The genetics of psoriasis: a complex disorder of the skin and immune system

doi:10.1093/hmg/7.10.1537

This Article

	Abstract
	FREE Full Text (PDF)
	Alert me when this article is cited
	Alert me if a correction is posted

Services

	Email this article to a friend
	Similar articles in this journal
	Similar articles in ISI Web of Science
	Similar articles in PubMed
	Alert me to new issues of the journal
	Add to My Personal Archive
	Download to citation manager
	Search for citing articles in: ISI Web of Science (111)
	Request Permissions

Google Scholar

	Articles by Bhalerao, J.
	Articles by Bowcock, A. M.
	Search for Related Content

PubMed

	PubMed Citation
	Articles by Bhalerao, J.
	Articles by Bowcock, A. M.

Social Bookmarking

	What's this?

Human Molecular Genetics Pages 1537-1545

The genetics of psoriasis: a complex disorder of the skin and immune system
Introduction
Twin Studies
Pathogenesis
Association With HLA
Location Of Primary Defect
   Skin
   Psoriasis as an autoimmune disease
Genetic Studies
Linked Regions Reported To Date
   17q25
   4q
   6p
   Other possible disease loci
   Variation in disease phenotypes observed in pedigrees
Animal Models
   The flaky skin mouse
   The integrin B1 transgenic mouse
   The HLA-B27 rat
Conclusion
Acknowledgements
References
Corrigenda

The genetics of psoriasis: a complex disorder of the skin and immune system

Jayant Bhalerao and Anne M. Bowcock*

Department of Internal Medicine, Pediatrics and McDermott Center for Human Growth and Development, University of Texas Southwestern Medical Center, 5323 Harry Hines Boulevard, Dallas, TX 75235-8591, USA

Received April 30, 1998

In the last few years, molecular genetics analyses have permitted novel insights into psoriasis, a disease characterized by uncontrolled proliferation of keratinocytes and recruitment of T cells into the skin. The disease affects ~1-2% of the Caucasian population and can occur in association with other inflammatory diseases such as Crohn's disease and in association with human immunodeficiency virus (HIV) infection. Given that psoriasis has characteristics of an autoimmune disease, it is not surprising that HLA studies revealed an association with certain alleles, notably HLA-Cw6. Despite this HLA component, psoriasis in some families is inherited as an autosomal dominant trait with high penetrance. Loci at chromosome 17q25 and 4q have been identified following genome-wide linkage scans of large, multiply affected families. In the case of at least the susceptibility locus at 17q25, the development of psoriasis does not require the presence of HLA-Cw6. Sib-pair analyses have confirmed the association with HLA-Cw6, confirmed the existence of a locus at 17q25 and identified other possible susceptibility loci. Two independent groups have reported a third region on chromosome 20p. Despite these findings, the extent of genetic heterogeneity and the role of environmental triggers and modifier genes is still not clear. The precise role of HLA also still needs to be defined. The isolation of novel susceptibility genes will provide insights into the precise biochemical pathways that control this disease. Such pathways will also reveal additional candidate genes that can be tested for molecular alterations resulting in disease susceptibility.

INTRODUCTION

Although the skin disease psoriasis was first recognized as a distinct disease as early as 1808 (1), its pathogenic mechanisms have eluded investigators for decades. Recently, it has attracted the attention of molecular geneticists having fallen under the somewhat elusive descriptor `complex disease'. The development of localized lesions in response to skin trauma (Koebner reaction or isomorphic response) was first described by Koebner in 1872 (2) and is a hallmark of psoriasis, a disease that derives its name from the Greek `psora' meaning `to itch' (3). Psoriasis commonly begins between ages 20 and 40, with an average age of onset of 27. It is estimated that up to 80% of juvenile guttate psoriasis cases are preceded by streptococcal infections (4). Drugs, stress and climate can also be predisposing factors in susceptible individuals (5).

Prevalence varies with race and geography, being highest in the northernmost regions of the Soviet Union and Norway (5-10% of individuals), of moderate prevalence in the UK, USA and Holland (2-3%) and of low prevalence among North American Indians, Latin American Indians, Mongoloids and Western Africans (0-0.3%) (6,7).

Lesions generally occur on extensor surfaces with or without scalp involvement. However, they may occur on the flexor surfaces, the genitals and the hands and feet. Although plaque-forming psoriasis is most common, and affects 90% of patients, other forms include guttate, erythrodermic and palmar-plantar psoriasis (reviewed in ref. 8). Five to ten percent of patients develop a co-existing, and sometimes debilitating, psoriatic arthritis. One confounding aspect of psoriasis is its waxing and waning; its recurrence and regression, and the variable extent of body involvement. Psoriasis can occur in association with other inflammatory disease such as inflammatory bowel disease (Crohn's disease) and in association with human immunodeficiency virus (HIV) infection (9,10).

Recent advances in semi-automated genonotyping and a variety of methods of genetic analysis are starting to yield insights into complex human diseases such as diabetes (11), asthma (12,13), rheumatoid arthritis (14), Crohn's disease (15,16) and psoriasis (17-20). This article reviews the biological alterations in psoriasis and summarizes the latest genetic findings. For psoriasis, no predisposing gene has yet been identified, although susceptibility frequently appears to have a major genetic component.

TWIN STUDIES

Danish twin registry studies reveal that monozygotic twins have a concordance of psoriasis of 72% compared with 15% in dizygotic twins (21). This corresponds to a heritability of 90-100% given the 2-3% prevalence in the Danish population (21-24). Similarly, Farber et al. observed concordance rates of 70% in monozygotic twins and 23% in dizygotic twins in the USA (25). However, in an Australian study, the concordance rate was only 35% in monozygotic twins and 12% in dizygotic twins (26). Interestingly, when monozygotic twins are concordant for the disease, it tends to be similar in age of onset, body distribution, severity and course; an observation not made in dizygotic twin pairs (25). This would suggest that genetic factors play a role in these variables. Also evident from twin studies is that although genetic factors play a significant role in the pathogenesis of psoriasis, the actual expression of disease is under environmental influence, since concordance never reaches 100% in any given population.

PATHOGENESIS

Although clinical features and severity vary between individuals and with time, psoriasis is characterized by four abnormalities (27). (i) Vascular changes where the papillary blood vessels become dilated and tortuous. This results in redness or erythema, one hallmark of psoriasis. (ii) Inflammation, where polymorphonuclear leukocytes from the dermal vessels enter the epidermis. Lesions are also rich in activated CD4⁺ and CD8⁺ T cells that release proinflammatory cytokines. (iii) Hyperproliferation of the keratinocytic layer (acanthosis). (iv) Altered epidermal differentiation where keratinocytes retain their nuclei in the cornified layer (parakeratosis) and the granular layer is lost. These changes in the epidermis result in scaling, another hallmark of psoriasis.

ASSOCIATION WITH HLA

The autoimmune nature of psoriasis prompted investigations in the early 1970s into associations with the HLA complex on chromosome 6p. Russell et al. (28) first reported association with HLA-B13. Subsequently, strong associations with Cw6 and DR7 were identified in the Finnish population and in Germany (29-31), and were estimated to confer relative risk factors of 4-15 for the disease. Henseler and Christophers (30,32) proposed that there are two types of psoriasis: a familial, early age of onset (<40 years) form, frequently associated with HLA-Cw6, DR7, B13 and B57; and a non-familial, later age at onset form that is associated with HLA-Cw2 and B27. However, in our initial examination of eight multiply affected families collected as part of the National Psoriasis Tissue Bank and described elsewhere (17), the average age of onset in all families was <40 years, and only two of these families showed any association with HLA-Cw6.

Only ~10% of HLA-Cw6-positive individuals develop psoriasis, suggesting a major role for additional genes and/or environmental triggers. This conjecture is supported by studies such as those of Leder and Farber (33) who report that the incidence of psoriasis is lower among non-Bantu speaking western Africans and their African-American descendents, than the Bantu-speaking easterners, despite the fact that predisposing alleles such as HLA-Cw6 and B17 are relatively common in the West African populations. Moreover, we and others have provided evidence for other susceptibility genes. The observation that a large, multiply affected family demonstrated linkage of psoriasis susceptibility to 17q25 (17) and not to HLA suggests that other genes can confer susceptibility. In this family, the average age at onset of the disease was 19 years, although several members had developed the disease in infancy. In a study of 23 multiply affected families collected as part of the National Psoriasis Tissue Bank, we have observed that 25% are HLA-Cw6 positive, so that all affected members harbor at least one Cw6 allele. In one family, all three affected members are HLA-B27 (34).

The contribution of HLA to familial clustering can also be estimated by measuring the expected proportion of affected sib-pairs that share zero haplotypes identical by descent (IBD) divided by the observed proportion [[lambda]_s= [phis]_s/P_s (35)]. In a set of 74 sib-pairs, we recently ascertained [lambda]_s for HLA-C to be 1.4 (unpublished data). The same value was obtained by Trembath et al. (19) who examined 106 affected sibling pairs. The overall [lambda]_s value for psoriasis [risk to siblings divided by prevalence in the general population (35)] is estimated to be between 4 (8/2%) in a Swedish population (36) to 11.5 (23/2%) in the USA (25). If HLA-C accounts for 1.4 of these values, the remainder must be due to other genetic contributions. It is also still not clear if it is the HLA-Cw6 allele per se that predisposes to the disease or an allele in strong linkage disequilibrium with it. In fact, in the recent study of Trembath et al. (19), there was a stronger association with TNF-[alpha] and D6S273 than with HLA-C.

LOCATION OF PRIMARY DEFECT

Although considerable progress has been made in understanding the phenotype of psoriasis at the cellular and molecular levels, many questions remain unresolved. These include the following: what molecular lesions are associated with predisposition to psoriasis?; how frequently are these lesions germline versus sporadic in origin?; what role does the environment play?

Knowledge of genetic defects predisposing to this disease should provide an answer to the important question regarding the origin of the disease; namely, is it a primary defect of the skin (keratinocytes and fibroblasts), cells involved in the immune response [antigen-presenting cells (APCs ) or T cells] or the vascular system (endothelial cells)? Some current concepts pertaining to these are discussed below.

Skin

Krueger et al. recently proposed a `framework hypothesis' stating that `there is an aberration throughout the skin of patients with psoriasis that is modified to disease expression by circulating factors'. They propose that disease is caused by failure to efficiently produce an unidentified skin-specific immunosuppressor molecule. Activity of this suppressor would be affected by certain HLA antigens and its gene would be highly mutable to explain the high incidence of sporadic cases. They propose that the primary skin defect could reside in the epidermis (keratinocyte) or dermis (fibroblast) and that the altered gene could be a regulator of cytokines or growth factors (37). Several studies have demonstrated that both involved and uninvolved skin from psoriatic individuals differ from the skin of normal subjects in several ways. (i) Keratinocyte stem cells ([beta]1 integrin⁺ K1/K10^-) in the uninvolved skin of psoriatic subjects are inherently hyperresponsive to proliferation signals mediated by T cell lymphokines, for example those induced by a combination of granulocyte-macrophage colony-stimulating factor (GM-CSF), interleukin-3 (IL-3) and interferon-[gamma] (IFN-[gamma]) (38). (ii) Keratinocytes from both involved and uninvolved psoriatic skin are 100 times less sensitive to growth inhibition by 1,25-dihydroxyvitamin D3 than normal keratinocytes (39). (iii) Psoriatic keratinocytes show abnormal patterns of integrin [alpha]2[beta]1, [alpha]3[beta]1, [alpha]5[beta]1 and [alpha]6[beta]4 expression (40,41). Aberrant integrin expression by keratinocytes can have dramatic results as is seen in transgenic mice for integrin [beta]1 (see Animal Models). (iv) Overexpression of platelet-derived growth factor (PDGF), transforming growth factor-[alpha] (TGF-[alpha]), IL-6, IL-8, IFN-[gamma] and monocyte chemoattractant protein-1 (MCP-1) are seen in lesional skin (37) and PDGF is overexpressed in uninvolved skin (42,43). PDGF induces IL-6 and MCP-1 whereas TGF-[alpha] induces IL-8 and MCP-1 (37). TGF-[alpha] and IL-8 have been shown to stimulate keratinocyte proliferation (44,45). (v) Fibroblasts from psoriatic skin but not normal skin can induce hyperproliferation of normal keratinocytes (46,47). Fibroblasts from uninvolved psoriatic skin proliferate faster than normal fibroblasts in the presence of normal human serum and proliferate to an even greater extent in the presence of serum from psoriatic patients (48). (vi) The expression of insulin-like growth factor-1 (IGF-1) and epidermal growth factor (EGF) is increased in psoriatic lesional skin (49,50). (vii) Fibroblasts from lesional and non-lesional psoriatic skin produce increased amounts of IL-8 compared with fibroblasts from normal skin (51).

Psoriatic skin, therefore, appears to be sensitized toward disease which may be triggered by exogenous (environmental) or endogenous factors (e.g. a component of the immune or vascular systems).

Psoriasis as an autoimmune disease

The immune system has been strongly implicated in the pathogenesis of psoriasis since it resembles a T cell-mediated autoimmune disease (reviewed in refs 52-54). Several studies supporting this are discussed below.

Drugs such as cyclosporin and FK506, that act by suppressing the activity of T cells, are effective in treating psoriasis (52). During lesion formation, inflammation precedes epidermal hyperproliferation and increased numbers of T cells have been demonstrated in the uninvolved skin of psoriatics (55,56). IFN-[gamma], which is produced exclusively by activated T cells, is increased in psoriatic epidermis (57). T cells isolated from lesional psoriatic skin enhance keratinocyte proliferation via secreted products (58). Furthermore, psoriasis has been reported to be cured (59) or induced by bone marrow transplants (60), and a favorable response to anti-CD4 and -CD3 monoclonal antibodies and lymphocytic toxins has been documented (61-63).

The association of psoriasis with HLA-Cw6 suggests that class I molecules in the skin might initiate activation of CD8⁺ T suppressor cells which would then lead to the activation of CD4+ cells and release of inflammatory cytokines (64). APCs are key elements in launching an immune response. They include macrophages, dendritic cells and Langerhans' cells. APCs present foreign antigens within the antigen-binding pocket of class II MHC molecules to T cells for activation. During activation, intercellular adhesion molecule-I (ICAM-I) and co-stimulatory accessory molecules such as B7 expressed on the surface of the APC bind to their respective ligands on T cells, namely, lymphocyte function-associated molecule-1 (LFA-1) and CD28/CTLA4 (65). Increased numbers of Langerhans' cells are found in psoriatic skin (66), and the CD1a-DR⁺ subset of these cells demonstrate an increased capacity for T cell activation (67). Keratinocytes can also function like APCs in regulating T cell activation (68). Endothelial leukocyte adhesion molecule-1 (ELAM-1) is induced in psoriatic dermal vasculature (69) and it facilitates the recruitment of CD4⁺ T helper cells. The dermal endothelium is known to play an important role in the recruitment of inflammatory cells in psoriasis (4,70-72).

Streptococcal throat infections frequently precede outbreaks of guttate psoriasis which can then lead to chronic plaque psoriasis. In recent years, interesting insights into the immunological effects of streptococcal infections have been made (5,73-75). Streptococcal superantigens can stimulate T cells directly without requiring prior intracellular processing (76-79). Streptococcal superantigens can also induce the expression of the skin homing receptor cutaneous lymphocyte-associated antigen (CLA) on T cells, thus directing them toward the skin (80). The streptococcal M protein is homologous to the 50 kDa type I keratin (K14) protein (81), suggesting that specific T cells may be directed against keratinocytes via a cross-reactive process (73-75). The streptococcal M protein also shares structural and immunological similarities with myosin (82), tropomyosin (83) and vimentin (84), and streptococcal lipoteichoic acids can bind laminin, a component of the basement membrane (85). Furthermore, a 67 kDa streptococcal protein is homologous to myosin and the [beta] chain of the class II HLA antigens (86). All of these studies suggest plausible initiating factors for susceptible individuals, and future investigations should help to clarify the role that such environmental factors play in the disease process.

GENETIC STUDIES

Elucidating the mode of inheritance of psoriasis has been plagued with all the problems associated with a common and complex disease. Sometimes sporadic cases can be mistaken for familial segregations because the disease is so common. Other factors that confound linkage analyses are incomplete penetrance of the trait in susceptible individuals and variations in phenotypic expression that may depend on age, gender, modifier genes and environmental trigger factors. The existence of genetic heterogeneity is likely, again decreasing the ability to detect linkage by combining scores from different families. The mode of transmission of psoriasis may also be variable. Some early studies suggested autosomal dominant inheritance with reduced penetrance (87,88), while others indicated a recessive mode of inheritance (89). Other pedigrees were explained with an oligogenic (90,91) or multifactorial model (92).

The involvement of more than one gene (multilocus model) was suggested by Elder et al. (93) who applied the method of Risch (94) to calculate [lambda]_R- 1 using data gathered from the Faroe Islands (95) and Sweden (36). One would expect [lambda]_R- 1 to decrease by a factor of 2 for each degree of relationship if a single gene is involved or if several genes contribute additively to the trait. On the other hand, if several genes interact epistatically then their individual contributions become multiplicative and [lambda]_R- 1 decreases by more than a factor of 2. Elder et al. (93) found that [lambda]_R- 1 decreased by a factor of 7 in patients from the Faroe Islands, and by a factor of 6 in patients from Sweden. We now have evidence that some cases of psoriasis are due to a highly penetrant, dominantly acting susceptibility allele (17,18; see below); however, it is not currently clear what proportion of cases are due to such predisposing alleles, and what proportion of cases are polygenic.

To map genes successfully in complex multifactorial diseases like psoriasis, a combination of several approaches may be required and include: (i) linkage analysis; (ii) allele sharing methods; (iii) association studies; and (iv) animal models (96,97). All approaches are now being performed by a variety of groups. Results from these studies are described below.

Table 1. Summary of linked regions identified and replicated in independent studies

Region Gene/locus Map position^a No. of families Model LOD score P-value Reference

17q D17S784 116.86 8 Dominant 5.33 17

17q D17S802 106.80 115 (29 extended/86 nuclear) Dominant 2.09 0.0056 (GH) 20

17q D17S785 103.53 115 0.0034 (GH) 20

17q D17S928 126.46 23 extended 0.0006 (GH) b

17q D17S795 89.32 23 extended 0.06 (GH) b

6p TNF-[alpha] 44.9 68 nuclear Recessive 6.5 0.00000058 (HRR) 19

6p D6S273 44.96 68 nuclear Recessive 5.01 0.000014 (HRR) 19

6p HLA-C 44.7 68 nuclear Recessive 3.07 0.00041 (HRR) 19

6p TNF-[beta] 44.9 115 Recessive 2.31 20

6p HLA-C 44.7 101 0.00002 (HRR) b

20p D20S186 32.30 68 nuclear Recessive 2.01 0.0012 (HRR) 19

20p D20S851 24.70 115 Recessive 2.62 0.0604 (GH) 20

GH, GENEHUNTER (ref. 102); HRR, haplotype relative risk (Terwilliger, 1996 program ANALYZE. ftp://linkage.cpmc.columbia.edu ). Map positions were obtained from the Marshfield genetic maps at http://www.marshmed.org/genetics
^acM from the top.
^b R. Barnes et al. (manuscript in preparation).

LINKED REGIONS REPORTED TO DATE

17q25

Human Molecular Genetics