Searching for Psoriasis Susceptibility Genes in Italy: Genome Scan and Evidence for a New Locus on Chromosome 1

Searching for Psoriasis Susceptibility Genes in Italy: Genome Scan and Evidence for a New Locus on Chromosome 1 Francesca Capon,* Giuseppe Novelli,* Sabrina Semprini,* Maurizio Clementi,† Maurizio Nudo,‡ Pietro Vultaggio,‡ Cinzia Mazzanti,§ Tommaso Gobello,§ Annalisa Botta,* Giuseppe Fabrizi,‡ and Bruno Dallapiccola*¶ *Chair of Human and Medical Genetics, Department of Biopathology, ‘‘Tor Verngata’’ University of Rome, Italy; †Department of Paediatrics, Medical Geneitcs Unit, University of Padova, Italy; ‡Department of Dermatology, Catholic University of Rome, Italy; §VIII Dermatology Division, Istituto Dermatopatico dell’Immacolata, IRCCS, Rome, Italy; ¶CSS Hospital, IRCCS San Giovanni Rotondo, Italy

Psoriasis is a chronic inflammatory dermatitis, affecting µ2% of the population. Major clinical features include red, scaly patches on scalp, elbows, and knees, with or without severe arthritis. Several putative susceptibility loci have been mapped by parametric and non-parametric linkage analysis to chromosome regions 2p, 4q, 6p, 8q, 16q, 17q, and 20p; however, the most significant results and confirmation of linkage are only available for the 17q and 6p chromosome regions at present. In this study, 22 multiplex Italian families were investigated for linkage to 6p and 17q susceptibility regions, using a set of four microsatellites. These analyses failed to detect significant linkage with any of the examined markers.

A genome-wide scan was then performed on one of the largest pedigrees, searching for an additional susceptibility locus. This study disclosed a putative linkage to chromosome 1cen-q21 markers. When these microsatellites were analyzed in the remaining families of the sample, a significant linkage was observed using both parametric and non-parametric methods. The highest two-point lod score value was obtained with D1S305 marker (3.75 at θ J 0.05). Non-parametric analysis at this locus also demonstrated a significant excess of allele sharing (p J 0.0001). Key words: genome scanning/linkage/ multigenic disorder/psoriasin. J Invest Dermatol 112:32–35, 1999

P

reported (Heneseler and Christophers, 1985; Jenisch et al, 1998). Several putative loci for familial PS susceptibility have been assigned to chromosomes 2p, 4q, 6p, 8q, 16q, 17q, and 20p, using parametric and non-parametric linkage analysis (Tomofohrde et al, 1994; Matthews et al, 1996; Nair et al, 1997; Trembath et al, 1997). The most significant results and confirmation of linkage, however, are available at present only for the 17q and 6p chromosome regions (Trembath et al, 1997; Nair et al, 1997; Jenisch et al, 1998). In this study, we investigated 22 multiplex Italian families with a set of microsatellite markers mapping to chromosome regions 6p and 17q. As this analysis failed to detect any significant linkage, a large pedigree was selected for a genome-wide scan. This provided evidence for a novel susceptibility locus mapping to chromosome 1cen-q21.

soriasis (PS; MIM* 177900) is a chronic inflammatory disorder affecting both sexes with a population prevalence of about 2% (Nevitt and Hutchinson, 1996). The disease is characterized by sharply marginated red, scaly plaques. About 5% of PS patients will develop arthritis, which is frequently mild but can be severe and mutilating (Christophers and Sterry, 1993). The pathogenesis of this disorder is as yet unknown; however, abnormal keratinocyte proliferation and infiltration of inflammatory elements into the dermis and the epidermis has been documented (Granstein, 1996). Despite the determining influence of several environmental factors, familial clustering of PS is well established and empiric recurrence risk in first-degree relatives of isolated patients ranges from 8% to 23% (Lomholt, 1963; Hellgren, 1967). Based on twin studies, PS heritability has been estimated in the range of 70%–90% (Farber et al, 1974). Inheritance of PS is unclear, but the transmission observed in large pedigrees suggests dominance with a reduced penetrance and variable expressivity (Kimberling and Dobson, 1973). On the other hand, the same segregation patterns could be observed in the presence of a polygenic disorder characterised by high disease allele frequencies (Elder et al, 1994). In fact, association with several HLA antigens, including Cw6 and B57, has been

MATERIALS AND METHODS Criteria for entry in the study and pedigree selection Twenty-two families (Fig 1) were selected by a search of the medical records among those referred to the Department of Dermatology of the Catholic University and to the Istituto Dermatopatico dell’Immacolata of Rome, Italy. Families were ascertained via an affected offspring and included in the study if they had at least another first or second degree affected relative. Consensus diagnosis of PS was assessed by two expert dermatologists who carried out examinations of skin, nails, and joints of all psoriatic subjects and their relatives. Individuals were considered affected if they had plaque PS. Involvement at the seborrhoeic sites was not considered a phenotypic indicator of PS and individuals displaying seborrhoeic eczema without PS plaques were considered as unaffected; however, in the cohort of studied families, no individuals showed signs of seborrhoeic dermatitis or steatoid PS in the classical seborrhoeic skin areas (scalp, presternal area, interscapular region, and face). Families comprised a total of 108 affected and 116

Manuscript received January 5, 1998; revised October 12, 1998; accepted for publication October 13, 1998. Reprint requests to: Professor Giuseppe Novelli, Cattedra di Genetica Umana, Dipartimento di Biopatologia e Diagnostica per Immagini, Universita` di Roma ‘‘Tor Vergata,’’ Via di Tor Vergata 135, 00133 Roma, Italy. Abbreviations: NPL, nonparametric linkage; PS, psoriasis.

·

0022-202X/99/$10.50 Copyright © 1999 by The Society for Investigative Dermatology, Inc.

32

VOL. 112, NO. 1 JANUARY 1999

A NEW PSORIASIS LOCUS TO 1cen-q21

33

Figure 1. Psoriasis pedigrees participating in this study. Solid symbols, affected individuals; open symbols, unaffected.

unaffected (Fig 1). Because of age-dependent expressivity, unaffected individuals under 30 could not be classified, and were not included in this study. Genotyping Blood was taken with informed consent of each subject. DNA was obtained from peripheral blood lymphocytes by standard phenolchloroform extraction. Microsatellite markers were typed by electrophoresis of PCR products onto 6% denaturing polyacrilamide gels, and subsequent autoradiography (Weber and May, 1989). Markers D6S273, D17S784, D17S928, and D17S802 were selected on the basis of their demonstrated high lod score value in previous PS studies (Tomofohrde et al, 1994; Trembath et al, 1997), and were used to test for linkage with susceptibility loci mapping on chromosomes 6p and 17q. Markers D2S134, D4S408, D4S1535 D8S284, D16S3039, and D20S851 were used to test family PS31 for linkage to 2q, 4q, 8q, 16q, and 20p chromosome regions. The PCRbased genome-wide search was performed using Ge´ne´thon microsatellite markers selected at 10 cM intervals (Dib et al, 1996). Marker distances for some intervals were integrated into the map using information available from Genome DataBase (GDB:http://www.gdb.org/). Linkage analysis and homogeneity tests For initial data analysis, PS was considered an autosomal dominant trait, segregating with reduced penetrance (Tomofohrde et al, 1994; Matthews et al, 1996). The following genetic map and distances were used for linkage calculations: cen-D1S514(0.04)-D1S498-(0.025)-D1S1664-(0.015)-D1S305-tel (Dib et al, 1996; Marenholz et al, 1996). Haplotypes were constructed by minimising crossing-over events; marker data causing apparent high-order recombinations were re-inspected and/or repeated. The MLINK and LINKMAP programs (Lathrop and Laouel, 1984) were used to calculate two-point and multipoint lod scores. A penetrance value of 70% was assumed, and a

corresponding gene frequency of 0.014 was derived by Hardy–Weinberg equilibrium. Non-parametric linkage analysis implemented with GENEHUNTER (Kruglyak et al, 1996) was used to assess multipoint lod score at each marker, at four points within each interval between markers, and in the flanking regions upstream D1S514 and downstream D1S305 (setting skip large off, NPLall, score both, increment step 5). Evidence for linkage was arbitrarily defined at the level of p ø 0.01. Results obtained by these methods were expressed as lod scores and normal deviate units. The HOMOG program (HOMOG package, 1996) was used to analyze heterogeneity with respect to a single locus under the hypothesis of two family types, one with linkage between a trait to a marker and the other without linkage. Linkage disequilibrium analysis An approach was adopted in which one affected individual per pedigree was ascertained and the allele transmitted by the affected parent was counted as case. Control alleles were those not transmitted by affected parents. Further controls were obtained by analysing 25 unrelated unaffected individuals. Significance levels were computed using CLUMP program, which works on 2 3 N contingency tables and is designed to be especially useful when N is large and the table is sparse. A Monte Carlo approach is used to perform repeated simulations generating tables with the same marginal totals as the one under construction. The significance is then calculated by counting the number of times that a chi-square associated with the real table is achieved by the randomly simulated data (Sham and Curtis, 1995)

RESULTS None of our families showed significant linkage with either 6p or 17q susceptibility regions. Therefore, we selected a large three-

34

CAPON ET AL

THE JOURNAL OF INVESTIGATIVE DERMATOLOGY

Figure 2. Chromosome 1q21 haplotypes for family PS31. Solid bar, at risk haplotype; solid symbols, affected individuals; open symbols, unaffected.

generation pedigree (PS31) for a genome-wide search. As a matter of fact, our sample includes two families that are more informative (PS48 and PS49), but none of them was available at the time the scan was performed. In a first stage, putative susceptibility regions on chromosomes 2p, 4q, 8q, 16q, and 20p were tested on PS31 and excluded. Then, a preliminary analysis of 198 microsatellite markers spanning the 22 autosomes was undertaken. This study revealed a putative linkage with D1S498 marker on 1q21 (Zmax 5 1.69 at θ 5 0.00). This was confirmed by typing three neighboring microsatellites (D1S514, D1S1664, D1S305) in both PS31 (Fig 2) and the other families of the sample. A maximum two-point lod score of 3.75 was observed with D1S305 marker at θ 5 0.05, whereas Zmax (at θ 5 0.1) for D1S514, D1S498, and D1S1664 were 3.07, 3.36, and 2.13, respectively. Non-parametric analysis generated significant p values and nonparametric linkage (NPL) scores throughout the linked region on chromosome 1cen-q21. Multipoint lod score values were computed under heterogeneity, which allows alpha to vary until the HLOD is maximised. This analysis generated a multipoint lod score curve peaking between D1S1664 and D1S305, at α 5 0.57 (Table I). This is consistent with the observation of two putative recombination events observed in families PS26 and PS38. In either case, the occurrence of these cross-overs could not be unambiguously demonstrated because the haplotypes of recombining subjects had to be derived, their parents being deceased (data not shown, available on request). The sample genetic heterogeneity was also tested using the HOMOG program to analyze LINKMAP multipoint lod score data. This generated an alpha value of 0.55 (H2 versus H1, χ2 5 7.98, 1 df, p 5 0.005), well correlating with GENEHUNTER results. Interestingly, a closer inspection of our pedigrees revealed that PS22, P23, PS25, and PS30 (all displaying negative lod scores and posterior probabilities of linkage , 0.4) did not originate from the Lazio region (central Italy), as did the other families of the sample. In order to test the hypothesis of a founder effect, disequilibrium studies were performed in the 1q-linked set of families (i.e., pedigrees displaying positive lod scores and posterior probabilities of linkage . 0.6). Alleles of 15 non-consanguineous patients and 25 unrelated unaffected controls were analyzed at D1S514, D1S498, D1S1664, and D1S305 loci; however, chi-square analysis on 2 3 N tables failed to detect any significant distortion in allele transmission (data not

Table I. Output of GENEHUNTER analysis on 1cen-q21 markers Position cM –2.00 –1.00 D1S514 0.80 1.60 2.40 3.20 D1S498 4.30 4.60 4.90 5.20 D1S1664 6.00 6.50 7.00 7.50 D1S305 9.00 10.00

Multipoint lod scorea

NPL score

1.5 1.2 0.59 1.70 1.91 1.81 1.34 –0.66 0.26 0.71 1.02 1.27 1.47 1.66 1.82 1.95 2.06 2.15 2.49 2.77

3.53 3.72 3.92 3.93 3.93 3.92 3.92 3.88 3.91 3.94 3.98 4.01 4.04 4.04 4.05 4.05 4.06 4.07 3.86 3.66

p value HLOD

alpha

0.0005 0.0003 0.0002 0.0002 0.0002 0.0002 0.0002 0.0002 0.0002 0.0002 0.0002 0.0001 0.0001 0.0001 0.0001 0.0001 0.0001 0.0001 0.0002 0.0004

0.58 0.57 0.55 0.56 0.56 0.56 0.54 0.50 0.53 0.54 0.55 0.56 0.56 0.57 0.57 0.57 0.57 0.58 0.59 0.61

5.30 5.30 5.38 5.46 5.52 5.54 5.48 5.34 5.50 5.63 5.73 5.81 5.88 5.91 5.94 5.97 5.99 6.00 5.91 5.81

aBecause of memory constraints, most unaffected were eliminated from extended pedigrees (skip large off). This accounts for the discrepancy between lod score and NPL, as the missing individuals would add very little to the NPL statistic (which measures sharing among affected) but affect somewhat the lod score.

shown). Only the D1S498 marker yielded a borderline chi-square of 15.1 that was reached 109 times in 2000 simulations (p 5 0.054). DISCUSSION In this study, we report PS linkage analysis in a cohort of 22 Italian families. Using the lod score method, we failed to detect linkage to 6p and 17q chromosome regions. Therefore, we performed a genome scan on one of the largest pedigrees, after excluding that this family may be assigned to any other putative PS locus (i.e., 2q, 4q, 8q, 16q, and 20p chromosome regions). Our results suggested the existence of a novel susceptibility locus on chromo-

VOL. 112, NO. 1 JANUARY 1999

some 1cen-q21. When analysis of linked markers was extended to the whole sample, significant lod scores were obtained with D1S514, D1S498, D1S1664, and D1S305. An excess of allele sharing was also observed at these loci, using a non-parametric method. Haplotype analysis suggested incomplete penetrance in families PS17, PS24, PS26, PS28, PS38, PS45, PS48, and PS49, where unaffected individuals had inherited and/or transmitted predisposing alleles. This can also be due to the difficulty of diagnosing the mild PS forms, intrafamilial variability of age of onset, and occurrence of spontaneous remissions (Lomholt, 1963). The high prevalence of PS is an additional factor that complicates linkage analysis. This is well illustrated by families PS17 and PS31, in which marriages occurred between two subjects with a positive family history for the disease. In both cases, linkage analysis disclosed the different haplotypes segregating in the two branches of each family (Fig 2). Because four 1q-unlinked families did not originate from the Lazio region, we undertook a disequilibrium study to search for a founder effect in Lazio. These analyses, however, failed to detect any significant distortion in allele transmission for the examined markers. Only D1S498 generated a chi-squared value that is on the threshold of statistical significance (p 5 0.054). This may be due to the small size of the sample (15 affected patients from 1q-linked families) and to the large inter-marker distances (1.5–3 cM). The putative PS candidate region on chromosome 1cen-q21 is of interest because it contains the so-called Epidermal Differentiation Complex, which is a cluster of at least 20 genes expressed during epithelial differentiation (Volz et al, 1993; Borglum et al, 1995; Hardas et al, 1996; Mischke et al, 1996). Many of these genes belong to the small proline-rich protein (SPRR) and S100 gene families (Backendorf and Hohl, 1992). The SPRR genes are expressed at high levels during differentiation of epidermal keratinocytes (Gibbs et al, 1993). The S100 gene products play an important role in exerting the effects of calcium on cell growth and differentiation (Heizmann and Hunzinker, 1991). Additional skin-expressed genes also map to the 1q21 region, including loricrin, involucrin, filaggrin, trichohyalin, cellular retinoic acid-binding protein II (CRABP II), dermatopontin, and β-glucocerebrosidase (Kingsmore et al, 1990; Backendorf and Hohl, 1992; Elder et al, 1992a, b; Superti-Furga et al, 1993). Five members of this gene cluster – psoriasin (S100A7), calgranulin A (MRP8), calgranulin B (MRP14), SPRR2–1, and CaN19 – are markedly overexpressed in psoriatic lesions (Madsen et al, 1991; Hardas et al, 1996). Interestingly, psoriasin acts as a potent and selective chemotactic inflammatory protein for CD41 T lymphocytes and neutrophils and could therefore be involved in the inflammatory response observed during PS onset (Jinquan et al, 1996). Experiments are currently underway to study the expression status and the nucleotide sequence of these genes in affected members of our 1q-linked PS families. These studies are expected to elucidate the role of psoriasin and the other tightly linked 1q21 genes in the PS pathogenesis.

We thank A.DI.PSO (Italian Association for the Defence of Psoriatic Patients) and the International Center for Study and Research in Dermatology (Rome) for help and co-operation in this study. We are indebted to anonymous reviewers for constructing criticism of the manuscript. We are grateful to the UK HGMP Resource Center for computer assistance. This work was supported by grants from Italian Telethon E.557 (to G.N.), and Italian Ministry of Health (to B.D.). All the experiments described comply with the current Italian law.

REFERENCES Backendorf C, Hohl D: A common origin for cornified envelope proteins? Nature Genet 2:91, 1992 Borglum AD, Flint T, Madsen P, Celis JE, Kruse TA: Refined mapping of the psoriasin gene S100A7 to chromosome 1cen-q21. Hum Genet 96:592–596, 1995

A NEW PSORIASIS LOCUS TO 1cen-q21

35

Christophers E, Sterry W: Psoriasis. In: Fitzpatrick TB, Eisen AZ, Wolff K, Austin KF (eds). Dermatology in General Medicine. New York: McGraw-Hill, 1993, pp. 489–514 Dib C, Faure` S, Fizames C, et al: A comprehensive genetic map of the human genome based on 5264 microsatellites. Nature 380:152–154, 1996 Elder JT, Astrom A, Pettersen U, Voorhees JJ, Trent JM: Assignment of the human CRABP-II gene to chromosome 1q21 by nonisotopic in situ hybridization. Hum Genet 89:487–490, 1992a Elder JT, Astrom A, Pettersen U, Voorhees JJ, Trent JM: Assignment of the human CRABP-II gene to chromosome 1q21 by non-isotopic in situ hybridization. Hum Genet 89:487–490, 1992b Elder JT, Nair RP, Guo SW, Henseler T, Christophers E, Voorhees JJ: The genetics of psoriasis. Arch Dermatol 130:216–224, 1994 Farber EM, Nall ML, Watson W: Natural history of psoriasis in 61 twin pairs. Arch Derm 109:207–211, 1974 Gibbs S, Fijneman R, Wiegant J, van Kessel AG, van De Putte P, Backendorf C: Molecular characterization and evolution of the SPRR family of keratinocyte differentiation markers encoding small proline-rich proteins. Genomics 16:630– 637, 1993 Granstein RD: Psoriasis: further evidence for a key role for leucocytes J Clin Invest 98:1695–1696, 1996 Hardas DB, Zhao X, Zhang J, Longquin X, Stoll S, Elder JT: Assignment of psoriasin to human chromosomal band 1q21: coordinate overexpression of clustered genes in psoriasis. J Invest Dermatol 106:753–758, 1996 Heizmann CW, Hunzinker W: Intracellular calcium-binding proteins: more sites than insights. Trends Biochem Sci 16:98–103, 1991 Hellgren I: Psoriasis, The Prevalence in Sex, Age, and Occupational Groups in Total Population in Sweden: Morphology, Inheritance and Association with Other Skin and Rheumatic Diseases. Stockolm: Almqwist and Wiksell, 1967 Heneseler T, Christophers E: Psoriasis of early and late onset: characterisation of two types of psoriasis vulgaris. J Am Acad Dermatol 13:450–456, 1985 Jenisch S, Henseler T, Nair RP, et al: Linkage analysis of Human Leukocyte Antigen (HLA) markers in familial psoriasis: strong disequilibrium effects provide evidence for a major determinant in the HLA-B/-C region. Am J Hum Genet 63:191–199, 1998 Jinquan T, Vorum H, Larsen CG, Honore´ B, Celis JE, Thestrup-Pedersen K: Psoriasin: a novel chemiotactic protein. J Invest Dermatol 107:5–10, 1996 Kimberling WJ, Dobson RL: The inheritance of psoriasis. J Invest Derm 60:538– 540, 1973 Kingsmore SF, Moseley WS, Watson ML, Sabina RL, Holmes EW, Seldin MF: Long-range restriction site mapping of a syntenic segment conserved between human chromosome 1 and mouse chromosome 3. Genomics 7:75–83, 1990 Kruglyak L, Daly MJ, Reeve-Daly MP, Lander ES: Parametric and non-parametric linkage analysis: a unified multipoint approach. Am J Hum Genet 58:1347– 1363, 1996 Lathrop G, Laouel J: Easy calculation of lod scores on small computers. Am J Hum Genet 36:460–465, 1984 Lomholt G: Psoriasis: Prevalence, Spontaneous Course and Genetics. Copenaghen: GEC Gad, 1963 Madsen P, Rasmussen HH, Leffers H, et al: Molecular cloning, occurence and expression of a novel partially secreted protein ‘‘psoriasin’’ that is highly upregulated in psoriatic skin. J Invest Dermatol 97:701–712, 1991 Marenholz I, Volz A, Ziegler A, Davies A, Ragoussis I, Korge BP, Mischke D: Genetic analysis of the Epidermel Differentiation Complex (EDC) on human chromosome 1q21: chromosomal orientation, new markers and a 6 Mb YAC contig. Genomics 37:295–302, 1996 Matthews D, Fry L, Powles A, et al: Evidence that a locus for familial psoriasis maps to chromosome 4q. Nature Genet 14:231–233, 1996 Mischke D, Korge BP, Marenholz I, Volz A, Ziegler A: Genes encoding structural proteins of epidermal cornification and S-100 calcium binding proteins form a gene complex (‘‘Epidermal Differentiation Complex’’) on human chromosome 1q21. J Invest Dermatol 106:989–992, 1996 Nair RP, Henseler T, Jenisch S, et al: Evidence for two psoriasis susceptibility loci (HLA and 17q) and two novel candidate regions (16q and 20p) by genomewide scan. Hum Mol Genet 6:1349–1356, 1997 Nevitt GJ, Hutchinson PE: Psoriasis in the community: prevalence, severity and patients’ beliefs and attitudes towards the disease. Br J Dermatol 135:533– 537, 1996 Sham PC, Curtis D: Monte Carlo tests for associations between disease and alleles at hihly polymorphic loci. Ann Hum Genet 59:97–105, 1995 Superti-Furga A, Rocchi M, Schafer BW, Gitzelman R: Complementary DNA sequence and chromosomal mapping of a human proteoglycan-binding cell adhesion protein (dermatopontin). Genomics 17:463–467, 1993 Tomofohrde J, Silverman A, Barnes R, et al: Gene for familial psoriasis susceptibility mapped to the distal end of chromosome 17q. Science 264:1141–1145, 1994 Trembath RC, Clough RL, Rosbotham JL, et al: Identification of a major susceptibility locus on chromosome 6p and evidence for further disease loci revealed by a two stage genome-wide search in psoriasis. Hum Mol Genet 6:813–820, 1997 Volz A, Korge BP, Compton JG, Ziegler A, Steinert PM, Mischke D: Physical mapping of a functional cluster of epidermal differentiation genes on chromosome 1q21. Genomics 18:92–99, 1993 Weber JL, May PE: Abundant class of human DNA polymorphisms which can be typed using the polymerase chain reaction. Am J Hum Genet 44:388–396, 1989