Traitement en cours

Veuillez attendre...

Paramétrages

Paramétrages

Aller à Demande

1. WO2003024990 - VARIANTS DU RECEPTEUR D'HORMONE DE LIBERATION DE LA CORTICOTROPINE DE TYPE 1, ET UTILISATIONS CORRESPONDANTES

Note: Texte fondé sur des processus automatiques de reconnaissance optique de caractères. Seule la version PDF a une valeur juridique

[ EN ]

VARIANTS OF CORTICOTROPIN RELEASING HORMONE
RECEPTOR TYPE 1 AND USES THEREOF

BACKGROUND OF THE INVENTION

Federal Funding Legend
This invention was produced in part using funds obtained through grant IBN-049087 from the National Science Foundation. Consequently, the federal government has certain rights in this invention.

Field of the Invention
The present invention relates generally to the field of biochemical endocrinology of corticotropin releasing hormone receptors. More specifically, the present invention relates to th e identification of alternative splicing variants of human and mouse corticotropin releasing hormone receptor- 1.

Description of the Related Art
Corticotropin releasing hormone (CRH, also known a s corticotropin releasing factor or CRF) is the most proximal element of the hypothalamic-pituitary-adrenal (HPA) axis that coordinates the complex array of behavioral, autonomic, endocrine and immune responses to stress. The peptide signal is translated into phenotypic effects through interaction with G protein-coupled, membrane-bound CRH receptors (1). Two subtypes of receptors, Type 1 (CRH-Rl) and Type 2 (CRH-R2), have been characterized in humans (2, 3), rats (4-6), mice (7-9) and Xenopus (10). Most recently a third subtype, CRH-R3, has been identified in catfish (1 1 ).
CRH-Rl is a protein with 98% sequence homology among different mammalian species and approximately 30% homology with receptors for the gut-brain family of neuropeptides (1, 12, 13). The human CRH-Rl gene contains 1 4 exons (14). Four alternatively spliced CRH-Rl transcripts have been identified in humans. These are CRH-Rl α, in which exon 6 is spliced out to generate a 13-exon transcript that produces a 4 1 5 -amino acid protein (2); CRH-Rl β, which contains all 14 exons to produce a 444-amino acid protein (2); a CRH-Rlc isoform, where exons 3 and 6 are spliced out to generate a 12-exon transcript producing a 375-amino acid protein (15); and an CRH-Rld isoform, where exons 6 and 13 are spliced out to produce a 401 -amino acid protein (16).
CRH-Rl isoforms have different affinity for receptor ligands, resulting in differences in coupling of the isoforms to cAMP production signaling. The major ligand-binding determinant of mammalian CRH-Rl has been mapped to its first extracellular domain (17). This domain is encoded .by exons 1.- 4 of CRH-Rl. Exon 3 contains two regions that are critical for high-affinity ligand binding; thus, mutations in this region abolish CRH binding (18). The CRH-Rlc isoform, which lacks exon 3, should therefore have a decreased CRH binding capacity. A 29 amino acid insert corresponding to exon 6 of CRH-Rl β has also b een reported to decrease binding affinity as well as coupling of th e receptor to G proteins (19). A CRH-Rld isoform lacking exon 1 3 has been recently cloned from human myometrium (16). This isoform is poorly coupled to G proteins. Thus, it appears that CRH-Rl is the most efficient receptor isoform in transducing a CRH signal into cAMP-mediated pathways, while other isoforms either have a poor ligand-binding capacity or are poorly coupled to cAMP production. Because a spectrum of receptor isoforms expressed by a cell can determine its response to a ligand, full molecular characterization of CRH-Rl transcripts is necessary in order to understand the pleiotropic role of CRH.
Skin, the largest body organ, maintains internal homeostasis by serving as a barrier between the external environment and the internal milieu. Being continuously exposed to noxious stimuli of varying intensities, including solar radiation, thermal energy and biological agents, the skin requires a highly localized and precise mechanism for dealing with the immediacy of these interactions (20, 21). Analogous to the central response to stress centered on the HPA axis, it was proposed that similar mediators could activate peripheral responses to stress with a CRH-based signaling system playing a major regulatory role (22-24).
Both CRH and urocortin are produced in human an d rodent skin, accompanied by the expression of functional CRH-Rl (21-26). It has been proposed that the flow of information involving cutaneous CRH peptides could be arranged hierarchically, from CRH through CRH-Rl to the activation of POMC peptide production and corresponding activation of the respective receptors for these peptides (22, 24). Alternatively, they could act directly through CRH-Rl activated pathways to regulate epidermal integrity, barrier function, immunomodulation, dermal vascular function, and hair growth and pigmentation (20, 22, 24) . Such functional diversity requires specific molecular mediators, and functional selectivity could be achieved through differential expression of CRH-Rl isoforms.
The prior art is deficient in a full molecular characterization of CRH-Rl isoform expression for understanding the pleiotropic effects of CRH. The present invention fulfills this long-standing need and desire in the art.

SUMMARY OF THE INVENTION

In the present study, the expression of CRH-Rl isoforms was examined in various human and mouse skin samples and cell lines having different physiological and pathologic status, as well as the effects of exposure to UN radiation. The observed expression patterns were compared to pituitary, brain, adrenal and spleen CRH receptors.
Four new isoforms of human CRH-Rl (e-h) and three of mouse (mCRH-Rlc, e and f) were identified. Human CRH-Rle was characterized by the deletion of exons 3 and 4; exon 12 w as deleted from CRH-Rlf; exon 11, 27 bp of exon 10, and 28 bp of exon 12 were deleted from CRH-Rlg; and CRH-Rlh w as characterized by the addition of a cryptic exon. In mouse CRH-Rl c, exon 3 was spliced out; in mCRH-Rlβj exons 3 and 4 were spliced out; and in mCRH-Rlf, exon 11 was spliced from th e mRNA.
CRH-Rl was expressed in all skin specimens in patterns dependent on the cell type, physiological status and presence of pathology. CRH-Rlα, the most prevalent form, w as detected in almost all samples. Ultraviolet radiation (UV) changed the splicing pattern and induced or increased expression of CRH-Rig in cultured skin cells. Continuing UV treatment of succeeding generations of cells resulted in a progressive increase in th e number of CRH-Rl isoforms, suggesting that receptor heterogeneity might favor cell survival. TPA, forskolin, dbcAMP and IBMX also changed the splicing patterns. These data suggest that polymorphism of CRH-Rl expression is related to anatomic location, skin physiological or pathologic status, specific cell type, and external stress (UV); and that cAMP-dependent pathways and TPA may regulate CRH-Rl expression.
In one embodiment of the current invention, a DNA encoding a corticotropin releasing hormone receptor type 1 protein amino acid is provided. This sequence may be selected from the group consisting of: SEQ ID No. 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 31 , 32, 33, 34, 35, 36, and 37. Also provided is DNA encoding the protein selected from the above group, that differs from the above DNA in codon sequence due to the degeneracy of the genetic code.
In another embodiment of the current invention, the instant invention is directed to a vector capable of expressing the DNA.

The instant invention is also directed to a host cell transfected with and expressing a corticotropin releasing hormone type 1 receptor protein from such a vector.
In yet another embodiment of the instant invention, an isolated corticotropin releasing hormone receptor type 1 protein is provided, encoded by the DNA described above. Preferably, the purified protein has an amino acid sequence corresponding to SEQ

ID No: 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 31, 32, 33, 34, 35, 36, or 37.
In another embodiment of the instant invention, an antibody directed against the corticotropin releasing hormone receptor type 1 protein is provided.
In yet another embodiment of the instant invention, a pharmaceutical composition is provided comprising a corticotropin releasing hormone receptor type 1 protein.
Still another embodiment provides a method of treating a pathophysiological state.
The present invention also provides a method of protecting skin cells against damage by inducing the expression of corticotropin releasing hormone receptor type lg in said skin cells, wherein the expression of the corticotropin releasing hormone receptor protects said skin cells against damage induced b y environmental factors, of which solar radiation is an example.
In another embodiment of the present invention, there is provided a method of screening for a compound that induces the expression of corticotropin releasing hormone receptor type I f or lg, comprising the steps of: contacting said compound with skin cells; and determining the expression of the corticotropin releasing hormone receptor in cells that are or are not treated with th e compound, wherein increased expression of the corticotropin releasing hormone receptor in treated cells compared to untreated cells indicates the compound induces expression of the corticotropin releasing hormone receptor type If or lg.
In yet another embodiment of the present invention, there is provided a method of regulating the extracellular concentration of corticotropin releasing hormone or corticotropin releasing hormone related peptides, comprising the step of: administering corticotropin releasing hormone receptor type le or lh to an individual, wherein the receptor regulates extracellular concentration of corticotropin releasing hormone or corticotropin releasing hormone related peptides by binding and slowly releasing the hormone in said individual.
Other and further aspects, features, and advantages of the present invention will be apparent from the following description of the presently preferred embodiments of the invention. These embodiments are given for the purpose of disclosure.

BRIEF DESCRIPTION OF THE DRAWINGS

So that the matter in which the above-recited features, advantages and objects of the invention as well as others which will become clear are attained and can be understood in detail, more particular descriptions and certain embodiments of the invention briefly summarized above are illustrated in the appended drawings. These drawings form a part of the specification. It is to be noted, however, that the appended drawings illustrate preferred embodiments of the invention and therefore are not to be considered limiting in their scope.
Figure 1 shows a UV spectrum characteristic of th e BioRad Transilluminator 2000 (250-500 nm).
Figure 2 shows alternatively-spliced isoforms of

CRH-Rl. Shadowed boxes: translated exons; . open boxes: exons situated after a frame-shift; lines: exons absent in mRNA; dashed lines: homologue of human exon 6, which is not detected in mouse or hamster mRNA. The site of insertion of a cryptic exon in CRH-Rlh is indicated. Arrows indicate positions of primers.
Figure 3(A-D) shows amplification of human CRH-Rl to detect transcripts (exons 2-7; primers PI 12 and P113). Figure 3A shows screening of human tissue samples: lane 2, pituitary; lane 3, adrenal gland; lanes 4 and 5, normal skin samples; lane 6, normal keratinocytes; lanes 7 and 8, skin samples containing basal cell carcinomas; lanes 1 and 9, DNA ladders.
Figure 3B shows expression in the immortalized human keratinocyte cell line HaCat: lane 2, untreated cells (control); lane 3, treated by TPA; lane 4, treated by forskolin; lane 5, treated by IB MX and dbcAMP. Cells in lanes 6-9 w ere irradiated by 50 mJ/cm2 of UVB. Lane 6, cells detached 24 hours after irradiation; lane 7, cells treated by UVB and incubated a t standard conditions for 2 weeks; lane. 8, two successive cycles of treatment by UVB and incubation for 2 weeks; lane 9, three successive cycles of treatment by UVB and incubation for 2 weeks; lanel, DNA ladder. Amplification of GAPDH is shown below the pictures. Arrows indicate sequenced bands.
Figure 3C shows expression in human squamous cell carcinoma C^: lane 2, untreated cells (control); lane 3, treated b y TPA; lane 4, treated by forskolin; lane 5, treated by LBMX and dbcAMP. Cells in lanes 6-8 were irradiated by 50 mJ/cm2 of UVB. Lane 6, cells detached 24 hours after irradiation; lane 7, cells treated by UVB and incubated at standard conditions for 2 weeks ; lane 8, two successive cycles of treatment by UVB and incubation for 2 weeks; lane 1, DNA ladder. Amplification of GAPDH is shown below the pictures. Arrows indicate sequenced bands.
Figure 3D shows expression in human melanoma cell line SKMEL188: lane 2, untreated cells (control); lane 3, treated b y TPA; lane 4, treated by forskolin; lane 5, treated by IBMX and dbcAMP; lane 6, cells treated by UVB (50 mJ/cm2) and incubated at standard conditions for 2 weeks; lane 1, DNA ladder. Amplification of GAPDH is shown below the pictures. Arrows indicate sequenced bands.
Figure 4(A-E) shows amplification of human CRH- Rl(exons 9-14) or hamster CRH-Rl (exons 8-13) to detect transcripts. Figure 4A shows results of screening human tissue samples: lane 2, pituitary; lane 3, adrenal gland; lane 4, normal skin; lane 5, neonatal keratinocytes; lane 6, neonatal melanocytes; lanes 7-9, skin containing basal cell carcinomas; lane 1, DNA ladder. Arrows indicate sequenced bands.
Figure 4B shows expression in the immortalized human keratinocyte HaCaT cell line: lane 2, untreated cells (control); lane 3, treated by TPA; lane 4, treated by forskolin; lane 5, treated by IBMX and dbcAMP. Cells in lanes 6-8 w ere irradiated by 50 mJ/cm2 of UVB. Lane 6, cells detached 24 hours after irradiation; lane 7, cells treated by UVB and incubated a t standard conditions for 2 weeks; lane 8, two successive cycles of treatment by UVB and incubation for 2 weeks; lane 9, three successive cycles of treatment by UVB and incubation for 2 weeks ; lane 1 , DNA ladder. Arrows indicate sequenced bands.
Figure 4C shows expression in the human squamous cell carcinoma cell line C^: lane 1, untreated cells (control); lane 2, treated by TPA; lane 3, treated by forskolin; lane 4, treated b y IBMX and dbcAMP. Cells in lanes 5-8 were irradiated by 5 0 mJ/cm2 of UVB. Lane 5, cells detached 24 hours after irradiation; lane 6, cells treated by UVB and incubated at. standard conditions for 2 weeks; lane 7, two successive cycles of treatment by UVB and incubation for 2 weeks. Arrows indicate sequenced bands.
Figure 4D shows expression in human melanoma cell line SKMEL188: lane 2, untreated cells (control); lane 3, treated b y TPA; lane 4, treated by forskolin; lane 5, treated by LBMX and dbcAMP. Cells in lanes 6-7 were irradiated by 50 mJ/cm2 of UVB. Lane 6, cells detached 24 hours after irradiation; lane 7, cells treated by UVB and incubated at standard conditions for 2 weeks ; lane 1, DNA ladder. Arrows indicate sequenced bands.
Figure 4E shows expression in the hamster melanoma cell line AbCl. Lines 2 and 5, untreated AbCl cells; 3, AbCl cells irradiated by UV (50 mJ per cm2 of UVB); 6, AbCl cells after induction of melanogenesis. Lines 1 and 4, DNA ladder. Arrows indicate sequenced bands. The bands are described in Table 1.
Figure 5(A-E) shows the predicted amino acid sequences of human CRH-Rle (GenBank Accession No. AF369651 ), CRH-Rlf (AF369652), CRH-Rlg (AF369653), and CRH-Rlh (AF374231); mouse CRH-Rlc (AF369654), CRH-Rle (AF369655) and CRH-Rlf (AF369656); and hamster CRH-Rle (AF387669), CRH-Rlf (AF387671), CRH-Rlh (AF387667), CRH-Rlj (AF387668), CRH-Rlk (AF387670), CRH-Rlm (AF387672), and CRH-Rln (AF387673) isoforms. Figure 5A-5C shows human and mouse sequences. Previously sequenced isoforms are shown for comparison: human CRH-Rlα (L23332), CRH-Rlβ (L23333), CRH-Rlc (U16273), CRH-Rld (AF180301), and mouse CRH-Rlα (NM_007762). Arrows indicate the positions of introns. The putative transmembrane domains are indicated by rows of # symbols above the appropriate amino acids. The numbers in th e right-hand column refer to the amino acid number. Underlined are new sequences after the frame shift. Figure 5D-5E shows hamster sequences. Previously sequenced isoforms are shown for comparison: hamster CRH-Rlα (A Y034599), and hamster CRH-Rld (AF416616). Arrows indicate the positions of introns. The putative transmembrane domains are indicated by rows of # symbols below the appropriate amino acid. The numbers in th e right-hand column refer to the amino acid number. Predicted amino acid sequences situated after the frameshift are underlines . Dots represent untranslated sequences.
Figure 6(A-D) shows amplification of mouse an d hamster CRH-Rl to detect transcripts. Figure 6 A shows amplification of the mouse fragment spanning exons 2-6 (primers P158 and P159): lane 2, mouse brain; lane 3, mouse pituitary; lanes 4-6, mouse anagen IV, V and VI skin respectively; lane 7 , mouse spleen. Figure 6B shows amplification of the mouse fragment spanning exons 8-13 (primers P162 and P163): lane 2, mouse brain; lane 3, mouse pituitary; lanes 4-6, mouse anagen IV, V and VI skin respectively; lane 7, mouse spleen; lane 8, mouse melanoma S91 (subline M3). Lane 1 on both pictures represents DNA ladder. Arrows indicate sequenced bands. Figure 6C shows amplification of the hamster fragment spanning exons 2-6: 2, eye;

3, pituitary; 4, heart; 5, skin; 6, melanoma Ma; 7, melanoma MI; 8 , melanoma AbCl. Figure 6D shows amplification of the hamster fragment spanning exons 8-13 (primers P162 and P163):2, eye; 3 , pituitary; 4, heart; 5, spleen; 6, skin; 7, melanoma Ma; 8 , melanoma MI; 9, melanoma AbCl. Lane 1 in Figure 6C and 6D is a DNA ladder; arrows indicate sequenced bands.
Figure 7(A-C) shows that the CRH-Rle and CRH-Rlh isoforms are translated into biologically active proteins that inhibit cAMP induction by CRH-Rlα in COS stimulated with either CRH or urocortin. Figure 7 A shows expression constructs containing different isoforms of CRH-Rl (equivalent to CRF-R1). Open boxes represent exons. Arrows indicate the positions of primers used for assembling the constructs. Hindlll and EcoRI restriction sites are situated in the flanking primers. Isoforms amplified by the flanking primers were cloned in the expression vector. Constructs were named according to the isoforms they contain: pCRFRlα (CRH-Rlα isoform), pCRFRlg, pCRFRlf, pCRFRle, pCRFRlh, pCRFRlh2 (CRH-Rlh with 2 mutations). Figure 7 B shows Western blot analysis of expression of the CRH-Rle and CRH-Rlh isoforms in transiently transfected COS cells. Samples of protein extracts of untransformed COS cells (lane 1) and cells transformed by pCRFRlel-V5 (lane 2) or pCRFRlh-V5 (lane 3 ) were probed with mouse anti-V5 antibody and anti-mouse HRP. Figure 7C shows that coexpression of CRH-Rlα with CRH-Rle or CRH-Rlh inhibits cAMP accumulation mediated by CRH-Rlα, in CRF (I, III) or urocortin (II, IV) stimulated COS cells. COS cells were cotransfected by pCRFRle and pCRFRlα (I and II) or b y pCRFRlh and pCRFRlα (III and IV). PCRFRlα was used in all experiments as a positive control. pcDNA was used as an empty vector.

DETAILED DESCRIPTION OF THE INVENTION

Four isoforms of the human CRH receptor type 1 have been described: CRH-Rlα (lacking exon 6), CRH-Rl β (containing all 14 exons), CRH-Rle (lacking exons 3 and 6) and CRH-Rld (lacking exons 6 and 13). In the mouse, only one isoform equivalent to human CRH-Rlα has been characterized (7).
In the present study four new types of human CRH-Rl mRNA (hCRH-Rle, f, g and h) and three new mouse isoforms homologous to human CRH-Rle, e and f were identified (mCRH-Rlc, e and f). In addition, seven new hamster isoforms were identified (hamCRH-Rle, f, h, j, k, m and n). Isoforms e, f, and h are homologous to the corresponding human isoforms, while j, k, m and n have so far only been identified in the hamster. I n humans, in addition to exon 6, exon 12 was spliced from CRH-Rlf; exons 3 and 4 were spliced from CRH-Rle; exon 11, 27 bp of exon 10 and 28 bp of exon 12 were spliced from CRH-Rl g; and CRH-Rlh had a cryptic exon, i.e., an insertion 110 base pairs between exons 4 and 5 (Figure 2). The mouse and hamster sequences do not contain exon 6. Exon 3 was spliced out in mCRH-Rlc; exons 3 and 4 were spliced out in mCRH-Rle, and exon 11 was missing in mCRH-Rlf. The hamster hamCRH-Rle and f splicing patterns are similar to those of the mouse. (Figure 2). Among the additional hamster isoforms, exon 5 is spliced out of hamCRH-Rlj, exon 10 is spliced out of isoform k, exons 11-12 are spliced out of the m isoform, and exons 10-12 are spliced out of isoform n. The hamster h isoform, similar to the human homologue, contains a sequence inserted between exons 4 and 5.
Alternative splicing is a tightly regulated process, generating different mRNAs and increasing the coding capacity of genes (34-36). Approximately 33-59% of human genes have a t least two variants (37). For example, 576 possible alternative forms of a K+ channel are expressed in a gradient along the 10, 000 sensory-receptor cells present in the inner ear of birds, enabling perception of different sound frequencies (38). Furthermore, 15% of the point mutations that cause diseases in humans alter th e normal splicing pattern of genes(36, 39). Thus, the described spectrum of CRH-Rl isoforms in human and mouse may reflect th e diverse phenotypic functions of CRH and related peptides, requiring precise and selective coupling of signal transduction pathways .
Some information on the potential role(s) of the new isoforms in phenotypic regulation can be obtained from th e analysis of predicted structures of the protein products and the gene expression patterns in different cellular compartments (Figure 5, Table 2). Human and mouse CRH-Rle isoforms of th e CRH-Rl receptor contain two reading frames. One reading frame (CRH-Rlel) encodes a soluble protein of 194 amino acids in humans and 139 amino acids in mice. It contains the first 4 0 amino acids of the N-terminal sequence, the remaining sequence being different from the CRH-Rl receptor due to the frame shift. Because it contains the first 400 amino acids from the N terminus , it can act either as a CRH or CRH-related peptide-binding protein. The second frame (CRH-Rle2) encodes a human protein of 240 amino acids and a mouse protein of 309 amino acids. The beginning of the protein sequence contains the third transmembrane domain in humans, and the first transmembrane domain in the mouse (Figure 5). It will not be able to bind a ligand because it lacks the N terminus of the receptor. Similarly, the newest CRH-R2 isoform detected in the stomach also comprises only the C-terminal part of the CRH-R2 gene (GeneBank accession No. E12750; (40)).
The CRH-Rlh isoform encodes a truncated protein having only the CRH-binding domain (coded by exons - 1 -4) , because the cryptic exon 4 contains a translation terminator (Figure 5). It can potentially interfere with the binding of CRH or serve as an analog of a CRH-binding protein. Of note, tested mouse tissues did not express the CRH-Rlg and CRH-Rlh forms, emphasizing interspecies differences.
CRH-Rle and CRH-Rlh contain sequences leading to premature termination of translation, consequently producing soluble forms of the receptor containing a CRH-binding domain. These proteins can therefore act as regulators of extracellular concentrations of CRH or CRH-related peptides by (i) making them unavailable for interaction with cell surface receptors; (ii) acting as transporters in systemic circulation or as slowly-releasing deposits of bound CRH and CRH-related peptides; or (iii) protecting them from degradation and making them available for phenotypic action in the periphery. Thus, the soluble protein products of CRH-Rle and CRH-Rlh can be injected intravenously to control systemic levels of CRH and CRH-related peptides, or serve as slow-releasing deposits after intravenous or intratissue injection of complexes of receptor-ligand. Since others postulate that CRH in the periphery acts as an immunostimulator, these soluble forms can act as immunomodulators by binding to or storing th e corresponding ligands. In the skin, these forms can be used to treat inflammatory skin diseases such as psoriasis, allergic contact dermatitis and others. These isoforms can also be used to regulate hair growth, because there is differential expression of these genes in the skin in relation to the growing (anagen) or resting (telogen) phases of hair follicles. The presence of the soluble protein product of the CRH-Rlh gene in the pituitary suggests that it can control availability of CRH for regulation of POMC expression and production of ACTH and beta-endorphin, thus acting as a modulator of the systemic stress response.
Human CRH-Rlf encodes the. entire . CRH-binding domain and the first five transmembrane domains; therefore, it should bind CRH and fix it on the outer surface of the cellular membrane. Thus, it may decrease the local concentration of CRH or serve as a pool of bound hormone. The murine form of this receptor also encodes the entire N-terminus and the first five transmembrane domains. The type of signal transduction pathway to which it is coupled remains to be investigated.
The most unusual isoform identified in the present study was CRH-Rlg, in which the reading frame was preserved; but the protein sequence had a deletion of 74 amino acids corresponding to transmembrane domains 5 and 6. This kind of receptor can be potentially coupled to the production 'of cAMP.
Receptor forms CRH-Rlf and CRH-Rlg can serve a s targets for screening the most efficient drugs that can regulate the function of neuroendocrine cells and the phenotypes of skin and immune cells through CRH receptors. Expression of the CRH-Rlf and CRH-Rlg isoforms also suggests that their activation by a selective ligand can inhibit keratinocyte proliferation. Their activity can play a role in hyperproliferative epidermal disorders and in regulation of hair growth, because CRH-Rlf is expressed in anagen (hair follicles in growing phase) but not in telogen skin (hair follicles in resting phase).
Because expression of CRH-Rlg can be induced b y ultraviolet light in skin cells, it is assumed that the activity of this receptor plays a role in protection against damage induced b y solar radiation. Therefore, the specific activation of this receptor by a drug can inhibit epidermal carcinogenesis or malignant transformation of epidermal or dermal melanocytes.
In one embodiment of the current invention, a DNA encoding a corticotropin releasing hormone receptor type 1 protein amino acid is provided. This sequence may be selected from th e group consisting of: SEQ ID No. 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 3 1 , 32, 33, 34, 35, 36, or 37. Also provided is DNA encoding the protein selected from the above group, that differs from the above DNA in codon sequence due to the degeneracy of the genetic code.
In another embodiment of the current invention, the instant invention is directed to a vector capable of expressing the DNA. Such a vector consists of said DNA encoding a corticotropin releasing hormone receptor type 1 protein and regulatory elements necessary for expression in a cell. The instant invention is also directed to a host cell transfected with and expressing a corticotropin releasing hormone type 1 receptor protein from such a vector. The protein may be expressed in a cell type selected from bacterial cells, mammalian cells, plant cells and insect cells.. In one preferred embodiment, the protein is expressed in E. coli.

In yet another embodiment of the instant invention, an isolated corticotropin releasing hormone receptor type 1 protein is provided encoded from DNA as described above. Preferably, the purified protein has an amino acid sequence corresponding to SEQ ID No: 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 31, 32, 33, 34, 35, 36, or 37.
In another embodiment of the instant invention, a n antibody directed against the corticotropin releasing hormone receptor type 1 protein is provided. This antibody may be a monoclonal antibody.
In yet another embodiment of the instant invention, a pharmaceutical composition is provided comprising a corticotropin releasing hormone receptor type 1 protein. Such a pharmaceutical composition may be used to treat a pathophysiological state; in one embodiment, such a state may be a hyperproliferative epidermal disorder, allergic contact dermatitis, autoimmune disorder, epidermal carcinogenesis, or malignant transformation of epidermal or dermal melanocytes.
The present invention also provides a method of protecting skin cells against damage by inducing the expression of corticotropin releasing hormone receptor type lg in said skin cells, wherein the expression of the corticotropin releasing hormone receptor protects said skin cells against damage induced b y environmental factors, of which solar radiation is an example. The damage in this scenario will include mutagenic or carcinogenic effects, or oxidative damage to cellular components that m ay cause an inflammatory or autoimmune response. This may b e achieved by inhibition of cell proliferation that would protect DNA by keeping it longer in the chromatin-bound form or an increase in the controlled death of damaged cells, thus, preventing oncogenesis or induction of autoimmune processes. In one embodiment, the expression of receptor type lg in skin cells regulates production of cAMP.
In another embodiment of the present invention, there is provided a method of screening for a compound that induces the expression of corticotropin releasing hormone receptor type I f or lg, comprising the steps of: contacting said compound with skin cells; and determining the expression of the corticotropin releasing hormone receptor in cells that are or are not treated with the compound, wherein increased expression of the corticotropin releasing hormone receptor in treated cells compared- to untreated cells indicates the compound induces expression of th e corticotropin releasing hormone receptor type If or lg. The expected effects in the skin would include regulation of proliferation or immune functions, or modification of the activity of other isoforms such as CRH-Rlα. In one embodiment, said compound comprises a treatment for a pathophysiological state. In a preferred embodiment, the pathophysiological state is a hyperproliferative epidermal disorder or a neuroendocrine disorder.
In yet another embodiment of the present invention, there is provided a method of regulating the extracellular concentration of corticotropin releasing hormone or corticotropin releasing hormone related peptides, comprising the step of: administering corticotropin releasing hormone receptor type le or lh to an individual, wherein the receptor regulates extracellular concentration of corticotropin releasing hormone or corticotropin releasing hormone related peptides by binding and slowly releasing the hormone in said individual.

In still another embodiment, administering the type le or lh receptor comprises a treatment for a pathophysiological state.

In a preferred embodiment, the pathophysiological state is a n inflammatory skin disease, which may include psoriasis, allergic contact dermatitis, or abnormal hair growth.
Another embodiment provides that the receptor type le or lh may be administered to the individual by injecting said receptor type le or lh intravenously. In an additional embodiment, the administration of the receptor type le or 1 h inhibits production of cAMP in the individual.
In an additional embodiment, said receptor type le or lh comprises a complex between such receptor type and a corticotropin releasing hormone or corticotropin releasing hormone related peptides. One embodiment provides that .such a complex may be administered to the individual by intravenous or intratissue injection.
The following examples are given for the purpose of illustrating various embodiments of the invention and are not meant to limit the present invention in any fashion.

EXAMPLE 1

Tissues And Cell Culture
Human tissue samples included pituitary, adrenal gland, and non-lesional normal skin and pathologic skin containing basal cell carcinoma. Skin and adrenal gland specimens w ere obtained from tissue removed during surgery, while pituitaries were obtained from the National Hormone and Pituitary Program, NIDDK. The tissues were stored at -80 °C until the time of analysis. The University of Tennessee Health Science Center (UTHSC) Committee on Research Involving Human Subjects approved the use of human tissues.
Murine samples consisted of brain, pituitary, spleen, and skin isolated at telogen and anagen IV, V and VI stages of th e hair cycle. Female C57BL/6 mice (8 weeks old) were purchased from Taconic (NY) and housed in community cages at the animal facilities of the Albany Medical College (AMC), Albany, NY. The animals were sacrificed under pentobarbital anesthesia, and selected organs as well as back skin were collected following protocols described previously (23, 27). Tissue specimens w ere frozen rapidly in liquid nitrogen and stored at -80°C until further analysis. The Institutional Animal Care and Use Committee a t AMC originally approved the experimental protocol, and a similar protocol for mice was approved at UTHSC.
The tissues were pulverized in liquid nitrogen with a mortar and then suspended in Trizol (Gibco-BRL, Gaithersburg, MD), and RNA was isolated following the manufacturer's protocol.
Hamster eyes, pituitary, heart, spleen, and skin w ere used for the hamster studies. Syrian hamsters (males 3 months old) were purchased from Taconic (New York) and housed in community cages at the animal facilities of the Albany Medical College (AMC), Albany, NY. The animals were killed under pentobarbital anesthesia and selected organs as well as back skin were collected following protocols routinely used in the laboratory (47). Tissue specimens were frozen rapidly in liquid nitrogen. Hamster Bomirski Ma melanotic, MI . hypomelanotic and . AbCl amelanotic melanomas were propagated in male Syrian hamsters by subcutaneous inoculation of tissue suspension as described previously (48). After killing the animals, tumor tissue was freed from connective and necrotic tissues and frozen rapidly in liquid nitrogen. Hamster tissues as well as melanoma transplants w ere stored at -80°C until further analysis. The experimental protocol was originally approved by the Institutional Animal Care and Use Committee at AMC.
Human and mouse cell lines were cultured according to standard protocols as described previously, and the media w ere changed every second day (28, 29). The C02 concentration was 5% except for mouse normal melanocytes (see below). Human immortalized keratinocytes (HaCaT) and squamous cell carcinoma cells (C^j) were propagated in DMEM medium (GIBCO), while human melanoma (SKMEL188) and mouse Cloudman S91 melanoma (sublines #6 and M3) cells were grown in Ham's F10 medium as described previously. The media were supplemented with 10% fetal bovine serum and antibiotics (GIBCO) (28, 29) . Additional human melanoma cells included those established from the radial growth phase (WM 35. .and SBCE2), vertical growth phase (WM 98 and WM 1341D) and metastatic phase (WM 164) (gift of Dr. M. Herlyn, Wistar Institute, Philadelphia, PA). These cells were cultured in DMEM supplemented with 10% fetal bovine serum, insulin (5 μg/ml) and antibiotics.
Immortalized normal mouse skin melanocyte line

MelA (from Dr. D. Bennett, Saint George Hospital, London, England) was cultured in RPMI 1640 medium supplemented with 10% bovine serum and 200 nM TPA (phorbol 12-myristate 13-acetate) in the presence of 10% C02. Normal human neonatal keratinocytes from passages 2 and 3 were used for experiments (25). Primary cell cultures were established from foreskin as described previously (25). The cells were propagated in low-calcium (0.15 mM) serum-free Keratinocyte Growth Medium (KGM) containing bovine pituitary extract (BPE) and antibiotics (Clonetics Corp., San Diego, CA). Normal human neonatal melanocytes were cultured in medium 154 (Cascade Biologicals, Portland, OR) supplemented with 5% FBS, 13 μg/ml BPE, and 8 nM TPA, 1 μg/ml α-tocopherol, 0.6 ng/ml basic fibroblast growth factor, 1 μg/ml transferrin and 5 μg/ml insulin (all from Sigma)(30). After washing with PBS, melanoma cells were detached with Tyrode's solution containing 1 mM EDTA (28), while keratinocytes and normal melanocytes w ere trypsinized (26, 29). The cells were centrifuged and suspended in RNA isolation solution (Trizol reagent).
Bomirski AbC-1 hamster melanoma cells were grown in Ham's F10 medium as described previously; the media were supplemented with 10% fetal bovine serum and antibiotics (Gibco BRL, Gaithersburg, MD) (49). To induce melanogenesis the cells were cultured for 3 days in Dulbecco's minimal Eagle's medium plus 10% fetal bovine serum. Melanoma cells were detached with Tyrode's solution containing lmM ethylenediamine tetraacetic acid after prior washing with PBS (49). The cells were centrifuged at 4°C, washed with cold PBS and cell pellets were used for RNA isolation.
In some experiments, SKMEL188, HaCaT and C4_. cells were also treated with TPA, forskolin or a mixture of IBMX (3 -isobutyl-1-methylxanthine) and dbcAMP (N6, 2'-0-dibutyryladenosine 3' :5'-cyclic monophospate sodium) (all from Sigma). Briefly, the cells were transferred to 75 cm2 flasks ( 106 cells/flask) and cultured for 24 hours in standard culture medium. Then the following compounds were added to the separate cultures: TPA (200nM), forskolin ( l O"5 M), LBMX (5x l 0"4 M), and dbcAMP ( 10"3 M). Controls were represented by untreated cultures. The cells were incubated for 24 hours, detached, collected by centrifugation, and dissolved in RNA isolation solution (Trizol reagent).
COS cells were propagated in DMEM medium (GIBCO) supplemented with 10% Fetal Bovine serum and antibiotics (GIBCO). Ten thousand cells were routinely transfected in each well of opaque 96-well plate (Packard) by Lipophectamine (Invitrogen,

Carlstand, CA) according to the manufacturer's protocol.

EXAMPLE 2

Irradiation Of Cell Lines By UN
SKMEL188, HaCaT, C^ and hamster AbCl cells were exposed to ultraviolet radiation produced by a UV transilluminator 2000 (BioRad). Cells were transferred to 9 cm Petri dishes at a concentration 106 cells/dish and grown for 24 hours under standard conditions as described above. Before irradiation, the medium was aspirated and replaced by 10 ml of PBS. The dishes were placed on the UV transilluminator and incubated for 3, 12 or 31 seconds, corresponding to 5, 20 or 5 0 mJ/cm2 doses of UVB respectively. Times of exposure and th e corresponding UV doses were calculated by the standard formula: Time (s) = Dose (J/cm2)/ Intensity (W/cm2) The spectrum of UV irradiation of the transilluminator

2000 was measured using an Optronic spectroradiometer, model

754 (Figure 1). The calculated intensity of UVB was 1.58xl 0'3

W/crn2, that of UV A was 1.16xl0"3 W/cm\ and that of UVC w as 2.26x 10-5 W/cm2.
After irradiation, the PBS was replaced by standard culture medium. Cells were incubated for 24 hours, detached, collected by centrifugation and dissolved in RNA isolation solution (Trizol reagent). Alternatively, cells were cultured in standard medium for 2 weeks until their full recovery. At this point cells were collected for RNA isolation or irradiated by an additional UV dose (50 mJ/cm2 of UVB). The irradiated cells were then incubated for two additional weeks in standard medium until full recovery and then irradiated again. The process of UV irradiation was repeated three times.

EXAMPLE 3

RNA Extraction. cDNA Preparation And Polymerase Chain Reaction
Total RNA was extracted using a Trizol isolation kit

(Gibco-BRL, Gaithersburg, MD). The synthesis of first-strand cDNA was performed using the Superscript preamplification system (Gibco-BRL). Five μg of total RNA per reaction was reverse transcribed using oligo(dT) as the primer.
Nested PCR was used to detect different CRH-Rl isoforms. The first round of amplification of the human CRH-Rl fragment spanning exons 2-7 was conducted using 2 μl of cDNA and primers PI 10: 5'-TCCGTCTCGTCAAGGCCCTTC-3' (sense) (SEQ ID No. 15) and Pi l l : 5'-GGCTCATGGTTAGCTGGACCAC-3' (antisense) (SEQ LD No. 16). An aliquot of the PCR mixture from the first round of amplification was transferred to a new tube, and a second round of PCR was conducted. Primers for the second round of PCR were PI 12: 5'-TGTCCCTGGCCAGCAACATCTC-3' (sense) (SEQ ID No. 17) and PI 13: 5'-AGTGGATGATGTTTCGCAGGCAC-3' (antisense) (SEQ ID No. 18).
Amplification of exons 9 through 14 of human CRH-Rl was done in the same way. Primers for the first round of PCR were PI 14: 5'-CCATTGGGAAGCTGTACTACGAC-3' (sense) (SEQ ID No. 19) and PI 15: 5'-GCTTGATGCTGTGAAAGCTGACAC-3' (antisense) (SEQ ID No. 20). Primers for the second round of PCR were PI 16: 5'-GGGTGTACACCGACTACATCTAC-3' (sense) (SEQ ID No. 21) and PI 17: 5'TCTTCCGGATGGCAGAACGGAC-3' (antisense) (SEQ ID No. 22).
Primers for the first round of amplification of th e mouse CRH-Rl fragment spanning exons 2-6 were P156: 5 ' -TCCGGCTCGTGAAGGCCCTTC-3' (sense) (SEQ LD No. 23) and P157: 5 ' -GCTCAGGGTGAGCTGGACCAC-3' (antisense) (SEQ ID No. 24) . Primers for the second round of PCR were P158: 5 ' -TGTCCCTGGCCAGCAATGTCTC-3' (sense) (SEQ ID No. 25) and PI 59 : 5'-AGTGGATGATGTTCCTCAGGCAC-3' (antisense) (SEQ ID No. 26).
Primers for the first round of amplification of th e mouse CRH-Rl fragment spanning exons 8-13 were P160: 5 ' -CCATTGGGAAACTTTACTACGAC-3' (sense) (SEQ ID No. 27) an d P161 : 5'-CTTGATGCTGTGGAAGCTGACTC-3' (antisense) (SEQ ID No. 28). Primers for the second round of PCR were P162: 5 ' -AAAAGTGCTGGTTTGGCAAACGTC-3' (sense) (SEQ ID No. 29) and PI 63: 5'-CTTCCGGATGGCAGAGCGGAC-3' (antisense) (SEQ LD No. 30).

Primers for the first- and second rounds of amplification of exons 2-6 and exons 8-13 of hamster CRH-Rl isoforms were the same as used for the mouse.
All samples were standardized for the analysis by th e amplification of housekeeping gene glyceraldehyde phosphate dehydrogenase (GAPDH). Primers for the GAPDH gene were a s described by Robbins and McKinney (31). GAPDH gene expression was tested in all samples to assure the integrity of isolated RNA.

Integrated DNA Technology, Inc. synthesized all primers.
The reaction mixture (25 μl) contained 2.5 mM MgCl2,

0.25 of each dNTP, 0.4 μM of each primer, 75 mM Tris-HCl (pH

8.8), 20 mM (NH4)2S04, 0.01% Tween 20 and 0.25 μl of Taq polymerase (Promega). The mixture was heated to 94°C for 2.5 minutes and then amplified for 35 cycles: 94°C for 30 seconds (denaturation), 65°C for 45 seconds (annealing) and 72°C for 1 minute (extension).
GAPDH amplification products were separated b y agarose gel electrophoresis and visualized by ethidium bromide staining according to standard protocol (23).

EXAMPLE 4

Sequencing
The identified PCR products were electrophoresed in an agarose gel, then excised from the gel and purified with a GFX PCR DNA and gel band purification kit (Amersham-Pharmacia-Biotech). PCR products were sequenced from both ends. Sequencing was performed in the Molecular Resource Center a t the University of Tennessee HSC (Memphis) using an Applied Biosystems 3100 Genetic Analyzer and BigDye™ Terminator Kit. The sequence data have been submitted to the GeneBank database under accession numbers AF369651, AF369652, AF369653 , AF374231 , AF369654, AF369655, AF369656, . AF3.87667, AF387668, AF387669, AF387670, AF387671 , AF387672, and AF387673.

EXAMPLE 5

Alternate Splicing Variants Of Human Corticotropin Releasing Hormone Receptor Type 1
Two sets of nested primers were designed to amplify the regions of human CRH-Rl mRNA spanning exons 2 through 7 and 9 through 14 (Figure 2). These regions contain exons 3, 6 and 13, which are more likely to be spliced out from human CRH-Rl mRNA. CRH-Rl mRNA expression was detected in all human tissues and cell lines tested including pituitary, adrenal gland, skin, normal neonatal melanocytes and keratinocytes, immortalized HaCaT keratinocytes, squamous cell carcinoma C4A and melanoma cell lines (Figures 3A and 4A). The visualized amplification products were cut from the gels and sequenced.
The characteristics of detected isoforms are presented in Table 1. In addition to the previously described CRH-Rlα, c and d isoforms, the present invention identified 4 new isoforms named CRH-Rle (AF369651), CRH-Rlf (AF369652), CRH-Rlg (AF369653) and CRH-Rlh (AF374231). The splicing pattern is presented in Figure 2 and Table 1. All of these isoforms have exon 6 spliced out from the final transcript. Furthermore, in CRH-Rle exons 3 and 4 are spliced out (Figure 3A, 163 bp fragment) .

CRH-Rlf has a deletion of exon 12 (Figure 4 A and B, 200 b p fragment). In CRH-Rlg exon 11, 27 bp of exon 10 and 28 bp of exon 12 are deleted from the mRNA transcript (Figure 2 and 4, 114 bp band; Table 1). CRH-Rlh contains an insertion of a cryptic exon (110 bp) between exons 4 and 5 (Figure 4B, Table 1).
Predicted protein sequences of these new isoforms in comparison to the CRH-Rl variants α, β, c and d are presented in Figure 5. CRH-Rle and CRH-Rlf have frame shifts (Figure 2). CRH-Rle mRNA has two potential reading frames of 585 bp and 723 bp (Figure 2, 5). The first one encodes a 194-amino acid protein (CRH-Rlel) containing only the first 40 amino acids from the N-terminus encoded by exons 1 and 2, the remaining (Figure 5, CRH-Rlel, underlined) amino acid sequence is different from that encoded by other isoforms and does not contain transmembrane binding domains (Figure 2, 5). The second reading frame can potentially encode a membrane-bound protein of 240 amino acids (CRH-Rle2) with a sequence starting from th e third transmembrane domain and containing the C terminus (Figure 5). The reading frame for CRH-Rlf is 1113- bp long and encodes a receptor protein of 370 amino acids containing the entire CRH-binding domain and the first five transmembrane domains (215 amino acids); the remaining C terminal sequence (underlined) is different from other forms due to a frame shift (Figure 5).
The most unusual isoform was CRH-Rlg, which has preserved the reading frame of the CRH receptor. It encodes a membrane-bound protein of 341 amino acids that has a deletion of 74 amino acids corresponding to transmembrane domains 5 and 6 (Figure 5). The insertion of a 110 bp cryptic exon between exons 4 and 5 in the CRH-Rlh isoform would generate a truncated protein of 145 amino acids having only a CRH-binding domain (coded by exons 1-4), because the inserted exon contains a translation terminator (AF374231).

TABLE 1
Characteristics Of CRH-Rl Isoforms Detected Bv Nested RT-PCR



EXAMPLE 6

Expression of Alternate Splicing Variants Of Human Corticotropin Releasing Hormone Receptor Type 1
The expression pattern of CRH-Rl isoforms in tested tissues and cell lines is summarized in Table 2. CRH-Rlα was the most widely expressed isoform (the 369 bp fragment in Figure 3A ; 336 bp in Figure 4 A), being detectable in pituitary, adrenal gland and in all cell lines as well as skin samples, with exception of one skin biopsy containing basal cell carcinoma. The CRH-Rle isoform was expressed only in the skin containing basal cell carcinoma (Figure 3 A , lanes 7 and 8, 249 bp fragment). The CRH-Rld isoform was detected in the pituitary, neonatal normal keratinocytes and in five melanoma lines (the 294 bp fragment in Figure 4A, lanes 2 and 5, Table2). CRH-Rle transcripts w ere detected in skin containing basal cell carcinoma ' and in four melanoma lines, as well as hamster pituitary (Table 2). CRH-Rlf transcripts were detected in one skin biopsy with basal cell carcinoma, in normal neonatal keratinocytes, a squamous cell carcinoma line and three melanoma lines, and also in hamster heart, skin, spleen, melanomas Ma and MI, and melanoma line AbCl (Table 2). Among the newly characterized isoforms, th e most widely distributed isoform was CRH-Rlg that was detected in pituitary, adrenals, normal and pathologic skin, neonatal keratinocytes and five melanoma lines (Table 2). Hamster isoforms CRH-Rl j, k, and n were detected in hamster pituitary (j), eye and skin (k), or eye (m). Hamster isoform m was found in melanized but not UVB -irradiated AbCl cells (Table 2). The only isoform that could not be detected in the tested samples was CRH-Rlβ.

TABLE 2
Expression Of CRH-Rl In Different Tissues And Cell Lines



To test the hypothesis that environmental stress can change cutaneous expression of CRH-Rl in human skin cells, as was shown in Table 2 above for hamster AbCl cells, human HaCaT immortalized keratinocytes, SKMEL188 melanoma and C^ squamous cell carcinoma cells were exposed to UV radiation. As shown in Figures 3 - 4 and Table 3, ultraviolet irradiation changed significantly the spectrum of CRH-Rl isoforms detected. In SKMEL188, UV switched expression from the CRH-Rld isoform to CRH-Rlα and CRH-Rlg (Figure 4D, Table 3). In HaCaT keratinocytes, a first exposure to UV increased only expression of CRH-Rlα without changing the isoform pattern (compare lanes 2 and 6 on Figure 4B; Table 3). In C^ cells, UV inhibited expression of CRH-Rlf and induced the expression of CRH-Rlg (Figure 4C, lanes 1 and 5, Table 3) .
Since approximately 50% of cells exposed to UV die within 3 days after treatment, cells that survived such treatment were investigated in order to establish whether a new pattern of CRH-Rl splicing is maintained in the succeeding generations. Such a pattern could represent a factor affecting the survival of cells under stressful conditions. To study these questions, CRH-Rl splicing patterns in cells that started rapid growth 2-3 weeks after UV irradiation were investigated (UV plus incubation, Table 3). Alternatively, these cultures were irradiated by a new dose of UV and again incubated for 2-3 weeks under standard conditions. It was found that the CRH-Rl splicing did not return to th e original pattern (Table 3, Figure 3B-D, Figure 4B-D). Thus, cultured human melanoma cells preserved UV-induced CRH-Rlα and CRH-Rlg expression (Figure 4D, Table 3). However, expression of the CRH-Rlα isoform increased in cells cultured for 2 weeks after irradiation (compare lanes 2 and 6 in Figure 3D and lanes 2 and 7 in Figure 4D).
HaCaT keratinocytes gained expression of the CRH-Rle and CRH-Rlg isoforms after a second UV treatment, and CRH-Rle after a third UV treatment (Figures 3B and 4B, Table 2). C^ cells cultured for two weeks after second UV treatments expressed the CRH-Rle, CRH-Rle and CRH-Rlg isoforms in addition to CRH-Rlα (Figure 3C, lane 8; Figure 4C, lane 7). Thus, repeated exposure of the epithelial cells (HaCaT and CM cells) to UV increased the number of CRH-Rl isoforms expressed (Table 3).
To test whether cAMP-dependent and • TPA-induced pathways can change CRH receptor expression, cell lines w ere incubated in the presence of TPA, forskolin or a mixture of IBMX and dbcAMP. Differential and cell-specific splicing patterns w ere observed. In human melanoma cells, TPA shifted the CRH-Rl splicing pattern from the CRH-Rld to the CRH-Rlα isoform (Figure 4D, Table 3). Forskolin or dbcAMP plus IBMX inhibited expression of CRH-Rld and stimulated expression of CRH-Rlα and CRH-Rlg; the pattern was identical to that induced by UV (Table 3, Figure 4D). In the C4_! cell line, all of these compounds switched off CRH-Rlf isoform expression and induced CRH-Rlg; the pattern was again identical to that induced by UV (Table 3, Figure 4C).
In HaCaT keratinocytes, TPA induced insertion of a 110 bp fragment between exons 4 and 5 of CRH-Rl (Figure 3B, lane 3, 479 bp fragment) that led to a premature translation termination due to the presence of a termination codon in th e inserted sequence. This isoform was named CRH-Rlh. In these cells forskolin had no effect on splicing, while dbcAMP plus IBMX induced expression of CRH-Rle and CRH-Rlg. In summary, forskolin and dbcAMP plus IBMX, but not TPA, stimulated CRH-Rlg isoform expression in all cell lines tested (with the exception of HaCaT cells in the case of forskolin), and also increased the level of CRH-Rlα expression (compare lanes 2-5 in Figs. 3C, 4B and 4D).

TABLE 3
Environmental Regulation Of CRH-Rl Expression In Human Skin Cells


EXAMPLE 7

Alternate Splicing Variants Of Mouse and Hamster Corticotropin Releasing Hormone Receptor Type 1
Only one CRH-Rl isoform has been previously described in mice, i.e., an analog of human CRH-Rlα (7). To test CRH-Rl expression patterns in the mouse, different mouse tissue samples were screened (Table 2) with a set of specific primers, in which exonal locations are listed in Figure 2. The exonal allocation is based on the rodent (rat) gene structure that is similar to that in humans, except that it does not contain exon 6 (14, 32). The detected isoforms have been marked with the letter "m" to emphasize their murine origin, as counterparts of the human receptor form.
Amplification of mouse cDNAs showed that mRNA of mCRH-Rl α was expressed in mouse brain, pituitary, spleen, mouse anagen IV, V and VI skin, normal melanocytes and Cloudman S91 melanoma cells (Figure 6(A-B) and Table 2). The α isoform was absent in resting (telogen) skin. In addition to mCRH-Rlα, three new isoforms were detected (Figure 2, Table 1). One of them is analogous to the human CRH-Rle isoform (AF369654), lacking part of the CRH-binding domain due to the absence of exon 3 (Table 1). This isoform encodes a protein of 375 amino acids, and analogous to the human counterpart (17, 18, 33) would h ave a decreased affinity to CRH. It is expressed only in anagen (IV-VI) skin and spleen (Figure 6 (A-B) and Table 2). The other two isoforms are homologues of human CRH-Rle and CRH-Rlf. These isoforms also have deletions of exons 3 and 4 (AF369655) or 11 (AF369656), respectively; these deletions lead to frame shifts and consequent changes in amino acid sequences of th e receptor proteins (Figure 5A-5C). Mouse CRH-Rle mRNA also has two potential reading frames of 420 bp and 930 bp (Figure 5A-5C). The first one (mCRH-Rlel) encodes a protein of 1 39 amino acids that is similar to the human counterpart; it contains only first 40 amino acids from the N-terminus of CRH-Rl and lacks transmembrane domains due to a frame shift (Figure 5 A -5C, underlined). The second reading frame (mCRH-Rle2) can potentially encode a membrane-bound protein of 309 amino acids with a sequence starting from the first transmembrane domain and containing the C terminus (Figure 5A-5C). The reading frame for CRH-Rlf of 990 bp encodes a receptor protein of 329 amino acids containing the entire CRH-binding domain, seven transmembrane domains and a proximal part of the C terminus ; the distal part of the C terminus is missing due to the absence of exon 12 (Figures 2 and 6(A-B)). mCRH-Rle is expressed in the brain, pituitary, and telogen and anagen skin, while mCRH-Rlf is expressed in anagen skin and in the M3 subline of Cloudman S91 melanoma cells (Table 2) .
To test CRH-Rl expression patterns in the hamster, the regions of the hamster CRH-Rl mRNA spanning exons 2-6 and 8 -13 were amplified from various tissues, two melanoma types propagated in hamsters, and the AbCl melanoma cell line (Table 2). CRF-Rl expression was detected in all hamster tissues tested, including pituitary, eye, skin, spleen, and melanoma lines (Figure 6 (C-D)). The visualized amplification products were cut from a n agarose gel and sequenced. The splicing patterns are- presented in Figure 2 and Table 1. Apart from CRF-Rla, CRF-Rle and f isoforms were also detected, and the h isoform similar to that found in humans. mRNA corresponding to the hamster, mouse, and human CRH-Rle isoform does not have exons 3 and 4 (Figure 2). The reading frame for the hamster e isoform is 420 base pairs long and contains only the first two in-frame exons of the original receptor. It can be translated into a 129-amino acid peptide (Figure 5D-5E). There is another potential reading frame containing seven transmembrane domains, but the biologic role of this protein is not clear. CRH-Rle was expressed only in the hamster pituitary gland (Table 2) .
The hamster CRH-Rlf isoform is also similar to th e murine form in that it does not have exon 11 and also contains a 963-base pair long reading frame potentially translating into a 320- amino acid protein. This protein has the CRH-binding domain and also contains the first five transmembrane domains; thus, it can potentially bind CRH. Hamster CRH-Rlf was expressed in skin, melanomas, heart, and spleen but not in the pituitary (Table 2) .
The hamster CRH-Rlh isoform was also similar to the isoform that was found in humans. Hamster CRH-Rlh mRNA has a 148-base pair insertion, representing a cryptic exon between exons 4 and 5 that should lead to the production of a truncated protein (114 amino acids) due to the presence of several terminator codons (Figures 2 and 5D-5E). The human isoform has a 110-base pair insertion. For comparison, insertions of a cryptic exon have been found in other G protein-coupled receptors, such as the serotonin 2 A receptor (50). Both hamster and human CRF-Rlh isoforms contain CRF-binding domains only. Thus this isoform may represent a soluble protein with the binding activity for CRH-related peptides.

Of great interest is the finding of new types of CRF-Rl mRNA in the hamster. CRH-Rlj has a deletion of exon 5 (Figure 6C; 276 base pair fragment), and its reading frame is 519 base pairs long, coding for a 172-amino acid peptide comprising only the CRH-binding domain (Figure 5D-5E). Thus this isoform can encode a soluble CRH-Rl isoform with properties similar to CRH-Rlh.
Exon 10 is deleted from CRH-Rlk mRNA (Figure 2; 280 base pair fragment); the reading frame is 1020 base pairs long, and it can encode a 339-amino acid peptide containing a CRH-binding domain and two transmembrane domains, lacking the C terminus of the receptor (Figure 5D-5E). If this isoform w as expressed it should be able to bind CRH and fix it on th e membrane surface. mRNA corresponding to this isoform in hamster eye and skin was found (Table 2) .
The CRH-Rln isoform mRNA preserves the original reading frame of the CRH receptor, although it lacks exons 10- 12. This deletion does not cause a reading frameshift. The resulting protein has 327 amino acids, with complete absence of transmembrane domains 6 and 7. Nine amino acids of the fifth domain are also deleted, but the C terminus is probably inside of the cell as there are 15 amino acids of the fifth hydrophobic domain left. This composition is .reminiscent of . the CRF-Rlg isoform found in humans. Human mRNA corresponding to CRH-Fig contains a deletion of exon 11, 27 base pairs of exon 10, and 28 base pairs of exon 12; it does not have transmembrane domains 5 and 6 but is should preserve the intracellular localization of the C terminus, as the last hydrophobic domain is present. Hamster CRF-Rln and human CRF-Rlg isoforms might have properties different from other isoforms. First although their mRNAs have extensive deletions, they preserve an original reading frame and C terminus. Second, the C terminus should b e situated inside the cell, allowing receptor coupling to a signal transduction pathway. On the other side, they also differ from CRF-Rlc and d isoforms by having extensive in-frame deletions of either fifth and sixth or sixth and seventh transmembrane domains.
The expression patterns of CRH-Rl isoforms in tested tissues and cell lines is summarized in Table 2. CRF-Rlα w as expressed in all samples tested (369 base pair fragment in Figure 6C and 363 base pairs in Figure 6D). CRH-Rlf was also widely expressed, being detected in hamster heart, skin, spleen, and melanomas (Figure 6D, 226 base pairs; Table 2). CRH-Rle w as detected only in the pituitary (Figure 6C, lane 3, 163 base pair fragment) and CRH-Rln only in the hamster eye (Figure 6D, lane 2, 98 base pair fragment). Isoform k was expressed in hamster eye and skin, and the j isoform only in the eye (Table 2) .
Ultraviolet light is known to induce melanin synthesis in skin. Melanin synthesis is a multistep process of transformation of L-tyrosine to a melanin biopolymer that includes free radical formation and can potentially generate intracellular oxidative stress (51). The hamster AbCl cell line w as tested, which is amelanotic when cultured in Ham's F10 medium and produces melanin pigment when propagated in Dulbecco's minimal Eagle's medium (49). It was found that , induction of melanin synthesis (Figure 4E) changed the CRH-Rl splicing pattern. Thus, a 184 base pair band appeared that corresponded to an isoform, which was named CRH-Rlm (Table 1). mRNA corresponding to this isoform has a deletion of exons 11 and 12, resulting in a reading frame 1071 base pairs long, and encoding a protein of 356 amino acids. As the receptor reading frame is shifted after exon 10, CRH-Rlm resembles the k and f isoforms, with a CRH-binding domain and the first two transmembrane domains. UV light also changed the pattern of CRH-Rl alternative splicing but differently from the effect of melanogenesis. UV light induced expression of an additional isoform that was analogous to human CRH-Rld (Figure 4E). This form did not have the exon 1 2 comprising the largest part of the seventh transmembrane domain (Figure 2), and it was poorly coupled to cAMP production ( 16) . As in other models, the quantity and ratios between splicing factors have a pronounced impact on splice site selection (34); b y analogy, UV irradiation and melanin synthesis may act b y changing splicing factor availability or activity. Although the mechanism involved in this process is unknown, the different spliced patterns evidence that different regulatory pathways activated either by UV or melanogenesis become operational.

EXAMPLE 8

Regulation And Functions Of Alternate Splicing Variants Of Corticotropin Releasing Hormone Receptor Type 1
Among the already described isoforms, CRH-Rlα is the most efficient receptor variant in transducing a peptide signal into cAMP-mediated pathways. Other forms (β, c, and d) either have a decreased ligand binding capacity (β, c) or are poorly coupled to cAMP production (β, d) (16, 18, 19). CRH-Rlα was the. mo st prevalent form detected in almost all samples tested, with th e exception of one human melanoma cell line (SKMEL188), one skin biopsy containing basal cell carcinoma, and the mouse telogen skin. The dominant role of CRH-Rlα in the skin is further emphasized by the induction of expression in human melanoma cells and keratinocytes by UV and other factors raising cAMP. There was also hair cycle-dependent expression in murine skin (present in anagen and absent in telogen). Of note, pigmentary, metabolic, endocrine and immune, activities of . mouse skin fluctuate during the hair cycle, being low in telogen (resting phase) and high in anagen (growing phase) (20, 21, 41). The second most frequent form detected in human tissues was CRH-Rlg, which can be potentially coupled to cAMP production. Therefore, data in the present study suggest that the main pathway activated by CRH (or related peptides) in the skin involves increased production of cAMP.
CRH-Rl β was not detected in samples derived from human corporal skin, suggesting lack of expression in the tested material. This is in agreement with a previous study showing th e absence of CRH-Rl β in corporal skin, cultured melanocytes and keratinocytes, and its restricted expression in the scalp (24) . Alternatively, the negative results may be due to the preferential amplification of shorter PCR fragments. However, most of the PCR reactions produced several bands of different lengths, implying that one isoform did not completely dominate the amplification reaction. The CRH-Rlα isoform, which is only 87 bp shorter, w as detected in almost all tested samples. Thus, even if the CRH-Rl β isoform is present in the corporal skin, the level of its expression is below the limit of detection of the PCR method.

Accepting that CRH signaling plays a central role in response to stress (20-22), it would be expected that in response to environmental stressors, the expression pattern of CRH-Rl would be changed to counteract the damaging effects of external or internal insults. The experiments with the ultraviolet irradiation that changed the pattern of receptor splicing in skin-derived cell lines support this concept. Thus, CRH-Rl mRNA splicing was changed from the d to the α and g isoforms in th e human melanoma cell line; CRH-Rlα and g also increased in UV-treated immortalized and malignant keratinocytes. Again, this pattern suggests that cutaneous stress stimulates the expression of isoforms that are or can be coupled to cAMP production. It is also significant that the newly gained pattern of CRH-Rl splicing appears to be stable, e.g., it does not regress even after prolonged cultivation (more that 2 weeks in culture). Although it is unclear how this splicing pattern is preserved, it suggests that this n e w pattern somehow promotes the survival of cells damaged b y radiation. Repeated treatments by UV led to an increase of CRH-Rl isoforms expressed in normal and malignant keratinocyte lines, with the resulting populations expressing the CRH-Rlc, e, and f and g isoforms in addition the initial CRH-Rlα.
Thus, its appears that repeated stress favors th e survival of cells having a diverse spectrum of CRH receptor isoforms, probably reflecting the induced cellular heterogeneity of these lines. By analogy with tumor biology (42), such heterogeneity could play a role in stabilizing the phenotype of th e cell line, making it resistant to external manipulation. Pawelek e t al. (43, 44) proposed that the melanocyte response to solar radiation is highly regulated, involving UV-stimulated expression, activation of MSH receptors and increased production of their ligands, e.g., POMC-derived MSH and ACTH. It was noted that UV induced CRH-Rl expression with preference for the most efficient α isoform, and production of the respective CRH ligand (45). Thus, this general molecular mechanism of UV action on epidermal cells (43, 44) may be conserved and would involve stimulation of CRH and POMC peptide production accompanied by induction and modification of the corresponding receptors (20, 45, 46).
To better understand the mechanism of differential CRH-Rl splicing, the effects of factors raising intracellular cAMP and of TPA were examined. Factors raising intracellular cAMP increased CRH-Rl expression and switched the pattern to predominant expression of the α and g isoforms. This pattern was similar or identical to that induced by ultraviolet radiation, suggesting that similar mechanisms regulate CRH-Rl expression that are tightly linked to a cAMP-activated signaling pathway(s). TPA also switched receptor splicing; however, the pattern of expression levels were different from those induced by UV or cAMP-dependent signals. Thus, it is suggested that cutaneous CRH-Rl gene expression can be regulated by at least two different signaling systems: one activated by UV and cAMP, and the second by TPA.
In summary, the present study finds that CRH-Rl is differentially spliced in a variety of human and mouse tissues. New isoforms of the receptors are identified, and a pattern of environmental regulation in cultured skin cells is found. I n conclusion, a polymorphism of CRH-rRl expression appears . to b e related to anatomic location, skin physiological and pathologic status and cell type. In addition, external stress (UV), cAMP dependent pathways and TPA can also regulate CRH-Rl expression in skin cells.

EXAMPLE 9

cAMP Accumulation in CRF or Urocortin-Stimulated COS Cells
Expression constructs were prepared that contain different isoforms of CRH-Rl (Figure 7A). The alpha isoform was amplified from the phCRF-R82 plasmid by primers E3 and El l.
Full-length CRF-Rlg DNA was obtained from three PCR reactions. First, a fragment spanning the 5' untranslated sequence and exons 1 through 10 was amplified using primers E3 and E9. A second fragment (exons 12-14) was amplified using primers E12 and El l. Finally, the first and second fragments were assembled together using primers E3 and El l; this was possible because primer E9 contained a sequence homologous to primer El 2.
Similarly, for the CRF-Rlf construct exons 1 through 11 of the

CRF receptor were amplified using primers E3 and E18, and exons

13 and 14 by using primers E16 and E17. The full- sequence w as obtained by combining these two fragments together using primers

E3 and El 7.
CRF-Rle DNA was constructed in a slightly more complicated way. Fragments spanning exons 1-2 and 5-14 were amplified using primer pairs E3, E24 and E25, El l respectively. The first fragment was slightly extended in nested PCR by primers E3 and E28. Finally, full-length CRF-Rle DNA was assembled by amplifying these two fragments with primers E3 and El l.
The CRF-Rlh isoform contained exons 1-4 and a fragment of the cryptic exon up to the translation terminator. This construct was also assembled in 3 steps. In the first step, exons 1 through 4 were amplified with primers E3 and E 24; in the second step, the cryptic exon was amplified with primers El 2 and El 9. In the third PCR step, the final CRF-Rlh DNA was assemble using primers E3 and El l.
Sequences of primers:
E3: 5'-AAAAGCTTAGGACCCGGGCATTCAGGA-3' (SEQ ID No. 38) E9: 5'-GAAGGAGTTGAAGTAGATGTAGTCGGTGTACA-3' (SEQ ID No. 39)
El 1: 5'-AAGAATTCTCAGACTGCTGTGGACTGCT-3' (SEQ ID No. 40) E12: 5'-CATCTACTTCAACTCCTTCCTG-3' (SEQ ID No. 41)
E16: 5'-CATTCAGTACAGGGCTTCTTTGTGTCTGTG-3' (SEQ ID No. 42) E17: 5'-AAGAATTCTCATCCCCCCAGCCACAG-3' (SEQ ID No. 43)
El 8: 5'-ACAAAGAAGC-CCTGTACTGAATGGTCTCAG-3' (SEQ ED No. 44) E19: 5'-AAGAATTCTTTGTCCCACCACGGTGTGCTC-3' (SEQ ID No. 45) E24: 5'-CTCCTCATTGAGGATCTCCT-3' (SEQ ID No. 46)
E25: 5'-CTTGCTTTTTTTGAGATGTTGCTGGCCAGGGA-3' (SEQ ID No. 47) E27: 5'-GGTAGTGCACCTTGCTTTTTTTCTCTCCCCA-3' (SEQ LD No. 48) E28: 5'-TGGTAGTGCACCTTGCTTTTTTTGAGATGTTGC-3'(SEQ ID No. 49)

Receptor isoforms e and h were tested for their ability to produce soluble forms of the CRH receptor that would affect the activity of the most prevalent CRH-Rlα isoform. To allow detection by immunoblotting, a cDNA for the V5 protein w as attached to the cDNA for the isoforms in the constructs. The constructs were thus transfected into COS cells and tested as to whether the isoform messages were translatable, and whether cotransfection of CRH-Rlα with CRH-Rle or h affects the activity of CRH-Rlα. . . . . .
COS cells were propagated in DMEM medium (GIBCO) supplemented with 10% Fetal Bovine serum and antibiotics (GIBCO). Ten thousand cells were routinely transfected in each well of opaque 96-well plate (Packard) by Lipophectamine (Invitrogen,

Carlstand, CA) according to the manufacturer's protocol.
Western blot analysis (Figure 7B) using antibody against V 5 protein shows that CRH-Rle and h are translated into final protein products (see the main bands in lines 2 and 3 of Figure 7B). Samples of protein extracts of untransformed COS cells and cells transfected with pCRFRlel-V5 or pCRFRlh-V5 were probed with mouse anti-V5 antibody (Invitrogen) and anti-mouse HRP.
Cyclic AMP concentration was measured by a cAMP functional assay kit (Packard BioScience, Meriden, CT). Stimulated cells were washed 3 times by PBS and incubated for 1 hour in 25 μl of lysis buffer at room temperature. The signal was measured using a Fusion α instrument (Packard BioScience), and cAMP concentrations were recalculated from the standard curve.
Figure 7C shows that coexpression of CRF-Rlα with

CRF-Rle inhibits CRH or urocortin-stimulated cAMP accumulation mediated by CRF-Rlα, and that coexpression of CRF-Rlα with CRF-Rlh stimulates CRH or urocortin-stimulated cAMP accumulation mediated by CRF-Rlα.
Taken together, Figures 7B and 7C demonstrate th at

CRH-Rle and CRH-Rlh are translated into final biologically active proteins. The soluble CRF-Rle would bind the receptor ligand, thereby decreasing the extracellular concentration of the CRH or CRH-related peptide available for interaction with CRH-Rlα. The final effect is the inhibition of biological activity, as demonstrated by the decreases in cAMP production presented in the upper panel of the Figure 7C. On the other hand, the CRH-Rlh isoform would bind the receptor ligand .(CRH) and stimulate . cAMP production by the alpha isoform, through either stabilization of the ligand or induction of receptor dimerization. The final effect of this latter interaction is the stimulation of biological activity, a s demonstrated by the decrease in cAMP production presented in lower right panel of Figure 7C.

The following references were cited herein:
1. Perrin, et al., (1999) Ann NY Acad Set 885, 312-328
2. Chen, R., Lewis, K.A., Perrin, M.H., and Vale W.W. ( 1993 ) Proc. Natl. Acad. Sci. USA 90, 8967-8971
3. Liaw, et al., (1996) Endocrinology 137, 72-77
4. Chang, et al., (1993) Neuron 11 , 1187-1195. J. Clin. Endocrinol. Metab. 85, 3575-3581
5. Perrin, et al., (1995) Proc. Natl. Acad. Sci. USA 92, 836-840 6. Lovenberg, et al., (1995) Proc. Natl. Acad. Sci. USA 92 , 836 - 840
7. Vita, et al., (1993) FEBS Lett 335, 1-5
8. Stenzel, P., Kesterson, R., Yeung, W., and Cone, R. D. ( 1995 ) Mol. Endocrinol. 9, 637-645
9. Kishimoto, T., Pearse, R. V., Lin, C. R., and Rosenfeld, M. G. (1995). Proc. Natl. Acad. Sci. USA 92, 1108-1112
10. Dautzenberg, F.M., Dietrich, K., Palchaudhuri, M.R., and Spiess J., (1997) J. Neurochem. 69, 1640-1649
11. Arai, et al., (2001) Endocrinology 142, 446-454 12. Chalmers, et al., (1996) Trends Pharm. Sci. 17, 166-172

13. Spiess, et al., (1998) Trends Endocrinol. Metab. 9, 140-145

14. Sakai, et al., (1998) Gene 219, 125-130
15. Ross, et al., (1994) Biochem. Biophys. Res. Commun. 205 , 1836- 1 842
16. Grammatopoulos, et al., (1999) Mol. Endocrinol. 13 , 2 1 89 - 2202
17. Perrin, M.H., Sutton, S., Bain, D., Berggren, W.T., and Vale, W.W. (1998) Endocrinology 139, 566-570
18. Wille, et al., (1999) Identification of amino acids in the N- terminal domain of corticotropin-releasing factor receptor 1 that are important determinants of high-affinity ligand binding. J. Neurochem. 72, 388-395
19. Nabhan, et al., (1995) The alternative spliced type I I corticotropin-releasing factor receptor, stably expressed in

LLCPK-1 cells, is not well coupled to the G proteins. Biochem. Biophys. Res. Com. 212, 1015-1021
20. Slominski, et al., (2000) Corticotropin releasing hormone and proopϊomelanocortin involvement in the cutaneous response to stress. Physiol. Rev. 80, 979-1020 .
21. Slominski, A., and Wortsman, J. (2000) Neuroendocrinology of the skin. Endocrine Rev. 21, 457-487
22. Slominski, et al., (1999) Cutaneous expression of CRH and CRH-R: is there a "skin stress system"? Ann. NY Acad. Sci. 885 , 287-311
23. Slominski, A., et al., (2000) The skin produces urocortin. J. Clin. Endocrinol. Metab. 85, 815-823
24. Slominski, A et al., (2001) FASEB J., 15, 1678-1693
25. Quevedo, et al. (2001) In Vitro Cell Develop. Biol. 37A, 50-54 26. Slominski et al., (2000) In Vitro Cell Develop. Biol. 36, 21 1 - 2 1 6
27. Slominski et al., (1996) Biochim. Biophys. Ada 1289 , 247 - 25 1
28. Slominski, A., Costantino, R., Wortsman, J., Paus, R., Ling, N. (1992) Life Sci. 50, 1103-1108
29. Slominski, et al. (1999) In Vitro Cell. Develop. Biol. 35, 564- 565
30. Abdel-Malek, et al., (1995) Proc. Natl Acad. Sci. U.S.A 92 , 1789- 1793
31. Robbins, et al. (1992) Mol. Brain Res. 13, 83-92
32. Tsai-Morris, et al., (1996) J. Biol. Chem. 271, 14519-14525

33. Liaw, et al., (1997) Mol. Endocrinology 11, 980-985
34. Smith and Valcarcel (2000) Trends Biol. Sci. 25, 381-387 35. Black, D.L. (2000) Cell 103, 363-370
36. Sorek, R., and Amitai M. (2001) Piecing together th e significance of splicing. Nature Biotech. 19, 196
37. International Human Genome Sequencing Consortium (2001 ) Nature 409, 860-921
38. Black. D.L. (1998) SpliNeuron 20, 165-168
39. Arts, et al., (2000) Hum Mol Genet. 9, 237-247
40. Hinuma, et al., (1997) Human CRF2 receptor protein, its production and use. Patent: JP 1997070289-A
41. Slominski, A., and Paus, R. (1993) J. Invest. Dermatol. 101 , 90S-97S
42. Owens, Jr. A.H., Coffey, D. S., and Baylin, S. B., editors. ( 1982) Tumor Cell Heterogeneity. Academic Press, New York
43. Pawelek, et al., (1992) MolecPigment Cell Res. 5, 348-356

44. Slominski, et al. (1998). Clin Dermatol. 16, 503-515 45. Slominski, et al., (1996) FEBS Lett. 399, 175-176
46. Slominski, et al. (2000) J. Clin. Endocrinol. Metab. 85 , 3575 - 358 1 . . . . . .
47. Slominski, et al.. (1988 J. Biol. Chem. 211, 12281-12286 48. Bomirski, A., Slominski, A., and Bigda, J. (1988) Cancer Metastasis Rev. 1, 95-118
49. Slominski, et al.. (1988) /. Cell Sci. 89, 287-296
50. Guest, et al.. (2000 Brain Res. 876, 238-244.
51. Wood, et al. (1999) Exp. Dermatol. 8, 153-164
Any patents or publications mentioned in this specification are indicative of the levels of those skilled in the art to which the invention pertains. Further, these patents and publications are incorporated by reference herein to the s ame extent as if each individual publication was specifically and individually indicated to be incorporated by reference.
One skilled in the art will appreciate readily that th e present invention is well adapted to carry out the objects and obtain the ends and advantages mentioned, as well as those objects, ends and advantages inherent herein. The present examples, along with the methods, procedures, treatments, molecules, and specific compounds described herein are presently representative of preferred embodiments, are exemplary, and are not intended as limitations on the scope of the invention. Changes therein and other uses will occur to those skilled in the art which are encompassed within the spirit of the invention as defined b y the scope of the claims.