- Email: [email protected]
Retinoids and psoriasis: Novel issues in retinoid pharmacology and implications for psoriasis treatment
SUPPORTED
BY A GRANT FROM ROCHE LABORATORIES, INC., NUTLEY, N.J.
Retinoids and psoriasis: Novel issues in retinoid pharmacology and implications for psoriasis treatment J-H Saurat, MD Geneva, Switzerland Oral synthetic retinoids have been established as effective systemic therapy for psoriasis since their introduction for clinical use in the 1970s. Acitretin, the free acid of etretinate and its active metabolite, has replaced etretinate as the retinoid of choice for treating psoriasis because of its more favorable pharmacokinetic profile. Despite the demonstrated clinical success of retinoid therapy in psoriasis and other proliferative skin disorders, their mechanism of action has not been fully elucidated. Altered vitamin A metabolism, characterized by an increase in the formation of retinoic acid, has been demonstrated in psoriatic lesions and is potentially influenced by cytokines such as interferon gamma, which is present in high levels in these lesions. Synthetic retinoids such as acitretin may interfere with such cytokine-induced alterations. Studies on nuclear retinoic acid receptors have shown that acitretin activates all 3 receptor subtypes (RAR-alpha, -beta, and -gamma) without measurable receptor binding; this paradox remains unexplained. Further studies on nuclear receptor binding and activity, including possible receptor crosstalk with vitamin D nuclear receptors, promise to enhance understanding of the usefulness of retinoids in treatment of psoriasis. (J Am Acad Dermatol 1999;41:S2-6.)
T
he discovery that vitamin A metabolites known as retinoids were effective agents in treating psoriasis represented a major advance in the development of therapies for the disease. Oral synthetic retinoids were introduced for clinical use for cancer and other proliferative disorders in the early 1970s. Etretinate, the first such compound to be used in the treatment of psoriasis, yielded positive results both as monotherapy and in combination with PUVA and other treatment modalities. However, concerns about the safety of etretinate related to its teratogenicity and long elimination half-life, resulting from storage in fat tissues, caused a waning of interest in its clinical use in the 1980s. With the late 1980s introduction of acitretin, the free acid of etretinate, interest in retinoids as systemic treatment for psoriasis has been renewed. Acitretin, the pharmacologically active metabolite of etretinate, is 50 times less lipophilic than etretinate
and has an elimination half-life of 2 days compared with 100 days for etretinate.1 Unlike etretinate, acitretin is not stored in adipose tissue, although evidence suggesting that small amounts of acitretin may be reesterified to etretinate under certain conditions has raised concerns about long-term effects.1 Like etretinate, acitretin has been shown to be effective both as monotherapy and in combination with other psoriasis treatments.1 Despite the clinical efficacy of retinoids in the treatment of psoriasis, the mechanism of their therapeutic action has not been established. Evidence to date has pointed to altered vitamin A metabolism in psoriasis. Studies aided by molecular biological techniques have suggested that synthetic retinoids may interfere with these changes. The results of ongoing studies correlating clinical activity and cellular effects of retinoids promise to improve their effectiveness in the treatment of psoriasis.
EARLY STUDIES From the Department of Dermatology, University Hospital, Geneva. This manuscript is based on a presentation given at the 5th European Congress on Psoriasis/7th International Psoriasis Symposium in Milan, Italy on September 2, 1998, with support from Roche Laboratories, Inc., Nutley, NJ. Reprint requests: J-H Saurat, MD, Department of Dermatology, University Hospital, CH-1211 Geneva 14–Switzerland. Copyright © 1999 by the American Academy of Dermatology, Inc. 0190-9622/99/$8.00 + 0 16/0/100479
S2
Interest in vitamin A metabolites for treating psoriasis resulted from research on their use as anticancer agents. In initial animal studies conducted in the early 1970s, etretinate was found to be effective in reducing skin papillomas induced by topical application of oncogenic substances in mice.2 This finding led to studies on the use of retinoids for other proliferative skin disorders.3 In addition, because psori-
Saurat S3
J AM A C A D D ERMATOL V OLUME 41, NUMBER 3, PART 2
Fig 1. Cellular retinoid pathway. RA, Retinoic acid; CRABP, cellular retinoic acid binding proteins; CRBP, cellular retinol binding protein; CYP, cytochrome P450 enzyme; RAR, nuclear retinoic acid receptor; RXR, nuclear retinoid X receptor. For further details see Napoli JL. Retinoic acid biosynthesis and metabolism. FASEB J 1996;10:993-1001.
Fig 2. Rate of retinoic acid formation by cytosolic extracts of human skin from [3H] retinol or [3H] retinaldehyde. RAL, [3H] retinaldehyde; ROL, [3H] retinol. Adapted with permission from Siegenthaler G, Gumovski-Sunek D, Saurat JH. Metabolism of natural retinoids in psoriatic epidermis. J Invest Dermatol 1990;95:S47-8. Reprinted by permission of Blackwell Science, Inc.
asis is a disease involving cellular proliferation and altered differentiation, vitamin A derivatives, known to have antiproliferative activity and to promote cellular differentiation, were expected to reverse the disease process.
ALTERATION IN THE CELLULAR RETINOID PATHWAYS IN PSORIASIS Although the mechanisms involved in the onset of psoriasis are not fully understood, studies have suggested that altered cellular metabolism of vitamin A may be associated with the disease process. Studies examining each step of the cellular retinoid pathway in psoriatic skin compared with normal skin have been conducted.
Basic cellular retinoid pathway The basic cellular pathway leading to conversion of retinol, the naturally occurring form of vitamin A, to all-trans-retinoic acid (ATRA), the biologically active ligand that binds to cellular receptors, has been well-characterized (Fig 1). Retinol is first converted to retinal by the enzyme retinol dehydrogenase in a reversible reaction. Retinal, in turn, is converted to ATRA by retinal dehydrogenase. A binding protein, designated cellular retinol binding protein-I (CRBP 1), facilitates these enzymatic reactions. Two additional binding proteins known as cellular retinoic acid binding proteins, or CRABP 1 and CRABP 2, are thought to act, at least in part, to buffer the levels of free ATRA in the cell by sequestrating
S4 Saurat
J AM A C A D D ERMATOL SEPTEMBER 1999
Table I. Pattern of CRBP and CRABP expression in normal skin vs psoriatic skin vs normal skin exposed to a retinoic acid load*
CRBP 1 CRBP 2 CRABP 1 CRABP 2
Normal skin
Psoriasis
RA load
++ — ++ +
++ — — +++
+++ — — +++
CRABP, cellular retinoic acid binding protein; CRBP, Cellular retinol binding protein; RA, retinoic acid. Data from references 8-12. *Both naturally occurring and synthetic retinoic acid compounds tested.
ATRA and by promoting metabolism of ATRA by cytochrome P450 enzymes belonging to the newly characterized family CYP 26.4 ATRA is the active ligand that binds to the 3 known types of nuclear retinoic acid receptors (RARs alpha, beta, and gamma), which mediate the cellular effects of retinoic acid. 9-cis-retinoic acid is the natural ligand of nuclear retinoid X receptors, (RXRs) and also binds to RARs.5 Alterations found in psoriatic skin Investigation of the cellular retinoid pathway in both normal skin and psoriatic skin lesions has involved assessment of retinoid levels, as well as the activity of enzymes, binding proteins, and retinoid receptors. Examining possible alterations in absolute levels of vitamin A metabolites associated with psoriasis, Rollman and Vahlquist6 found no significant difference in baseline levels of retinoic acid precursors, retinol and dehydroretinol, in nonlesional or lesional psoriatic skin compared with normal skin. In this study, however, levels of retinoic acid and dehydroretinoic acid were not detectable in either psoriatic or normal skin samples, likely because of the detection limits of the methods used. Assessment of the rate of retinoic acid formation by cytosolic extracts loaded with radiolabeled retinal and retinol precursors, however, has demonstrated increased production of retinoic acid by extracts from psoriatic skin compared with normal skin. This suggests increased activity of both enzymes involved in cellular ATRA production in psoriatic skin (Fig 2).7 Measurement of binding proteins showed significantly higher levels of the ATRA-buffering protein CRABP 2 in extracts from psoriatic plaques compared with those of normal skin.8-11 The overall pattern of binding protein expression seen in psoriasis parallels that seen in normal skin
exposed to exogenous retinoic acid (Table I).12 Thus, although the study by Rollman and Vahlquist6 failed to provide direct evidence of increased retinoic acid in psoriatic plaques because of methodologic limitations, indirect evidence suggests that psoriatic plaques are likely to have higher concentrations of retinoic acid than normal skin. In studies of retinoic acid nuclear receptors, no significant changes in RAR gene expression or levels of RAR and RXR proteins have been found to be associated with psoriasis.13,14 Taken together, studies of the cellular retinoid pathway suggest that alterations in retinoic acid metabolism in psoriatic skin are similar to those seen in normal skin in response to a retinoic acid load. These alterations include increased expression of the binding protein CRABP 2, which likely serves to sequester excess ATRA.4 Effect of interferon γ and other cytokines Psoriasis is increasingly considered to be a disease mediated by a Th 1 type cytokine network. Recent observations suggest that altered vitamin A metabolism in psoriatic lesions is a secondary event resulting from the activity of these cytokines. Support for this theory has included studies on interferon gamma (IFN γ), a Th1 type cytokine present in high levels in psoriatic skin. Törmä, Rollman, and Vahlquist15 have recently reported that exposure to IFN γ resulted in increased production of retinoic acid from precursors such as retinol in cultured keratinocytes. The extent to which the resultant increase in retinoic acid contributes to the expression of the psoriatic phenotype (characterized by aberrant differentiation) remains to be demonstrated.
THERAPEUTIC EFFECT OF ACITRETIN Given the evidence suggesting that psoriasis involves altered vitamin A metabolism resulting in high levels of retinoic acid in psoriatic plaques, the fact that retinoid therapy is effective for psoriasis treatment appears to represent a paradox. Explanations for this apparent paradox are presently hypothetical. Hypotheses as to how acitretin may act to disrupt the psoriatic disease process include several possible mechanisms. Although based on indirect observations, they provide useful models for further investigation into the mechanism of action of acitretin in psoriasis. First, acitretin may inhibit the cytokine-induced increase in retinoic acid formation. This is supported by the finding that acitretin inhibits the enzymatic formation of retinoic acid from retinol.16 The effect of acitretin on the expression of cytokines such as
Saurat S5
J AM A C A D D ERMATOL V OLUME 41, NUMBER 3, PART 2
Table II. Retinoic acid nuclear receptor selectivity and activity of selected synthetic retinoids Receptor activation* RAR α
Tretinoin 100 Isotretinoin 70 Etretinate 14 Acitretin 61
Receptor binding* RXR
RAR
RXR
β
γ
α
β
γ
α
β
γ
α
β
γ
100 64 21 71
100 70 20 79
98 90 0 0
160 -
140 -
100 0 0 0
100 0 0 0
100 0 0 0
0 10 0 0
0 0 0 0
0† 8 0 0
Adapted with permission from LeMotte PK, Keidel S, Apfel CM. Characterization of synthetic retinoids with selectivity for retinoic acid or retinoid X nuclear receptors. Biochem Biophys Acta 1996;1289:298-304. With permission from Elsevier Science. *% max. †Less than 20%.
interleukin 1 in skin tissue also should be considered in this context.17 A second hypothesis is that acitretin may stimulate metabolism and buffering of cytokine-induced increases in retinoic acid by inducing increased cellular levels of CRABP 2. This is consistent with studies in which systemic treatment with acitretin has been shown to increase expression of CRABP 2 in the epidermis.18 A third hypothesis is that synthetic retinoids such as acitretin may alter metabolism of endogenous retinoids at the level of degradation. For example, increased intracellular ATRA levels in psoriatic lesions could theoretically result from a defective CYP 26 enzyme, the activity of which may be modulated by synthetic retinoids.
NUCLEAR RECEPTOR BINDING OF SYNTHETIC RETINOIDS Pharmacologic activity of synthetic retinoids is likely to be mediated through binding and activation of nuclear receptors; these include alpha, beta, and gamma subtypes of both the RAR and RXR receptors.5 Current research is aimed at identifying molecules that selectively bind and activate specific receptor subtypes. Underlying this research is the expectation that each receptor subtype is tissue-specific and mediates a set of unique biological functions, and that use of subtype-selective retinoids will thus yield therapeutic specificity. This strategy, however, has several limitations. RARs are known to form heterodimers with RXRs.5 The resulting heterodimers have been shown to exert both specific and redundant functions in differentiation, proliferation, and apoptosis. However, to date, none of these heterodimers have been demonstrated to exert a unique function in skin. Furthermore, the concentrations of subtype-selective retinoids such as tazarotene (an
RAR-beta and RAR-gamma–specific ligand) achieved in the skin after topical use are in the micromolar range, levels associated with nonspecific binding. Acitretin has been shown to activate all 3 subtypes of RAR, despite the absence of measurable binding to any of the subtypes in vitro (Table II).19 Such data indicating receptor activation without demonstrated receptor binding represent another paradox regarding the mechanism of action of retinoids and underscore the need for further study on the interaction of acitretin with nuclear receptors. Nuclear receptor crosstalk Topical vitamin D derivatives have proven effective in the treatment of psoriasis. Like retinoic acid, thyroid hormone, and steroids, 1,25-dihydroxyvitamin D3 regulates a number of developmental and physiological processes in vertebrates by binding to a specific vitamin D receptor (VDR) that functions as a transcription factor. These ligand-dependent transcription factors constitute a superfamily of nuclear receptors. Upon ligand binding, nuclear receptors bind to short, specific DNA sequences called response elements in the vicinity of target genes and stimulate, or in some cases repress, transcription. According to the current model, VDRs function as heterodimers with the RXRs, which are also members of the nuclear receptor superfamily.5 It has also been suggested that VDRs may form heterodimers with RAR. Although in vivo interactions between retinoic acid and vitamin D signaling pathways need further investigation, these molecular models suggest that retinoids and vitamin D derivatives may act synergistically in the treatment of psoriasis possibly through cooperative regulation of a single gene.20 Some clinical observations suggest that the combination of calcipotriol and acitretin may be effective; however, much work is still needed in this area.1
S6 Saurat
CONCLUSION Retinoids are important therapeutic agents in the treatment of psoriasis. Their mechanism of action, however, is still unknown. Studies of possible mechanisms to date have suggested that synthetic retinoids such as acitretin may act to modulate changes in cellular retinoid metabolism induced by cytokines such as IFN γ, which has been implicated in the disease process. Further study is needed on the function of retinoic acid nuclear receptors, including their activation by acitretin and interaction with other nuclear receptors. REFERENCES 1. Saurat JH. Systemic retinoids: what’s new? Dermatol Clin 1998;16:331-40. 2. Bollag W. Therapeutic effects of an aromatic acid analog on chemically induced skin papillomas and carcinomas of mice. Eur J Cancer 1974;10:731-7. 3. Bollag W. Vitamin A and retinoids, from nutrition to pharmacotherapy in dermatology and oncology. Lancet 1983;1(8329): 860-3. 4. Napoli JL. Retinoic acid biosynthesis and metabolism. FASEB J 1996;10:993-1001. 5. Chambon P. The retinoid signaling pathway: molecular and genetic analysis. Semin Cell Biol 1994;5:115-25. 6. Rollman O, Vahlquist A. Psoriasis and vitamin A: plasma transport and skin content of retinol, dehydroretinol, and carotenoids in adult patients versus healthy controls. Arch Dermatol Res 1985;278:17-24. 7. Siegenthaler G, Gumovski-Sunek D, Saurat JH. Metabolism of natural retinoids in psoriatic epidermis. J Invest Dermatol 1990;95:S47-8. 8. Siegenthaler G, Saurat JH, Hotz R, Camenzind M, Mérot Y. Cellular retinoic acid, but not cellular retinol-binding protein, is elevated in psoriatic plaques. J Invest Dermatol 1986;86:42-5. 9. Didierjean L, Durand B, Saurat JH. Cellular retinoic acid-binding protein type 2 mRNA is overexpressed in human psoriatic skin as shown by in situ hybridization. Biochem Biophys Res Commun 1991;180:204-8.
J AM A C A D D ERMATOL SEPTEMBER 1999
10. Siegenthaler G, Tomatis I, Didierjean L, Jaconi S, Saurat JH. Overexpression of cellular retinoic acid-binding protein type II (CRABP2) and down-regulation of CRABP1 in psoriatic skin. Dermatology 1992;185:251-6. 11. Busch C, Siegenthaler G, Vahlquist A, et al. Expression of cellular retinoid-binding proteins during normal and abnormal epidermal differentiation. J Invest Dermatol 1992;99:795-802. 12. Hirschel-Scholz S, Siegenthaler G, Saurat JH. Ligand-specific and nonspecific in vivo modulation of human epidermal cellular retinoic acid binding protein (CRABP). Eur J Clin Invest 1989;2:220-7. 13. Elder JT, Astrom A, Pettersson U, et al. Differential regulation of retinoic acid receptors and binding proteins in human skin. J Invest Dermatol 1992;98:673-9. 14. Reichrath J, Munssinger T, Kerber A, et al. In situ detection of retinoid-X receptor expression in normal and psoriatic human skin. Br J Dermatol 1995;133:168-75. 15. Törmä H, Rollman O, Vahlquist A. Interferon-γ increases retinoic acid and 3,4-didehydroretinoic acid concentrations in cultured keratinocytes: a clue to the abnormal vitamin A metabolism in psoriatic skin [abstract]? J Invest Dermatol 1998;110:551. 16. Siegenthaler G, Saurat JH. Natural retinoids: metabolism and transport in human epidermal cells. In: Saurat JH, editor. Retinoids: 10 Years On. Basel: Karger; 1991:56-68. 17. Schmitt A, Hauser C, Didierjean L, Mérot Y, Dayer JM, Saurat JH. Systemic administration of etretin increases epidermal interleukin-1 in the rat. Br J Dermatol 1987;116:615-22. 18. Siegenthaler G, Saurat JH. Therapy with synthetic retinoid (Ro 10-1670) etretin increases the cellular retinoic acid-binding protein in nonlesional psoriatic skin. J Invest Dermatol 1986; 87:122-4. 19. LeMotte PK, Keidel S, Apfel CM. Characterization of synthetic retinoids with selectivity for retinoic acid or retinoid X nuclear receptors. Biochim Biophys Acta 1996;1289:298-304. 20. Carlberg C, Saurat JH. Vitamin D-retinoid association: molecular basis and clinical applications. J Invest Dermatol Symp Proc 1996;1:82-6.
BY A GRANT FROM ROCHE LABORATORIES, INC., NUTLEY, N.J.
Retinoids and psoriasis: Novel issues in retinoid pharmacology and implications for psoriasis treatment J-H Saurat, MD Geneva, Switzerland Oral synthetic retinoids have been established as effective systemic therapy for psoriasis since their introduction for clinical use in the 1970s. Acitretin, the free acid of etretinate and its active metabolite, has replaced etretinate as the retinoid of choice for treating psoriasis because of its more favorable pharmacokinetic profile. Despite the demonstrated clinical success of retinoid therapy in psoriasis and other proliferative skin disorders, their mechanism of action has not been fully elucidated. Altered vitamin A metabolism, characterized by an increase in the formation of retinoic acid, has been demonstrated in psoriatic lesions and is potentially influenced by cytokines such as interferon gamma, which is present in high levels in these lesions. Synthetic retinoids such as acitretin may interfere with such cytokine-induced alterations. Studies on nuclear retinoic acid receptors have shown that acitretin activates all 3 receptor subtypes (RAR-alpha, -beta, and -gamma) without measurable receptor binding; this paradox remains unexplained. Further studies on nuclear receptor binding and activity, including possible receptor crosstalk with vitamin D nuclear receptors, promise to enhance understanding of the usefulness of retinoids in treatment of psoriasis. (J Am Acad Dermatol 1999;41:S2-6.)
T
he discovery that vitamin A metabolites known as retinoids were effective agents in treating psoriasis represented a major advance in the development of therapies for the disease. Oral synthetic retinoids were introduced for clinical use for cancer and other proliferative disorders in the early 1970s. Etretinate, the first such compound to be used in the treatment of psoriasis, yielded positive results both as monotherapy and in combination with PUVA and other treatment modalities. However, concerns about the safety of etretinate related to its teratogenicity and long elimination half-life, resulting from storage in fat tissues, caused a waning of interest in its clinical use in the 1980s. With the late 1980s introduction of acitretin, the free acid of etretinate, interest in retinoids as systemic treatment for psoriasis has been renewed. Acitretin, the pharmacologically active metabolite of etretinate, is 50 times less lipophilic than etretinate
and has an elimination half-life of 2 days compared with 100 days for etretinate.1 Unlike etretinate, acitretin is not stored in adipose tissue, although evidence suggesting that small amounts of acitretin may be reesterified to etretinate under certain conditions has raised concerns about long-term effects.1 Like etretinate, acitretin has been shown to be effective both as monotherapy and in combination with other psoriasis treatments.1 Despite the clinical efficacy of retinoids in the treatment of psoriasis, the mechanism of their therapeutic action has not been established. Evidence to date has pointed to altered vitamin A metabolism in psoriasis. Studies aided by molecular biological techniques have suggested that synthetic retinoids may interfere with these changes. The results of ongoing studies correlating clinical activity and cellular effects of retinoids promise to improve their effectiveness in the treatment of psoriasis.
EARLY STUDIES From the Department of Dermatology, University Hospital, Geneva. This manuscript is based on a presentation given at the 5th European Congress on Psoriasis/7th International Psoriasis Symposium in Milan, Italy on September 2, 1998, with support from Roche Laboratories, Inc., Nutley, NJ. Reprint requests: J-H Saurat, MD, Department of Dermatology, University Hospital, CH-1211 Geneva 14–Switzerland. Copyright © 1999 by the American Academy of Dermatology, Inc. 0190-9622/99/$8.00 + 0 16/0/100479
S2
Interest in vitamin A metabolites for treating psoriasis resulted from research on their use as anticancer agents. In initial animal studies conducted in the early 1970s, etretinate was found to be effective in reducing skin papillomas induced by topical application of oncogenic substances in mice.2 This finding led to studies on the use of retinoids for other proliferative skin disorders.3 In addition, because psori-
Saurat S3
J AM A C A D D ERMATOL V OLUME 41, NUMBER 3, PART 2
Fig 1. Cellular retinoid pathway. RA, Retinoic acid; CRABP, cellular retinoic acid binding proteins; CRBP, cellular retinol binding protein; CYP, cytochrome P450 enzyme; RAR, nuclear retinoic acid receptor; RXR, nuclear retinoid X receptor. For further details see Napoli JL. Retinoic acid biosynthesis and metabolism. FASEB J 1996;10:993-1001.
Fig 2. Rate of retinoic acid formation by cytosolic extracts of human skin from [3H] retinol or [3H] retinaldehyde. RAL, [3H] retinaldehyde; ROL, [3H] retinol. Adapted with permission from Siegenthaler G, Gumovski-Sunek D, Saurat JH. Metabolism of natural retinoids in psoriatic epidermis. J Invest Dermatol 1990;95:S47-8. Reprinted by permission of Blackwell Science, Inc.
asis is a disease involving cellular proliferation and altered differentiation, vitamin A derivatives, known to have antiproliferative activity and to promote cellular differentiation, were expected to reverse the disease process.
ALTERATION IN THE CELLULAR RETINOID PATHWAYS IN PSORIASIS Although the mechanisms involved in the onset of psoriasis are not fully understood, studies have suggested that altered cellular metabolism of vitamin A may be associated with the disease process. Studies examining each step of the cellular retinoid pathway in psoriatic skin compared with normal skin have been conducted.
Basic cellular retinoid pathway The basic cellular pathway leading to conversion of retinol, the naturally occurring form of vitamin A, to all-trans-retinoic acid (ATRA), the biologically active ligand that binds to cellular receptors, has been well-characterized (Fig 1). Retinol is first converted to retinal by the enzyme retinol dehydrogenase in a reversible reaction. Retinal, in turn, is converted to ATRA by retinal dehydrogenase. A binding protein, designated cellular retinol binding protein-I (CRBP 1), facilitates these enzymatic reactions. Two additional binding proteins known as cellular retinoic acid binding proteins, or CRABP 1 and CRABP 2, are thought to act, at least in part, to buffer the levels of free ATRA in the cell by sequestrating
S4 Saurat
J AM A C A D D ERMATOL SEPTEMBER 1999
Table I. Pattern of CRBP and CRABP expression in normal skin vs psoriatic skin vs normal skin exposed to a retinoic acid load*
CRBP 1 CRBP 2 CRABP 1 CRABP 2
Normal skin
Psoriasis
RA load
++ — ++ +
++ — — +++
+++ — — +++
CRABP, cellular retinoic acid binding protein; CRBP, Cellular retinol binding protein; RA, retinoic acid. Data from references 8-12. *Both naturally occurring and synthetic retinoic acid compounds tested.
ATRA and by promoting metabolism of ATRA by cytochrome P450 enzymes belonging to the newly characterized family CYP 26.4 ATRA is the active ligand that binds to the 3 known types of nuclear retinoic acid receptors (RARs alpha, beta, and gamma), which mediate the cellular effects of retinoic acid. 9-cis-retinoic acid is the natural ligand of nuclear retinoid X receptors, (RXRs) and also binds to RARs.5 Alterations found in psoriatic skin Investigation of the cellular retinoid pathway in both normal skin and psoriatic skin lesions has involved assessment of retinoid levels, as well as the activity of enzymes, binding proteins, and retinoid receptors. Examining possible alterations in absolute levels of vitamin A metabolites associated with psoriasis, Rollman and Vahlquist6 found no significant difference in baseline levels of retinoic acid precursors, retinol and dehydroretinol, in nonlesional or lesional psoriatic skin compared with normal skin. In this study, however, levels of retinoic acid and dehydroretinoic acid were not detectable in either psoriatic or normal skin samples, likely because of the detection limits of the methods used. Assessment of the rate of retinoic acid formation by cytosolic extracts loaded with radiolabeled retinal and retinol precursors, however, has demonstrated increased production of retinoic acid by extracts from psoriatic skin compared with normal skin. This suggests increased activity of both enzymes involved in cellular ATRA production in psoriatic skin (Fig 2).7 Measurement of binding proteins showed significantly higher levels of the ATRA-buffering protein CRABP 2 in extracts from psoriatic plaques compared with those of normal skin.8-11 The overall pattern of binding protein expression seen in psoriasis parallels that seen in normal skin
exposed to exogenous retinoic acid (Table I).12 Thus, although the study by Rollman and Vahlquist6 failed to provide direct evidence of increased retinoic acid in psoriatic plaques because of methodologic limitations, indirect evidence suggests that psoriatic plaques are likely to have higher concentrations of retinoic acid than normal skin. In studies of retinoic acid nuclear receptors, no significant changes in RAR gene expression or levels of RAR and RXR proteins have been found to be associated with psoriasis.13,14 Taken together, studies of the cellular retinoid pathway suggest that alterations in retinoic acid metabolism in psoriatic skin are similar to those seen in normal skin in response to a retinoic acid load. These alterations include increased expression of the binding protein CRABP 2, which likely serves to sequester excess ATRA.4 Effect of interferon γ and other cytokines Psoriasis is increasingly considered to be a disease mediated by a Th 1 type cytokine network. Recent observations suggest that altered vitamin A metabolism in psoriatic lesions is a secondary event resulting from the activity of these cytokines. Support for this theory has included studies on interferon gamma (IFN γ), a Th1 type cytokine present in high levels in psoriatic skin. Törmä, Rollman, and Vahlquist15 have recently reported that exposure to IFN γ resulted in increased production of retinoic acid from precursors such as retinol in cultured keratinocytes. The extent to which the resultant increase in retinoic acid contributes to the expression of the psoriatic phenotype (characterized by aberrant differentiation) remains to be demonstrated.
THERAPEUTIC EFFECT OF ACITRETIN Given the evidence suggesting that psoriasis involves altered vitamin A metabolism resulting in high levels of retinoic acid in psoriatic plaques, the fact that retinoid therapy is effective for psoriasis treatment appears to represent a paradox. Explanations for this apparent paradox are presently hypothetical. Hypotheses as to how acitretin may act to disrupt the psoriatic disease process include several possible mechanisms. Although based on indirect observations, they provide useful models for further investigation into the mechanism of action of acitretin in psoriasis. First, acitretin may inhibit the cytokine-induced increase in retinoic acid formation. This is supported by the finding that acitretin inhibits the enzymatic formation of retinoic acid from retinol.16 The effect of acitretin on the expression of cytokines such as
Saurat S5
J AM A C A D D ERMATOL V OLUME 41, NUMBER 3, PART 2
Table II. Retinoic acid nuclear receptor selectivity and activity of selected synthetic retinoids Receptor activation* RAR α
Tretinoin 100 Isotretinoin 70 Etretinate 14 Acitretin 61
Receptor binding* RXR
RAR
RXR
β
γ
α
β
γ
α
β
γ
α
β
γ
100 64 21 71
100 70 20 79
98 90 0 0
160 -
140 -
100 0 0 0
100 0 0 0
100 0 0 0
0 10 0 0
0 0 0 0
0† 8 0 0
Adapted with permission from LeMotte PK, Keidel S, Apfel CM. Characterization of synthetic retinoids with selectivity for retinoic acid or retinoid X nuclear receptors. Biochem Biophys Acta 1996;1289:298-304. With permission from Elsevier Science. *% max. †Less than 20%.
interleukin 1 in skin tissue also should be considered in this context.17 A second hypothesis is that acitretin may stimulate metabolism and buffering of cytokine-induced increases in retinoic acid by inducing increased cellular levels of CRABP 2. This is consistent with studies in which systemic treatment with acitretin has been shown to increase expression of CRABP 2 in the epidermis.18 A third hypothesis is that synthetic retinoids such as acitretin may alter metabolism of endogenous retinoids at the level of degradation. For example, increased intracellular ATRA levels in psoriatic lesions could theoretically result from a defective CYP 26 enzyme, the activity of which may be modulated by synthetic retinoids.
NUCLEAR RECEPTOR BINDING OF SYNTHETIC RETINOIDS Pharmacologic activity of synthetic retinoids is likely to be mediated through binding and activation of nuclear receptors; these include alpha, beta, and gamma subtypes of both the RAR and RXR receptors.5 Current research is aimed at identifying molecules that selectively bind and activate specific receptor subtypes. Underlying this research is the expectation that each receptor subtype is tissue-specific and mediates a set of unique biological functions, and that use of subtype-selective retinoids will thus yield therapeutic specificity. This strategy, however, has several limitations. RARs are known to form heterodimers with RXRs.5 The resulting heterodimers have been shown to exert both specific and redundant functions in differentiation, proliferation, and apoptosis. However, to date, none of these heterodimers have been demonstrated to exert a unique function in skin. Furthermore, the concentrations of subtype-selective retinoids such as tazarotene (an
RAR-beta and RAR-gamma–specific ligand) achieved in the skin after topical use are in the micromolar range, levels associated with nonspecific binding. Acitretin has been shown to activate all 3 subtypes of RAR, despite the absence of measurable binding to any of the subtypes in vitro (Table II).19 Such data indicating receptor activation without demonstrated receptor binding represent another paradox regarding the mechanism of action of retinoids and underscore the need for further study on the interaction of acitretin with nuclear receptors. Nuclear receptor crosstalk Topical vitamin D derivatives have proven effective in the treatment of psoriasis. Like retinoic acid, thyroid hormone, and steroids, 1,25-dihydroxyvitamin D3 regulates a number of developmental and physiological processes in vertebrates by binding to a specific vitamin D receptor (VDR) that functions as a transcription factor. These ligand-dependent transcription factors constitute a superfamily of nuclear receptors. Upon ligand binding, nuclear receptors bind to short, specific DNA sequences called response elements in the vicinity of target genes and stimulate, or in some cases repress, transcription. According to the current model, VDRs function as heterodimers with the RXRs, which are also members of the nuclear receptor superfamily.5 It has also been suggested that VDRs may form heterodimers with RAR. Although in vivo interactions between retinoic acid and vitamin D signaling pathways need further investigation, these molecular models suggest that retinoids and vitamin D derivatives may act synergistically in the treatment of psoriasis possibly through cooperative regulation of a single gene.20 Some clinical observations suggest that the combination of calcipotriol and acitretin may be effective; however, much work is still needed in this area.1
S6 Saurat
CONCLUSION Retinoids are important therapeutic agents in the treatment of psoriasis. Their mechanism of action, however, is still unknown. Studies of possible mechanisms to date have suggested that synthetic retinoids such as acitretin may act to modulate changes in cellular retinoid metabolism induced by cytokines such as IFN γ, which has been implicated in the disease process. Further study is needed on the function of retinoic acid nuclear receptors, including their activation by acitretin and interaction with other nuclear receptors. REFERENCES 1. Saurat JH. Systemic retinoids: what’s new? Dermatol Clin 1998;16:331-40. 2. Bollag W. Therapeutic effects of an aromatic acid analog on chemically induced skin papillomas and carcinomas of mice. Eur J Cancer 1974;10:731-7. 3. Bollag W. Vitamin A and retinoids, from nutrition to pharmacotherapy in dermatology and oncology. Lancet 1983;1(8329): 860-3. 4. Napoli JL. Retinoic acid biosynthesis and metabolism. FASEB J 1996;10:993-1001. 5. Chambon P. The retinoid signaling pathway: molecular and genetic analysis. Semin Cell Biol 1994;5:115-25. 6. Rollman O, Vahlquist A. Psoriasis and vitamin A: plasma transport and skin content of retinol, dehydroretinol, and carotenoids in adult patients versus healthy controls. Arch Dermatol Res 1985;278:17-24. 7. Siegenthaler G, Gumovski-Sunek D, Saurat JH. Metabolism of natural retinoids in psoriatic epidermis. J Invest Dermatol 1990;95:S47-8. 8. Siegenthaler G, Saurat JH, Hotz R, Camenzind M, Mérot Y. Cellular retinoic acid, but not cellular retinol-binding protein, is elevated in psoriatic plaques. J Invest Dermatol 1986;86:42-5. 9. Didierjean L, Durand B, Saurat JH. Cellular retinoic acid-binding protein type 2 mRNA is overexpressed in human psoriatic skin as shown by in situ hybridization. Biochem Biophys Res Commun 1991;180:204-8.
J AM A C A D D ERMATOL SEPTEMBER 1999
10. Siegenthaler G, Tomatis I, Didierjean L, Jaconi S, Saurat JH. Overexpression of cellular retinoic acid-binding protein type II (CRABP2) and down-regulation of CRABP1 in psoriatic skin. Dermatology 1992;185:251-6. 11. Busch C, Siegenthaler G, Vahlquist A, et al. Expression of cellular retinoid-binding proteins during normal and abnormal epidermal differentiation. J Invest Dermatol 1992;99:795-802. 12. Hirschel-Scholz S, Siegenthaler G, Saurat JH. Ligand-specific and nonspecific in vivo modulation of human epidermal cellular retinoic acid binding protein (CRABP). Eur J Clin Invest 1989;2:220-7. 13. Elder JT, Astrom A, Pettersson U, et al. Differential regulation of retinoic acid receptors and binding proteins in human skin. J Invest Dermatol 1992;98:673-9. 14. Reichrath J, Munssinger T, Kerber A, et al. In situ detection of retinoid-X receptor expression in normal and psoriatic human skin. Br J Dermatol 1995;133:168-75. 15. Törmä H, Rollman O, Vahlquist A. Interferon-γ increases retinoic acid and 3,4-didehydroretinoic acid concentrations in cultured keratinocytes: a clue to the abnormal vitamin A metabolism in psoriatic skin [abstract]? J Invest Dermatol 1998;110:551. 16. Siegenthaler G, Saurat JH. Natural retinoids: metabolism and transport in human epidermal cells. In: Saurat JH, editor. Retinoids: 10 Years On. Basel: Karger; 1991:56-68. 17. Schmitt A, Hauser C, Didierjean L, Mérot Y, Dayer JM, Saurat JH. Systemic administration of etretin increases epidermal interleukin-1 in the rat. Br J Dermatol 1987;116:615-22. 18. Siegenthaler G, Saurat JH. Therapy with synthetic retinoid (Ro 10-1670) etretin increases the cellular retinoic acid-binding protein in nonlesional psoriatic skin. J Invest Dermatol 1986; 87:122-4. 19. LeMotte PK, Keidel S, Apfel CM. Characterization of synthetic retinoids with selectivity for retinoic acid or retinoid X nuclear receptors. Biochim Biophys Acta 1996;1289:298-304. 20. Carlberg C, Saurat JH. Vitamin D-retinoid association: molecular basis and clinical applications. J Invest Dermatol Symp Proc 1996;1:82-6.