, 2008, Vol. 17, No. 5

Figure 4. LOXL1 protein expression in ocular tissues. Protein extracts of mouse cornea, human iris, ciliary body, lens capsule and optic nerve were separated on 12% (top panel) and 10% (bottom panel) polyacrylamide gels and the LOXL1 protein detected with the anti-LOXL1 antibody (arrows). Molecular mass of protein standards are indicated.

degeneration of elastin containing tissues. Pseudoexfoliation causes severe chronic open angle glaucoma with blindness, is associated with cataract formation, increased complications at cataract surgery, and also may be associated with systemic vascular disease (9– 12). In this study we confirmed, using a non-Nordic population, the findings of Thorleifsson et al. (16), that two coding variants in the LOXL1 gene are strongly associated with pseudoexfoliation. The LOXL1 association is of such a magnitude that it is clearly identifiable in our population; despite the fact that a small proportion of the control samples in this study are likely to have pseudoexfoliation undiagnosed due to previous cataract surgery, or will develop it at a later age. We found that the high risk diplotype across the two coding SNPs conferred a high OR (7.20, 95%CI: 3.04– 20.75) for developing pseudoexfoliation. Despite our general population having an approximate 9-fold lower lifetime incidence of pseudoexfoliation than the Nordic population, a similar proportion of people in our unaffected control group were found to carry the high risk diplotype (25%), as compared with those studied by Thorleifsson and coworkers (13 – 16). This is even more remarkable considering that the control population in the Thorleifsson study was not specifically examined for the condition and hence would be expected to contain affected but undiagnosed individuals with pseudoexfoliation. This example of relatively uncommon diseases being caused by common variants is not likely to be unique. It is well appreciated that the frequency of disease-associated alleles at each genetic locus is subject to the joint effects of selection, mutation and random genetic drift (7). We found that the disease-associated variants in LOXL1 were well conserved across species and that, peculiarly, the less common ‘mutations’ that have arisen are protective against pseudoexfoliation. Given the interesting phenotypic observation of uterine prolapse in a LOXL1 knockout mouse (26), we speculate that pseudoexfoliation-associated ancestral variants in LOXL1 could result in lower miscarriage rates. Pseudoexfoliation material is comprised of a cross-linked, highly glycosylated and enzymatically resistant glycoprotein – proteoglycan complex (9). This structure generally bears epitopes of the basement membrane and the elastic fibre systems (9). LOXL1 belongs to a group of proteins responsible for catalysing the oxidative deamination of

715

lysine residues of tropoelastin (25,27,28). In turn, this deamination causes spontaneous cross-linking and formation of elastin polymer fibres (26). Our RT – PCR and western analysis clearly demonstrate the expression of LOXL1 in the anterior segment of the eye, in keeping with the now recognized clinical manifestation of disease-causing variants in this gene. Interestingly, on western blotting, the 32 and 52 kDa mature forms of LOXL1, present in the culture media of human, mouse and bovine cultured cells and in tissues (29), were not observed in ocular tissues. Although LOXL1 protein species of higher than predicted mass are not uncommon in tissues and cultured cells (25,27), their predominance in ocular tissues is intriguing. Pseudoexfoliation syndrome represents a complex, late-onset disease, and our finding of a similar allelic architecture to that described by Thorleifsson et al., in the face of their study population’s higher prevalence for this disease, suggests that other genetic or environmental factors are important in its genesis. Despite the high population attributable risk of LOXL1 alleles in the Nordic study, which is in large part due to the remarkably high population frequency of the ‘risk alleles’, our findings suggest that further research into the molecular underpinnings of pseudoexfoliation is clearly warranted.

WEB RESOURCES The HapMap Project: www.hapmap.org Genetics Power Calculator: http://pngu.mgh.harvard.edu/ purcell/gpc/ Ensembl: www.ensembl.org

Conflict of Interest statement. All authors have no conflict of interest or financial interest in this work.

FUNDING This work was supported by grants from the Ophthalmic Research Institute of Australia, Glaucoma Australia and the National Health and Medical Research Council (NHMRC). A.W.H. is an NHMRC Medical Postgraduate Scholar, K.P.B. an NHMRC Peter Doherty Research Fellow, D.A.M. a Pfizer Research Fellow and J.E.C. is supported in part by an NHMRC Practitioner Fellowship.

REFERENCES 1. Daiger, S.P. (2005) Genetics. Was the Human Genome Project worth the effort? Science, 308, 362–364. 2. International Human Genome Sequencing Consortium (2004) Finishing the euchromatic sequence of the human genome. Nature, 431, 931– 945. 3. The HapMap Consortium (2003) The International HapMap Project. Nature, 426, 789– 796. 4. The Wellcome Trust Case Control Consortium (2007) Genome-wide association study of 14,000 cases of seven common diseases and 3,000 shared controls. Nature, 447, 661– 678. 5. Reich, D.E. and Lander, E.S. (2001) On the allelic spectrum of human disease. Trends Genet., 17, 502–510. 6. Lander, E.S. (1996) The new genomics: global views of biology. Science, 274, 536–539. 7. Pritchard, J.K. (2001) Are rare variants responsible for susceptibility to complex diseases? Am. J. Hum.Genet., 69, 124– 137.

716

Human Molecular Genetics, 2008, Vol. 17, No. 5

8. Pritchard, J.K. and Cox, N.J. (2002) The allelic architecture of human disease genes: common disease-common variant or not? Hum. Mol. Genet., 11, 2417–2423. 9. Schlotzer-Schrehardt, U. and Naumann, G.O. (2006) Ocular and systemic pseudoexfoliation syndrome. Am. J. Ophthalmol., 141, 921–937. 10. Ringvold, A. (2001) Pseudoexfoliation and aortic aneurysms. Lancet, 357, 2139– 2140. 11. Naumann, G.O., Schlotzer-Schrehardt, U. and Kuchle, M. (1998) Pseudoexfoliation syndrome for the comprehensive ophthalmologist. Intraocular and systemic manifestations. Ophthalmology, 105, 951–968. 12. Mitchell, P., Wang, J.J. and Smith, W. (1997) Association of pseudoexfoliation syndrome with increased vascular risk. Am. J. Ophthalmol., 124, 685– 687. 13. Mitchell, P., Wang, J.J. and Hourihan, F. (1999) The relationship between glaucoma and pseudoexfoliation: the Blue Mountains Eye Study. Arch. Ophthalmol., 117, 1319–1324. 14. Ringvold, A. (1996) Epidemiology of glaucoma in northern Europe. Eur. J. Ophthalmol., 6, 26–29. 15. Forsius, H. (1979) Prevalence of pseudoexfoliation of the lens in Finns, Lapps, Icelanders, Eskimos, and Russians. Trans. Ophthalmol. Soc. UK, 99, 296–298. 16. Thorleifsson, G., Magnusson, K.P., Sulem, P., Walters, G.B., Gudbjartsson, D.F., Stefansson, H., Jonsson, T., Jonasdottir, A., Jonasdottir, A., Stefansdottir, G. et al. (2007) Common sequence variants in the LOXL1 gene confer susceptibility to exfoliation glaucoma. Science, 317, 1397–1400. 17. Mitchell, P., Smith, W., Attebo, K. and Healey, P.R. (1996) Prevalence of open-angle glaucoma in Australia. The Blue Mountains Eye Study. Ophthalmology, 103, 1661– 1669. 18. Attebo, K., Mitchell, P. and Smith, W. (1996) Visual acuity and the causes of visual loss in Australia. The Blue Mountains Eye Study. Ophthalmology, 103, 357–364. 19. de Bakker, P.I., Yelensky, R., Pe’er, I., Gabriel, S.B., Daly, M.J. and Altshuler, D. (2005) Efficiency and power in genetic association studies. Nat. Genet., 37, 1217–1223.

20. Barrett, J.C., Fry, B., Maller, J. and Daly, M.J. (2005) Haploview: analysis and visualization of LD and haplotype maps. Bioinformatics, 21, 263– 265. 21. Sole, X., Guino, E., Valls, J., Iniesta, R. and Moreno, V. (2006) SNPStats: a web tool for the analysis of association studies. Bioinformatics, 22, 1928– 1929. 22. Purcell, S., Cherny, S.S. and Sham, P.C. (2003) Genetic power calculator: design of linkage and association genetic mapping studies of complex traits. Bioinformatics, 19, 150. 23. Thompson, J.D., Higgins, D.G. and Gibson, T.J. (1994) CLUSTALW: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Res., 22, 4673–4680. 24. Bradford, M.M. (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem., 72, 248–254. 25. Thomassin, L., Werneck, C.C., Broekelmann, T.J., Gleyzal, C., Hornstra, I.K., Mecham, R.P. and Sommer, P. (2005) The Pro-regions of lysyl oxidase and lysyl oxidase-like 1 are required for deposition onto elastic fibers. J. Biol. Chem., 280, 42848–42855. 26. Liu, X., Zhao, Y., Gao, J., Pawlyk, B., Starcher, B., Spencer, J.A., Yanagisawa, H., Zuo, J. and Li, T. (2004) Elastic fiber homeostasis requires lysyl oxidase-like 1 protein. Nat. Genet., 36, 178–182. 27. Borel, A., Eichenberger, D., Farjanel, J., Kessler, E., Gleyzal, C., Hulmes, D.J., Sommer, P. and Font, B. (2001) Lysyl oxidase-like protein from bovine aorta. Isolation and maturation to an active form by bone morphogenetic protein-1. J. Biol. Chem., 276, 48944–48949. 28. Noblesse, E., Cenizo, V., Bouez, C., Borel, A., Gleyzal, C., Peyrol, S., Jacob, M.P., Sommer, P. and Damour, O. (2004) Lysyl oxidase-like and lysyl oxidase are present in the dermis and epidermis of a skin equivalent and in human skin and are associated to elastic fibers. J. Invest. Dermatol., 122, 621–630. 29. Maki, J.M. and Kivirikko, K.I. (2001) Cloning and characterization of a fourth human lysyl oxidase isoenzyme. Biochem. J., 355, 381– 387.