The Immunologic and Genetic Basis of Psoriasis FREE

In 1991, a comprehensive review was published highlighting the importance of cytokines in the immunopathogenesis of psoriasis.¹ Since then, many advances have been made regarding the critical role for pathogenic T cells as causing this common and enigmatic skin disease. While epidermal keratinocytes are clearly involved as participants in establishment of the appropriate cytokine milieu as presented earlier, more recent data point to T lymphocytes as triggering the chain reaction of cellular and molecular networks that culminate in the formation of a psoriatic plaque. In this review, 3 major topics will be addressed.

In 1986, Valdimarsson et al² suggested that psoriasis was a skin disease in which keratinocyte proliferation was initiated by T-cell infiltration and activation. A decade later using severe combined immunodeficient mice engrafted with symptomless (PN) skin, direct in vivo evidence was provided that T lymphocytes could indeed induce conversion of PN skin to psoriatic plaques.³ Figure 1 portrays a severe combined immunodeficient mouse with an engrafted psoriatic plaque. Clinical observation using a wide range of immunosuppressive agents also established that activated T cells were required for persistence of psoriasis.^{4 - 11} Many other groups have confirmed the importance of activated T cells in the initiation and maintenance of psoriasis.^{12 - 15}

As evidence accumulated that intraepidermal T-cell activation was important in psoriasis, 2 important questions arose. First, were CD4⁺ and CD8⁺ T cells both involved in the immune reaction? Second, what was the identity of the inciting agent that triggered T-cell activation? Initial attempts to answer the first question were approached by immunohistochemical staining of biopsy specimens obtained from early, late, and resolving lesions. Conflicting results were obtained, with some investigators observing an initial influx of CD4⁺ T cells and others finding early infiltration by CD8⁺ T cells.^{16 - 20} Likewise, when psoriatic plaques were examined after administration of various treatments, the resolving lesions were characterized by prominent reductions in CD4⁺ and/or CD8⁺ T cells from the epidermis.^{9 - 11 ,21 - 23} In contrast to these divergent observations, there is more of a consensus that cytokine production by activated T cells resembles a T-helper type-1 profile with interferon gamma (IFN-γ) and interleukin (IL) 12, but not IL-4, IL-5, or IL-10.^{24 - 25} To resolve the dilemma regarding relative roles for CD4⁺ and CD8⁺ T cells, engrafted PN skin was separately injected with highly purified CD4⁺ or CD8⁺ T-cell lines. On one hand, results appeared relatively straightforward: injection of autologous activated CD4⁺ T cells produced psoriatic plaques, whereas none of the CD8⁺ T cells produced psoriasis in 5 different patients.²⁶ However, as we examined engrafted skin after introduction of CD4⁺ T cells, it became apparent that the story was more complex, as the intraepidermal T-cell population displaying acute activation markers such as CD25 (a high-affinity IL-2 receptor) and CD69 included CD8⁺ cells.²⁷ The experimental result agreed with a recent report finding epidermal-derived activated CD8⁺ T cells in plaques.²⁸ Thus, we concluded that the injection of CD4⁺ T cells created an appropriate microenvironment in which dormant intraepidermal CD8⁺ T cells already present in PN skin became activated following receipt of a "help" signal from CD4⁺ T cells. The precise nature of this help message is under investigation. Tumor immunologists have also pondered the nature of help signals, and this long-standing question has been recently answered.²⁹ When CD4⁺ T cells become activated, they express CD40L, which can then interact with antigen-presenting cells (APCs) expressing CD40 to embue them with enhanced APC function, particularly by up-regulating costimulatory molecules such as CD80 (B7-1) and CD86 (B7-2). Following recognition of tumor antigens on activated APCs, cytotoxic CD8⁺ T cells can then attack and destroy tumor cells. Figure 2 portrays an overview of a dynamic immunologic model for psoriasis highlighting sequential interactions involving CD4⁺ T cells, dendritic APCs, and CD8⁺ T cells.

As shown in Figure 3, induction of costimulatory molecules on APCs is essential for T-cell activation, as 2 different signals are required for complete T-cell proliferation. Signal 1 is provided by a specific antigen or superantigen, but without signal 2 delivered by CD80 and/or CD86, T cells cannot become fully activated, and may even become anergic. Involvement of this CD28:B7 pathway has particular relevance to dermatologists because in many skin diseases (including psoriasis), the nature of the inciting agent (ie, signal 1) is not known; but by directing therapy at signal 2, one could still block T-cell activation and potentially anergize the pathogenic T cells in many diseases, regardless of the initiating stimulus.³⁰ Using a fusion protein (CTLA4Ig) that blocks CD28:B7, improvement in psoriasis has been observed,³¹ consistent with an earlier observation that autologous T-cell activation in patients with psoriasis was inhibited in vitro by CTLA4Ig.³² However, there is another, perhaps more subtle, lesson to be learned from CTLA4Ig clinical data related to the CD28:B7 pathway. In general this activation pathway is relatively cyclosporin A–resistant.³³ Thus, T-cell activation occurring in upper levels of the dermis and epidermis may be more sensitive to inhibition by CTLA4Ig than by cyclosporin A. This hypothesis has practical treatment implications because it may be possible that topical cyclosporin A may not be effective in psoriasis because this particular activation step involving CD28:B7 may be more important than previously believed. It should also be noted that several different costimulatory ligand-receptor pairs mediate CD8⁺ T-cell activation as depicted in Figure 3, including LFA-1/ICAM-1, CD2/LFA-3,³⁴ and VLA-4/VCAM-1 (LFA indicates leukocyte function–associated antigen; ICAM, intercellular adhesion molecule; VLA, very late antigen; and VCAM, vascular cell adhesion molecule).

The second question posed earlier deals with the nature of inciting the antigen responsible for producing the immunologic events that trigger psoriasis. This question is obviously a central issue for psoriasis investigators, and several different approaches have been taken. One involves examining the distribution of various T-cell receptors in the blood and skin of patients with psoriasis. Studies of guttate psoriasis support a role for polyclonal, superantigen-driven T-cell expansion, and analysis of biopsy specimens of more chronic psoriatic plaques has revealed conflicting results with regard to both the presence or absence of a restricted and/or clonal T-cell population and on a role for nominal antigen vs superantigen. While a detailed analysis of numerous reports is beyond the scope of this review, evidence indicates selective activation and proliferation of specific subsets of CD8⁺ T cells in the epidermis of psoriatic lesions. Another approach has been to try to identify specific antigens (both exogenous and endogenous) in patients with psoriasis that could activate the immune system with creation of pathogenic T cells. Both infectious and noninfectious T-cell activating factors have been suggested as candidate triggering factors, including retrovirus (human immunodeficiency virus 1³⁵ ), bacteria-derived superantigens,^{36 - 37} streptococcal M protein and homologous keratin-derived peptides,³⁸ neuropeptides such as substance P,³⁹ and human papillomavirus.⁴⁰

Thus far we have focused on acquired-type of immune response (Figure 2 and Figure 3) where CD4⁺ or CD8⁺ T cells are activated through their T-cell receptors (signal 1) along with costimulatory molecules (signal 2). However, as Jullien et al⁴¹ of Dr Robert Modlin's laboratory have uncovered, there are also T cells expressing T-cell receptors and CD3, but devoid of CD4 and CD8 (so-called double-negative T cells) that become activated in the presence of nonprotein antigen (ie, lipids or lipoglycans) recognized in the context of CD1 molecules present on dendritic APCs. Besides these T-cell recognition systems, I would like to suggest another pathway that may be relevant in psoriasis (and other diseases) involving a different T-cell recognition mechanism. This hypothesis features surface receptors on a subset of T cells that are more generally associated with natural killer (NK) cells (so-called NK–T cells). Recently there has been tremendous interest in NK–T cells because they may provide an important bridge between innate and adoptive (or acquired) immune systems. While NK–T cells have been overlooked in dermatology, it has become clear that they are present in acute and chronic lesions of psoriasis, and this discovery raises new interest and potential novel targets for both genetic analysis and therapeutic targeting.⁴²

The secondary epidermal response to the presence of activated T cells (or NK–T cells) includes prominent and persistent epidermal thickening with scale production. Traditionally, psoriasis has been viewed as a hyperproliferative disorder in which cell-cycle kinetics and pathways of differentiation are substantially altered. Pioneering studies of epidermopoiesis in psoriasis led to initial therapeutic strategies exploring antimitotic agents also used in patients with cancer to try to halt hyperkinetic keratinocytes in the epidermis. Thus, drugs such as methotrexate or dithranol became popular with dermatologists treating patients with psoriasis. In this section, 2 major points will be addressed. First, we will review data that support the notion that T cells drive epidermal responses. Second, as oncologists have begun to appreciate, tumors can be produced without an increase in rates of cell proliferation, but by a decrease in cell death. Thus, a brief review of cell survival gene products in psoriasis will be presented.

The focus of research has gradually shifted from the keratinocyte to activated immunocytes as the prime cellular culprits in psoriasis. The precise mechanism by which activated T cells trigger psoriasis is unknown, but it may involve release of cytokines that influence the extracellular matrix and their receptors (ie, integrins) on the surface of epidermal keratinocytes.^{43 - 44} Interferon gamma is present in psoriatic epidermis,⁴⁵ and is localized to CD4⁺ and CD8⁺ T-cell subsets in psoriatic lesions.⁴⁶ When injected into PN skin, IFN-γ can provoke the appearance of psoriatic plaques.⁴⁷ However, the ability of IFN-γ to induce dramatic epidermal thickening presents a paradox because IFN-γ is the most potent antiproliferative cytokine known to arrest keratinocyte growth in vitro at picomolar levels through interaction with specific high-affinity receptors.⁴⁸ While it is possible that in vivo keratinocytes within a psoriatic plaque have an abnormal response to IFN-γ, another explanation exists for this apparent paradox. This alternative view focuses not on the growth-regulatory effect of IFN-γ on keratinocyte proliferation, but rather its ability to influence the longevity of keratinocytes by rendering them less susceptible to apoptosis.

Currently, very little is known regarding the importance and/or regulation of the programmed cell death pathway in the epidermis of normal or diseased skin. To determine if a contributing factor to the overall thickness of a plaque could result from a resistance to apoptosis, several different studies were undertaken. Initially, expression of antiapoptotic proteins such as Bcl-x and Bcl-2 were measured; psoriatic plaques contain keratinocytes with abundant Bcl-x but not Bcl-2.⁴⁹ Next, the potential biological importance of increased keratinocyte Bcl-x was examined, and keratinocytes derived from a psoriatic plaque were found to be resistant to apoptosis compared with normal skin.⁵⁰ The enhanced survival of keratinocytes may be related to IFN-γ, as this cytokine can increase levels of Bcl-x in keratinocytes, and pretreatment of normal foreskin-derived keratinocytes with IFN-γ produced a death-defying phenotype.⁵¹

The ability of keratinocytes in a psoriatic plaque to acquire an apoptotic-resistant phenotype raises several interesting points of pathophysiological and treatment relevance. One previously confusing aspect of the immunobiology of psoriasis was how psoriatic keratinocytes could resist apoptosis when they express either the Fas antigen (CD95) alone or when, after exposure to UV light, they coexpress both CD95 and the Fas ligand (CD95L)—even when activated CD8⁺ T cells containing cytotoxic granules were abundant in the epidermis.²⁸ A plausible explanation is psoriatic keratinocytes. By overexpressing, Bcl-x and other cytokine-induced antiapoptotic proteins can protect themselves against Fas-mediated apoptosis, while normal skin-containing keratinocytes remain susceptible to cell death. In therapy, clinicians may be able to deliver high levels of UV-B light to diseased skin such that pathogenic intraepidermal T cells undergo apoptosis, and yet adjacent psoriatic keratinocytes remain relatively unscathed because of their apoptotic-resistant phenotype.⁵²

Before leaving this topic, 2 other points are worth mentioning. First, it is unlikely that overexpression of Bcl-x by itself is responsible for the acanthotic epidermis typical of a plaque, because transgenic mice overexpressing Bcl-x do not exhibit features that resemble psoriasis.⁵³ Thus, other growth-regulatory or cell-cycle changes, in addition to overexpression of cell-survival gene products, may play important roles in driving epidermal thickening with elongation of rete pegs. The second topic concerns the relatively infrequent transformation of psoriatic plaques into invasive squamous cell carcinoma (reviewed in Lindelof et al⁵⁴ ). If psoriasis was truly a hyperplastic disorder resembling lymphoid hyperplasia or colonic epithelial hyperplasia as seen with polyps, it would be expected to represent a premalignant state. In addition, many initiators and/or tumor promoters (ie, anthralin, crude coal tar, etc) have been used to treat psoriasis, not to mention chronic immunosuppressive regimens and UV light with or without psoralen. The relatively infrequent development of squamous cell carcinoma suggests that in addition to psoriatic plaques containing keratinocytes that are resistant to apoptosis, these same keratinocytes may also be resistant to transformation. We currently suspect that psoriatic keratinocytes resemble senescent cells in that they resist both apoptosis and transformation.⁵¹

The primary genetic cause of psoriasis remains elusive. Many possible molecular mediators that are part of complex cytokine networks have been implicated in secondary cascades, chain reactions, and vicious cycles that perpetuate psoriatic pathology. However, the gene locus (or loci) responsible for initiating the disease process has yet to be determined. This section will briefly review the current status of the genetic basis for psoriasis and then conclude with a summary of potential candidate genes. Many epidemiological and genetic studies point to a role for both inherited and environmental factors in psoriasis.⁵⁵ As early as 1972, a potential role for the HLA region on chromosome 6 was documented.⁵⁶ More recently several groups have suggested multiple loci on many different chromosomes besides chromosome 6, including chromosomes 1, 2, 4, 8, 16, 17, and 20.^{57 - 65} While there seem to be multiple genes that contribute to the psoriatic phenotype, most investigators have focused on the MHC class I region, with particular interest on the HLA-Cw6 allele.

There are at least 2 potential pathways by which such class I alleles could function in stimulating T-cell activation, proliferation, and cytokine release. The conventional pathway would highlight the acquired immune response in which an appropriately processed peptide antigen would be presented in the context of a specific class I allele, and this would be recognized by a memory-type CD8⁺ T-cell receptor–bearing lymphocyte with this self-peptide (or non–self-peptide) specificity. A second, nonconventional recognition pathway would be more related to the innate immune response involving NK–T cells such that the NK–T cell could recognize the class I allele through a different set of cell surface receptors.

Several different receptors for MHC class I alleles have been identified that can either activate or inhibit NK cells or NK–T cells. Figure 4 portrays some of these MHC class I receptors, including CD158a, which recognizes HLA-Cw2, Cw4, Cw5, or Cw6 alleles; CD158b, which recognizes HLA-Cw1, Cw3, Cw7, or Cw8; NKB1, which recognizes HLA-Bw4; or CD94, which can recognize class I signal peptides presented by HLA-E. CD94 is a lectin type C receptor differing from CD158a and CD158b, which belong to the immunoglobulin gene superfamily.⁶⁶ Recently, we documented that T cells expressing CD94, CD158a, and CD158b are present in psoriatic plaques, and hence such previously overlooked NK–T cells may be participating as pathogenic or regulatory immunocytes in psoriasis.⁴²

Besides the classical class I MHC molecules encoded by genes on chromosome 6, there are several other ligand-receptors worth mentioning in psoriasis. The nonclassical MHC-like CD1d antigen has recently been observed to be overexpressed by keratinocytes in psoriatic plaques.⁴² Previously, CD1d was believed to be restricted to thymus and intestinal epithelium where it could mediate T-cell activation. We have postulated that CD161-positive NK–T cells may be participating in psoriasis involving CD161 and CD1d.⁴² Whether epidermal-derived glycolipids, including ceramide-containing glycolipids, could play a role as antigens or as costimulatory molecules with the CD1d expressed by keratinocytes is also under active investigation. The CD1d molecule is located on chromosome 1q near other genes that are also implicated in psoriasis.⁶⁵ A potentially important transmembrane residue in CD1d, as well as in HLA-Cw6, that can influence interaction with NK cells requires further study in psoriasis.⁶⁷

Finally, returning to chromosome 6, there are also nonclassical MHC antigens that are highly polymorphic, including MICA (a polymorphic HLA-related antigen), PERB 11, and the S gene.^{68 - 72} While MICA is present within keratinocytes, it has not been demonstrated to be present on the surface of human keratinocytes in vitro or in vivo.⁷² However, in human intestinal epithelium, γδ⁺ T cells were activated by MICA-bearing epithelial cells,⁷³ suggesting that this pathway may be of interest to psoriasis investigators. Whether functional homologs to these types of γδ T cells exist in human skin⁷⁴ and whether there is overlap in phenotype between NK-T cells and MHC class I vs CD161 and CD1d vs γδ T cells and MICA is unknown.

In conclusion, many new discoveries and insights related to the immunology and genetics of psoriasis are emerging from both biological approaches and genome-wide searches. New insights into how soluble products of immunocytes can influence the phenotype of keratinocytes with respect to resistance to apoptosis and transformation have also been gained by recent studies in psoriasis. Previously overlooked subsets of immunocytes are being pursued, and novel ligand-receptor interactions that regulate immune response can now be considered for both pathophysiological and therapeutic relevance. The recognition of a possible involvement of the innate immune system through NK–T cells may also explain the long-standing question of why psoriasis is so prevalent. Could the presence of such NK–T cells afford protection against bacterial and/or viral and/or fungal infection and development of neoplasms in the skin⁴² ? Clearly, since the last comprehensive review in 1991, much progress has been made and new avenues worth pursuing using validated animal models are now in place to facilitate discovery of the cause of this common and chronic skin disease.^{1 ,75}

Accepted for publication June 22, 1999.

This work has been continuously supported by National Institutes of Health grant AR40065, Bethesda, Md.

A longer reference section was prepared for the initial version of this article, but because of space constraints, substantial shortening was necessary.

This review is dedicated to my mentor and friend, Dr Eugene Farber, who first introduced me to psoriasis 15 years ago, and to the memory of my father, Vassel Bill Nickoloff.

Reprints: Brian J. Nickoloff, MD, PhD, Department of Pathology, Skin Cancer Research Laboratories, Loyola University Medical Center, Cardinal Bernardin Cancer Center, 2160 S First Ave, Maywood, Ill 60153 (e-mail: bnickol@luc.edu).