CA2134550A1 - Human crabp-i and crabp-ii - Google Patents

Abstract

2134550 9322331 PCTABScor01 The sequences encoding two isoforms of human cellular retinoic acid binding proteins, CRABP-I and CRABP-II, have been cloned and sequenced and are set forth with their corresponding amino acid sequences in SEQ ID NOS. 1-4. The indentification of human CRABP
nucleic and amino acid sequences provides the basis for the construction of recombinant human CRABP vectors and expression constructs. Human CRABP can also be synthesized or produced ex vivo, e.g. in bacterial or other production systems. Ligand binding assays, including recombinant and chimeric receptor reporter assays, and direct and competition hybridization assays employing the human CRABP sequences herein described can be used to test drugs for retinoic induction and tissue specificity for pathologies in which retinoids are implicated. Immunoassays utilizing antibodies or binding fragments produced to human CRABP can also be used to test patient tissues for the presence and levels of CRABP for diagnosis and to monitor treatment. The indentification of the nucleic and amino acids sequences for human CRABP-I and CRABP-II also contributes to the elucidation of the function and interaction of the retinoid-binding proteins. The CRABP-II gene, isolated from a human placenta genomic library, spans 6 kilobases and includes 4 exons. One major transcription initiation site was mapped to an A
residue 137 nucleotides upstream of the ATG initiation codon.
CRABP-II mRNA was rapidly induced within 2-6 hours in culture by retinoic acid, primarily due to an increased rate of transcription which required on-going synthesis. The human CRABP-II gene is thus apparently transcriptionally regulated by a newly synthesized regulator protein. Once the CRABP-II is produced, message stabilization may provide means by which elevated CRABP-II in mRNA is maintained.

Description

213~5~ .
I-- wo 93/2~331 i Pcr/uss3/o3936 HUMAN CRABP-I AND CRABP-II
FIELD OF THE INVENTION
The present invention relates g~nerally to cellular retinoic acid binding proteins (CRABPs) and, more specifically, to human CRABP-I and CRABP-II and the sequences5 encoding them, and their use in various assay systems for screening and diagnostic applications and for Ulerapeutic purposes.
RELATED APPI ICl~TlONS
This is a continuation of U.S. Application Serial No. 07/874,847, entitled "Human CRABP-I and CRABP~ a filad April 28,1992 by Voorhees et al. which is a continuation-in-10 part of co-pending U.S. application Serial No. 751,8~3 entitl0d ~Human CRABP-I and CRA3P-IIR, filed August 30,1991 by Voorhees et al., all her~in incorporated by reference.
GENE!~ANK ACCESSION INFORMATlON
GENE ACCESSiON NO.
human CRABP-II M68867 E~ACKGROUND OF THE INVENTION
Retinoids are essential regulators of epithelial cell growth and cellular dfflerentiation, skin being a major target in both normal and pathological states. Spom, M.B. et al., Cancer Res. 43:3034 3040 (1983); Kopan, R. ~t al., J. Cell BioL 109:29~307 (1989);
Asselineau, D. at al., Dcv. Bio/. 133:3æ-3~5 (1989); and Uppman, S.M. ~t al., Pharmacol.
ZO Th~r. 40:107-122 (1989).1t has been shown that retinoids prevent cancer in skin and have efficacy as agents Tn human malignant and premalignant cutaneous disorders. Asselineau, D. et al., Dev. Biol. 133:322-335 (1989). It has also been shown that retinoids cause growth ir.hibition in many hyper-proliferating cell lines, a feature that makes the compounds of fundamental interest as anti-tumor and anti-psoriatic agents. Sporn, M.B. st aî., Cancer 25 Res. 43:303~3040 ~1983); and Asselineau, D. et al., Dev. Biol. 133:322-335 (1989).
Retinoids also play fundamental roles in directins the spatial organization of cells during development and the generation of vertebrate limbs. Eichele, G. Trends Genet. 5:24~251 ~1989); and Summerbell, D. et al., Trends Neurosci. 13:142-147 (1990).
The elucidation of tlie function of retinoids in the complex biologicai processes 30 involved in cell growth and differentiation requires the identffication of the specific components of the retir, ~ signal transduction system as well as the genes directly regulated by this system. ~everal intraccllular reUnoid-binding proteins have already bcen identified, including cel!ular retinol-binding proteins (CRBP), nuclear retinoic acid receptors (RAR), cellular retinoic acid binding proteins (CRABP) and, most recentiy, RXRs, also 35 belonging to the nuclear receptor superfamily of genes. See Sundelin, J. et al., J. Biol.
Chem. 260:6488 6493 (1986); Li~ E. et al., PNAS (IJSA) 83:577~5783 (1986); Nilsson, M.H.L et al., Eur. J. Biochem. 173:4~51 (1988); Stoner, C.M. et al., Cancer Res.

213~S'~
WO 93/22331 . ~ PCI~US93/03936 ~--

2 -49:1497-1504 (1989); Giguere, V. et al., PNAS (USA) 87:6233 6237 (1990); Petkovieh, M.
et al., Naturs 330:444450 ~1987); Brand, N. et al., Nature 332:850 853 (1988); Benbrook, D. et al., Nature 333:66~672 (1988); Zelent, A. et al., Nature 339:71~717 (1989); Krust, A.
et al., PNAS ~USA) 86:531~5314 ~1989); and Mangelsdo~, D.J. et al., Nature 345:22~229 5 (1 990).
Cellular retinoic acid binding proteins (CRABP) ar~ low molecular weight protsins prasent in human skin with increased levels found in psoriatic lesions and aftcr external and systemic retinoid ~reatment. Sieganthaler, a. et al., J. Inv~st. DermatoJ. 86:4245 (198~); Hirschel-Scholz, S. et al., Eur. J. Clin. Invest. 19:22~227 (1989); and Siegenthaler, 10 G. et al., Arch. De~matol. 123:169~1692 (1987). Although unde~ctable in keratinocytes grown in low c~lciurn medium, CRABP is expressed when a more dl~ferentiated phenotype is induced by growth to confluence in the presence of elevated extracellular calcium concentrations. Siegenthaler, G. et al., Exp. CellRes. 178:114126 (1988).
Although th~ precise role of CRABP in retinoic acid ~RA) action has not been 15 determined, it has been suggested tha~ CRABP might act as a shuttle protein, facilitaUng tha movement of RA to th~rnucleusl or that CRABP might sequester RA, thereby decreasing the cellular response. Takæe, S. et al., Arch Biochem. Riop~fs. 247:~28 334 ~1986); Maden, M. at al., Natu~e 335:733 735 (1988). In a recent stuciy, }t was found that all biologically active RA analogs in F9 cells bound to RARs, while h~o of them did not bind 20 to CRABP. 3enbrook, D. et al., Nature 333:66~672 (1988). This suggests that retinoid binding to RA, but not necessarily to CRABP is necessary to induce cell dfflerentiation.
The effects of retinoic acid on gene transcription can be mediated by retinoic acid receptors (RARs) and retinoid X receptors (R,YRs). Leed, M. et ai., Cell 68:377-395 (1992);
Mangelsdorf, D.J. et al., Genes DeY. 6:329 344 (1992). RARs have been shown to bind 25 retinoic acid (RA) with high affinity, while iV(Rs apparentiy have no affinity for this ligand.
Mangelsdort, D.J. et ai., Natv~ 345:~229 (1990). However, it was recently demonstrated that ~cis RA can bind to PXR - ~ with high affinity. Levin, A. A. et al., N2ture 35~: 35~361 (1992~; Heyman, R.A. et al., Cell 68:397406 (1992). CRABPs have beenshown to bind PA with high affinity, but thair ~unction is pooriy understood. However, it 30 was recently demonstrated that CRABP may be invoived in cytochrome P 450 metabolism of i~ Fiorella, P.D. et al., J. 8iol. Chem. 266:16572-16579 (1991).
Two isofonns of CRABP, CRA8P-I and ll, have been identifi~d and cloned In the mouse. Stoner, C.M. et al., Cancer Res. 49:1497-1504 (1989); Giguere, V. et al., PNAS
(USA) 87:6233 6237 (1990); Nilsson, M.H.L ~t ai., Eur. J. Biochem. 173:45-51 (1988). By 35 isoform is meant two amino acid sequences with substantial sequence idantity. Bovine - W 0 93/22331 ~ 21~4~50 PCT/US93/03936 CRABP-I has also been sequenced, and the NH2 terminalregions ofrat and chicken C:RABP-lhave also been recently determined. Bailey, J.S. etal.,J. Biol. Chem. 263:932 9332 (1988); K~amoto,J. et al., Biochem. Biophys. Res. Comm. 157:1302-1308 (1988).
Human CRABP has,however, not been pre~ous~ isolated or cloned.
SUMMARY OF THE INVENTION
The sequences encoding h~o isoforms of human c~llular retinoic acid binding proteins, designatad CRA~P-I and CRABP-il! and the gene for CRABP-II, have been cloned and sequenc~d. llleir nucleic acid and corresponding amino acid s~quences are sat ~orth in the sequence listing preceding the claims. Expression ot human CRA13P-II, but not CRABP-I was markedly increased in human skin in vivo and in skin fibroblasts fn vitro a~t~r treatment with retinoic acid (RA) and a~ter trea~ments which induoe keratinocytedfflerentiation. The importance of RA dependent mRNA stabilization tor ksying CRABP 11 message at induced levels once transcription has occurrsd is also described.
The cloning and sequencing of human CRABP provides the basis for the construction of hurnan CRABP-I and 11 viral, prokaryotic and eukaryotic expression vactors and recombinant expression constructs. Hurnan CRABP can now also be produced synthetically or ex vfvo (outside thc human body), for example, through the production of h sion proteins in bacteria and later cleavage and purification of CRABP therefrom.
Ugand binding studies utilizing human CRABP sequenc~s can determine ligand 20 binding amnity and the interaction of human CRABP with other human retinoid-binding proteins, and can be used to better identi~y tissu~speciffc drugs for pathologies in which retinoid are implicated. Various ass~v schemes, including reporter assay systems, direct and competition hybridization and Dinding assays employ the nucleic and amino acid sequenc~s hersin described. Artibodies or binding fragments ther60f produced to human 25 CB~BP can also be used in immunoassays of patient tissues for CRABP levals for diagnosis and the monitoring of treament. Purified or synthetic human CRABP can also be used for supplementaUon therapy.
Other ~atures and advantages of the present invention will become apparent from the following description and appended claims, taken in conjunction with the 30 accompanying drawings.
BRIEF DESCRIPTION OF THE DRAWINGS
Figure 1 gives the sequences of the fo~ward and reverse degenerate primers used in cloning human CRABP.
Figure 2 compares human, mouse, rat and chicken CRAaP. Panel A is a 35 comparison of the amino acid sequences of human (h) and mouse (m) CRABP. Dashes WO g3~22331 ' PCI'/US93~03936 ~, represent sequence identity; the asterisk at residue 118 represents a gap introduced in the CRABP-I sequence for ma~imum alignment. Panel B is a sequence comparison of the NH2-terrninai ends of hurnan (h), mouse (m), rat (r) and chicken (c) CRABP. Boxed residues represent those dissimilar to human CRABP-II.
Figure 3 is an autoradiogram of RNA blots deriv~d trom ~hree individuals which illustrates induction of CRABP-II mRNA in human skin by topically applied retinoic acid (RA). Lane (a) represents no treatments; lane (b) RA vehicle; and lane (c) 0.1% RA cream in RA vehicle under occlusion.
Figure 4 is a bar graph of the RNA blot hybridkation results (quanfflated by laser densitometry) of nine independent axperiments involvin0 five dzrmai fibroblast lines prepared from three individuals and three diploid human lung fibroblast lines. The results iHustrate the induction of human CRABP-II m~NA in human darmal fibroblasts compared to lung fibroblasts. Ths inset shows the relevant autoradiographic bands trom hvo representati~ s experiments comparing dermal and lung fibroblasts.
Figure 5 illustrates the expression of CRABP-II mRNA in cultured k~ratinocytes under various conditions. Figure 5A is a bar graph of the r~suts obtained tor seven independent keratinocyte strains and illustrates the effects of confluence on human CRABP-II mP~NA levels. The error bars rapresent ~ SE ~p<0.005, ~*~p~0.0005 relative to cells 2 days preconfluence. Figure 5B shows tha effects of RA and increased calcium 23 concentration on human CRABP-II mRNA levels in postconfluant keratinocytes, with treatment tor prolonged periods with low concentrations of RA (3x109M~ having nodetectable effect. Figure SC illustrates that prolonged treatment of pos~confluent keratinocytes with higher concentrations of RA (3x10~M) reduced Ci~9BP-II mRNA to undetectable levels.
Figure 6 is a schematic diagram of a retinoic acid receptor (RAR)-CRAi3P or RXR-CilABP cotransfection assay.
Figure 7 is a restriction map of a bacteriophage lambda clone (A2.1 ) isolated from a human placenta genomic library of the human CRABP-II gene is provided with the exons indicated as filled boxes numbered I to iV.
Figure 8 illustrates the nucleotide sequence and shown above the nucleotide sequence is the deduced amino acid sequence at the human CRABP-II gene.
Figure 9 illustrates the transcripUon start sites in the human CRAi3P-II gene asdetermined by primer extension anaiysis, with thQ major transcripbon site indicated by an arrow and the corresponding base in the sequenc~ indicated by an asterisk.

I -; W093/22331 2 1 ~ ~ 5 5 0 PCr/USs3/03936 Figure 10 illustrates a nuclear run~ff assay using nuclei isolated from cu~ured human skin ffbroblasts treated with retinoic acid.
Figure 11 illustrates the mRNA lavel and transcription tate o~ C:RABP-II after quantitation by phospho~maging and normalization to cyclophilin.
Figure 12 is an autoradiogram demonstrating the effects ot cycloheximide or actinomycin D on the induction of CRABP-II mRNA expression.
Figure 13 is a bar graph demonstrating the effec~s of eycloheximide on CRABP-II
transcription rates determined by nuclear run~n assays.
DETAILED DESCRIPTION OF THE PREFERP~ED EMBODIMENTS

The cloning and sequencing of two human CRABP cDNAs revealed one with a predicted amino acid sequence 99.3% identical to tha mouse and bovine CRABP-I, and a second with a predicted amino acid sequence, 93.5% identical to the mouse CRABP-II.
The CRABP-II described herein appears to be the human homolog to rnouse CRABP-II, 15 since both are expr6ssed in adult skin, and is therefor dasignatad as such. The high amino acid homstogy seen behNeen boYine, mouse and human CRABP-I indicates that this isofonn has been more conssrved throughout 0volution than th3 CRABi'-ll (Figure 2A
described below). This was especially evident when the NH2-~erminal sequence between rat, chicken, mouse and human CRABP-i and CRABP-II were compared. No dfflerences20 were seen between the CRABP-ls, while several dfflerences sxist betYeen the CRA8P-lls.
CRABP-I transcripts were undetectable in adult human epidermis by RNA blot hybr;dkation, while the CRABP-II cDNA probe d2tected an approximately 1.2 kilobase (kb) mRNA transcript. External applicaUon o~ 0.1% retinoic acid cream in vivo for 16 hours resulted in a 1 ~fold induction of CRABP-II, but not CRABP-I. CRABP-II mRNA, was aiso 25 markedly increased (>1 ~fold) by retinoic acid treatment of fibroblast cultured from human skin, whereas no significant induction of CRABP-II mRNA was observed in human lung fibroblast. Human CRABP-II, but not CRABP-I mRNA was significantly induced by treatrnents which induce keratinocyte dfflerentiaUon in vitr~. Highly increased levels of CRABP-II RNA were found in psoriatic apidermis when compared to normal epidermis for 30 10 dfflerent patients. CRABP-I message levels were on the other hand very low or undetectable in both normai and psoriatic skin ~data not presentsd in Figures). Previous studies Ot the induction of CRABP by natural and syntheUc retlnoTds and their increased Ievels during keratinocyte dfflerenffation are consist~r,t with the ~xpression of CRABP-II
message seen in the present investigation, but not with the exprsssion of CRABP-I. These 213~550 - ;
WO93~22331 ` ; PCI`~US93/03936 ~`, studies thus identify CRABP-II as the isoforrn likely to be expressed and regulated by RA
in adult human skin.
Since mRNA isolated from skin biopsiçs used in this study is approximately 95%
derived from keratinocytes (see Voorhees, J.J. et al., Arch. Dermatol. 105:69~701 (1972)), 5 the majority of RA induetion of CRABP-II seen in vivo.~(Figura 3 deserlbed below) cannot be explained by the presence of mRNA derived from dermal fibroblasts. This is in contrast to the reduction of CRABP-II transcripts observed in response to high concentrations of RA in keratinocytes in vi~o (Figure 5C describ~d below). When grown to postconfluence under the same conditions used hsrein, human keratinocytes undergo coordinate 10 increases in involucrin content and transglutaminase activity, h~o key determinants of comffled envelope formation during terminal dfflerentiation. Pillai, S. et al., J. Cell. Physiol.
143:294302 (1990). Moreover, these events are accelerabd and began to occur prior to confluence when the extracellular calciurn concentrations were increased to 1.2 or 2.4 mM.
Pillai, S. et al., J. Cell. Physiol. 1~3:29~302 (1990). CRABP-il mRNA levels responded to 15 confiuence and external calcium in an identical fashion (Figure 5 described below). These results strongly favor the concept that the stratified structure of th~ epidermis and/or the presence of dermis is an important detsrminan~ of CRABP-II rsgulation in vivo, and may help to account for ths dfflerenUal rssponsiveness of keratlnocytes to RA in w~o and /n vivo. The nsgative eflect of high concsntrations of RA on C:RABP-II exprsssion seen in 20 vitro is similar to the effect of this compound on transglutaminase type I mRNA levels in cultured keratino~tes. See Floyd, E.E. et al., Mol. Cell. BioL 9:4846~851 (1989).
Whether RA inductlon of tne human CRABP-II gene occurs at the level of transcription and whetherthis regulaUon is mediated by specffic nuclear receptors remains to bs investigated. It has been shown that human skin as well as cultured human skin 25 fibroblasts express RAR~. Krust, A. et al., PNAS (USAJ. 86:531~5314 (1989); and Elder, J.T. et al., J. Invest. Dermatol. 96:425 433 (1991). However, this does not explain the lack of RA induction ot CRAi3P-II mRNA seen in cuitured human lung fibroblasts and keratinocytes, also known to express RAR~. Elder, J.T. et al., J. Inves~ Dermatol.
96:425 433 (1991). H the CRABP-II gene is reguiated by the RARs, additional Ussue or 30 cell-specific factors may be required for RA induction.
Members o~ the nuclear superfamiiy of receptors have been shown to interact withtheir responsive elements as dimers. 61ass, C.K et al., Cen 59:697-708 (1989). RARs have also been shown to Interact ~vith other members of this tamily (I.e., the thyroid hormone receptors) forming heterodimers. Glass, C.~ et al., Cea 59:697-708 (1989). One 35 of the most exclting findings recently Ts that the thyroid hormone recsptors require an WO 93/22331 ~ 2 1 3 ~ 5 5 0 Pcr/l~ss3/o3936 auxiliary protein (T~AP) to interact wth the thyroid hormone responsive element in the growth hormone gene. O'Donell, A.L et al., Mol. Endocrinol. 5:9~99 (1991). TRAP
apparently is forming a heterodimer w~h the thyroid hormone receptor on the responsive element. Such dimerization betw0en nuclear receptors and oth~r transcription factors 5 could explain tissue-specific regulation. ff there is a skin-specific factor forming a neterodimer with the RARs, that could explain why the C:RABP-II gene is induced by RA
in skin fibroblasts, but not lung fibroblasts.
In summa~, we have demonstrated that CRABP-II is expres~ed in human skin in vwo and the CRABP-II gene appears regulated by RA in skin in YiVo and in cultured skin 1 O fibroblasts in V~D. CRABP-II was not, however, inducsd by RA in cu~ured lung fibroblasts, demonstrating cell-specific regulation of this gene. CRABP-I, on the other hand, does not appear regulated by RA and is found at very low or undetectable levels in human skin in vivo, as wall as in keratinocytes and fibroblasts. This suggests that CRABP-II may participate in a rsgulato~ feedback mechanism to control the ac~ion of RA on cell 1~ dfflerentiation in skin. The identification of human CRABPs, RARs and RXRs now allows studies, such as those describ~d below, on intera~ons between members of these families in Ule complex molecular and cellular mechanisms of RA action.
The identificaUon of the nucleic and amino acids sequences of human CRABP-I and CRABP-II provTde tha basis ~or a variety of recombinant products, including vectors 20 carrying the human CRA3P cDNA secuences and expression constrwts cotransfected or infected w th such vectors. For example, plasmid or viral vectors carrying human CRABP
cDNA have been constructed and are used to cotransfect or coinfe~t receptor-deficient CY-1 monkey kidnsy cells. Reporter assay systems utilizing CV-1 recombinant expr~ssion constructs which include human CRABP cDNA, a reporter element containing a retinoid responsive element and a reporter gene, and preferably internal control sequences, such as generaliy descrlbed in Astrom, A. et al., Biochem. Biophys. Res. Comm. 173:339 34S
~- 390), oan also now be constructed. Chimeric receptor prot~ins such as those described in U.S. Patent No. 4,981,784 to Evans et al., can aiso be synthesized and utilized in rsport6r assays.
Assay systems emp'-\ ing human CRABi' in conjunction with additional retinoid-binding proteins are aiso coMemplated as within the scope of the Inven~on. i~c ~ example, as shown schemaffcally in Figure 6, viral (e.g. SV40) vector carrying a human CRABP, a reporter plasmid carrying t5~-~ retinoid responsn~e element (e.g. RRE3~-CAT), a viral (e.g.
SV40) vector carrying a human RAR or i~R of interest and a ~-galactosidase vector (pcH110) for an internal control are coinfected or cotransfected into CV-1 cells. As shown ~1345~
WO 93/~2331 ' ` PCr/US93/03936 ~'~
- 8 - ~
in Figure 6, the recombinant CV-1 construct is exposed to the binding ligand (e.g. retinoid) of interest. By ligand is meant a molecule which binds to the receptor binding protein and induces the expression ot the gene of interest. In this assay system, induction of the reporter gene is used to assay for ligand binding to receptor protein. Absent any 5 additional regulatory requirements, functional ligand would bind RAR or i~(R, sUmulat~n~
the expression (and tran~lation) of the reporter gene ~CAT). By including human CRABP
in the system, the interaction between human CRA8P and RAR or i~YR or other binding receptors can be determined. For example, if CRABP sequesters PA in the cytoplasm, less ligand wili reach the nucleus, thereby reducing RAR or ~XR-mediated stimulation of the 10 reporter gene.
The sequencing of human CRABP also allows the raising or production of antibodies or binding *agments, e.g. F~ab), which can be used in immunoassays tofurther characterize binding or for diagnostic purposes and to monitor the course of patient treatment. For example, patient tissue can be assayed for the pressnce and levels of 15 human CRABP to diagnose particular conditions where retinoids are implicated and to monitor the effectiveness of drugs and other treatments in aitering patient levels of CRABP.
CRABP purified directiy from human tissue or cells, in cuiture or human CRABP produced synthetically or ex vivo will also provide CRABP ~or supplementation therapy where needed.
The gene for human cellular retinoic acid-binding protein ll (CRABP-II) has beencloned and sequenced. It was isolated from a human placenta genomic library and is contained within one bacteriophage clone. The gena spans 6 kilobases and cons~sts of four exons and three introns as do other m0mbers of the hydrophobic ligand binding protein sens family. Wei, LN. et ai., DNA Cell Biol. 9:471~78 (1990). The mouse CRABP-I
25 gene has previousiy been cloned, demonstrating a dfflerent intron, but similar exon arrangement. Wei, L.N. et al., DNA Cel/ Biol. 9:471478 (1990). One major transcription site was mapped to an A residue 137 nucleotides upstream of the ATG initiation codon.
The sequence of the upstream region of the CRABP-II gene is rather GC rich and has a TATA box at ~i, and severai possible binding sites for transcription tactors.
Of spedal interest is the presence ot potantiai AP2 (CCC/GCA/GGGC) sites in the 5' flanking region. AP2 has been shown to be idenUcal to the transcription factor KER1 which has been suggest0d to bs gen~rally invoived in e~idermal gene regulaUon. L~ask, A. et ai., PNAS ~JSA) 88: 794~7952 (1991). CRABP-II is predominanUy expressed in the skin of aduit mice and we have recentiy demonstrated that CRABP-II, but not CRA8P-I is expressed in human skin. Giguere, V. et al., PNAS ~JSA) 87:6233 6237 (1990); Astrom, A.

2~3455 0 WO g3/2~331 PCr/US93/03936 g et al., J. Biol. Chem. 266:17662-17666 (1991). In addition, both AP2 and CRAi3P-II mRNA
have been shown to bs induced by RA. Astrbm, A. et al., J. ~iol. Chem. 266:17662-17666 (1991); Luscher, B. et al., Genes Dev. 3: 1507-1517 ~1989). Whather AP2 is involved Tn skin-spec~ic expression and ~A Induction of the CRABP-II gene remains to be datermined.
5 As shown in Fi~ure 8, the upstream region also contains a high amnity Spl bindiny site (Gt3GGCGGAGC) close to tha TATA box, and two sequences (GCGGGGGCG) identical to ~ox-~4 binding sites. Kadonaga, J. T. et al., rrends E~lchem. Scl. 11:20 23 (1986).
Lemaire, P. et al., MoL C~ll. Biol. 10:3456 3467 (1990). K~ox-24 is a member of the early responsive gene family, suggested to be Tnvoived in regulation of cell proliferation and i O dfflerentiation. Lemaire, P. et al., Mol. Cell. Biol. 10:345~3467 (1990); Edwards, S.A. et al., Dev. Biol. 148:16~173 (1991). It was recently shown that Egr-1 Kox-24 was induced by RA in embryonal P19 cells. The induction of Egr-1 probin by RA in these cells, demonstrated~a transient-5 fold increase, peaking around 30-60 minutss and returned to basic levels between 60-90 minutes. Thus, as shown in Figure 11, the induction of Egr-1 15 protein by RA just precedes CRABP-II gene transcription. However, the invoivement of Egr-1 In RA induced transcriptional regulation ot the CRABP-II gene in skin fibroblasts is unlikely, since Egr-1 was found to be induced by treatment of the cells with cycloheximide while CRABP-II genB expression in this study was found to be inhibited by this compound as demonstrated in Figures 12 and 13. Edwards, S.A. et al., Dev. Biol. 148:16~173 (1991).
20 The upstream region of the CRABP-II gene also contains a direct repeat (G/AGTTCA) - spaced with one nucleoUde with homology to the RARE found in tne RAR-~2 promoter, except that the RAR-~2 RAREIs spaced by ~ve nucleoUdas. Umesono, K et al., Cell 65:1255 1266 (1991). It was recently shown tnat ~NO other memb~rs ot the hydrophobic ligand-binding family of genes contain a RARE(CRBP-I) and a RXRE(CRBP-II) in their 25 promoters. Smith, W.C. et al., EMBO J. 1 0:2223-2230 (1 991 ); Mangelsdorf, D.J. et al., Cell 68:5~561 (1991). Whetner this direct repeat Is functional remains to be determined. To determine whether RA inductlon of CRABP-II Is transcriptional, nuclear run-on assays were performed. As can be seen in Figures 10 and 11, treatment of cuitured human skinfibroblasts with RA resuited in a rapid transient ~fold increase of transcription, tollowed by 30 a ~fold inductlon of CRABP-II mRNA levels. lllus, the CRABP-II gene Is mainlytranscriptionally activated by RA. The CRABP-II mRi~lAwas rapidiy induced within 2~ hours In cultured human skln flbroblasts by retinolc acid, reaching a plateau after 6 hours of treabnent. However, removai of reUnoic acld from tha medlum aRer 12 hours caused a sharp decline in CRABP-II mRNA levels. The rapid increase of GRABP-II message was 35 mainiy due to an increased rate of transcription as determined by the nuclear run-on i 5 3 wos3/2233l ~ ~ Pcr/~iS93/03936 ;~
- ~0 - ,, experiments as shown in Figure 10. Increased transcription could be detected as early as 1 hour aftsr addition of RA, peakad at 2 hours and returned to basal levels within 6 hours.
in addition, both the accumulation of rnessage, as shown in Figure 12, and the inductlon of transcription, as shown in Figure 13, by RA was inhibited by cycloheximide, sugges~ng 5 that the CRABP-II gene is transcriptionally regulated by a newly synthesized protein. Once transcription had occurred the message reached a plateau and did not decline until RA
was removed from th~ medium. On-going protein synthesis was required for the transient inoreas~ in transcription, sinca the induction was blocked by cycloheximide. ~ is vely unlikely that the dacrease of CRABP-II mRNA seen was a result o~ reduced transcription 10 of the gene, since the RA induction of transcription was transient an~ back to control levels within 6 hours. A more likaly explanation would be that RA was involved in stabilkation of the ~:RABP-II message. The CRABP-II mRNA is likely to be unstable, since it contains AU rich sequences in the 3' untranslated region. Giguere, V. et al., PNAS (USA) 87:6233 6237 (1990); Astr~m, A. et al., J. Bfol. Chem. 266:17662-t7666 tl991). ~t has been 15 shown that many transiently expressed genes, including Iymphokine genes, c-myc, and c-fos contain AU rich sequences in their 3' untranslated regions, and that the presence of these sequences correlates with rapid mRNA degradation. Cleveland, D.W. et al., New BJol. 1:121-126. Mechanisms by which CRABP-II mRNA stabilization could occur, may involve the production of a factor after treatment ot the cells with RA, or interaction of RA
20 with a preexisting factor, stabilizing the message. Since cycloheximide blocked the RA
induced transcripUon of the gene and as a consequence, the induction of the message, the possibilities were not distTnguishable. It has been demonstrated that the human CRABP-il yene is transientlytranscriptionally induced by RA in human skin fibroblasts, and that this induction Ts dependent on on~oing protein synthesis. llle eariy induction of 25 transcription is followed by a rapid increase in CRABP-II message lavels that does not decrease until RA is removed from the medium, suggesting RA dependent mRNA
stabilization.
It will be appreciated that the nwleotide and amino acid sequences of the present invention can include some variaUon from the sequences represented by and 30 complementary to the sequences set forth in the Sequenc~ Usting, but must be substanUally tepresentcd by or complementary to thosQ set forth therein. By ~substantially I
represented by" or "substantially complementary to~ is meant that any variation therein does not impair the functionality of the sequence to any significant degree. As used herein, A rapresents adenine; T represents thymine; G repr~sents guanine; and C
35 represer~s cytosine; except where otherwise indicated.

- ~ WO 93/22331 '~ 1 3 4 5 ~ ~ PCI/US93/03936 SPEC:IFIC: EXAMPLES
SPECIFIC EXAMPLE 1. Clonin~ and Sequencln~ of Human CRABP
MATERIALS AND METHODS
Clonl.~g of CRABP ~rom human skln FtNA by Potymerase Chaln Reactlon (PCR) Total RNA was isolated *om human keratoma biopsies as describad in Elder, J.T.
et al., J. Invest. Derma~ol. 94:1~25 (1990) and c5NA was ~ynthesized by reverse transcription as described in Maniatis, T. et al., Molecular Cloning: A Labof~tory Manual, Cold Spring Harbor Labor~tory, Cold Spring Harbor, NY (t982). Degenerate primersderived from mouse and bovine CRABP-I were designed so as to amplify the coding 10 region. Xbal and ~amHI sites w~re contained in the forward and reverse primers, respectively, which are shown in Figure lA.
Tha cl:)NA was used as a template for PCR (2 minutss d~naturation at 92~C, 2 minutes, annealing at 4~C, 2 minutes amplffication at 72C and final ext3nsion for 10 rninut~s). After 40 cycles, th6 leng~h of the amplffied DNA was determined on a 1.5%
15 agarose gel. The amplified 43~bp region of CRABP was isolated ~rom the gel, subclone~
in~ Bluescnpt phagsmid and seguenced as described below.
Scr~ening ~nd sequencing of cDNA cion~s Pcly A+ RNA was prepared from human skin as dascribed in Elder, J.T. et al., J.
Invest. Dennatol. 94:1~2~ (1990) and used to prepare a cDNA library in i~mbda Zapll 20 (Stratagene Inc., La Jolla, CA). The library containsd 1.0 x 107 primary recombinants. The 4~bp PCR product was labeled by random hexamer priming (Boeringer Mannheim, Indianapolis, IN) and used to screen the adult human whole skin cDNA library and a human skin ~broblast cDNA library in Agt11 (Clontech, Palo Alto, CA). Duplicate nitrocellulose filters wera hybridized for 16 hours in 50% forrnamide containing 5 x SSC t1 25 x SSC = 150 mM NaCI, 15 mM sodium citrate), 1 x Denhardt's (0.02% Ficoll/0.02% bovine serum albumin/0.02% polyvinylpyrrolidone), 0.1% SDS (sodium dodecyl suifate) and 200 ~glml tRNA. The fli~ers were washed tWG ~,mes for 20 minutes in 0.2 x SSC, 0.2% SDS and one Ume for 20 minutes in 0.2 x SSCI 0.1% SDS at 55C. Two positive clones were isolated from the skin library and flve clones from the skin flbroblast library. llle clones 30 from the skin library were rescued, while the clones from the ffbroblast library were subcloned into Bluescript phagemlds (Stratagene Inc., La Jolla, CA).
DNA sequence anaiysis was performed on both strands by dideoxy chain - termination as generally described in Sanger, F. et al., PNAS ~JSA) 74:5463 5467 (1977), using mod-ffied 17 polymerase (Sequenase, U.S. Biochemical Corp.) and synthetic

3~ oligonucleot~des.

213 ~S a ~ -WO 93/22331 ; ' ~ ~ ' PCI`~US93/03936 .---Cell cuftur~
Primary cultures of normal human keratinocytes were prepared as described in Boyce, S.T. et al., in In Vitro Models for Cancer Research (Weber, M.M., and Sekely, Ll., eds) Vol. 3, pp. 24~274. CRC Press, Boca Raton, FL (1 9a6). Subcuitures were axpandad 5 in keratinocyte growth medium (KGM) (l:~lonetics, San Diego, CA). Human dermalfibroblast cultures were prepared from punch biopsies of buttock skin (see Harper, R.A.
et al., Science 204:52~527 (t979)) and propagated in modffled McCoy's 5A medium containing 10% caH serum. Human lung fibroblasts were obtained from the American Type Cultura Collection (ATCC) (Rockville, MD) and grown in the same medium.
10 Nor~hem analysls o~ mRNA
RNA was isolated from keratorne biopsies and cultured cells by guanidinium isothiocyanate iysis and uitracentri~ugation as previQus~y described in Elder, J.T. et al., J.
Invest. Dermatol. 94:1~25 (1990). For studies involving retinoic acid treatment, 0.1% RA
cream (Retin-A, Ortho Pharmaceutical Corp. Raritan, NJ) was applied once to skin and 15 maintained under plastic wrap for 4 hours to 96 hours prior to biopsy. Adjacent sites were treated with Retin-A vehicle or left untreated. After 4 hours, 12 hours, 16 hours or 96 hours, keratome biopsies were obtained and used for RNA isolation.
RiW~ concentrations were determined by absorbance at 260 nm and verified by nondenaturing agarose gel electrophoresis and ethidium bromide stainin~ as described 20 in mompson, C.B. et al., Natur~ 314:36~366 (1985). Equal quantities ot total Ri~lA were electrophoreticaliy separated in 1% formaldehyde-agarose gels containing 0.5 l~g/ml ethidium bromide and transferred to derivatized nylon membranes (Zeta-Probe, BioRad, Richmond, CA) as described in Elder, J.T. et al., J. Invest Oermatol. 94:1~25 (1990) and Maniatis, T. et ai., Molecul~r Cloning: A Laboratory Manual, Cold Spring HarborLaboratory, 25 Cold Spring Harbor, NY (1982). Fiiters were baked 2 hours at 80C in vacuo, then prehybridized for 2-4 hours at 42C in 50% formamide, 5 x SSC, 50 mM sodium phosphate, pH 7.0, 1 x Denhardt's solution, 250 ;-~g/ml yeast tRNA, 100 ~g/ml sonicated herring sperm DNA and 1% SDS. Hybridiza~ion was carried out for 1~24 hours at 42C in the same i buffer containing 1096 dextran suifate. Fiiters ware washed twic~ tor 10 minutes in 2 x 30 SSC, 0.1% SDS at room t2mperature, then twice for 20 mlnutes in 0.1 x SSC, 0.1% SDS
at 56C. Autoradiography was performed using Intensi~yTng screens at -70C. Fliters ware stripped by boiling 2 x 10 minutes In 0.1 x SSC, 0.5% SDS. Hybridization probes were prepared by random primlng (see Feinberg, A.P. et ai., Anal. E~loch~m. læ:~l 3 (1983)) of low meiting agarose-purdi6d insert fragments from the human CRABP-I PCR product, 213~5(~
Wo 93/22331 ' PCr/Uss3/~3s36 the CRABP-I cDNA (Af1.1) and rat cyclophilin. Sea Danielson, P.E. et al., DNA 7:261-267 (1 988).
Autoradiograms were quantitated using a laser densitom~tsr (LK8 Modal 2202) coupled to a Hewlett-Packard 3390 A integrator. On~ exposurec which were In the linear 5 ran~e of densitometric response were usad. When appropriate, autoradiographic intensities were normalized to cyclophilin. See Elder, J.T. et al., J Invest. Derrnatol.
94:1~25 (1990). Statistical analysis of data was per~ormed by one way analysis of variance using Scheffes correction for mufflple comparison and a ~No-tailed hypothesis.
RESlJLTS
t O Clonin~ of human C:RABP~I and CRABP~II
Using PCR and degenerate primers derhred from bovine and mouse CRABP-I, a 436 base pair (bp) product was obtained, subcloned and s0quenc*d. The predicted amino acid sequence of the PCR product was found to be 99.3% homologous to the mouse and bovine CRABP-I sequences. Sae Nilsson, M.H.L et al., Eur. J. Biochem.
15 173:4~51 (1988); and Stoner, C.M. et al., Cancer Res. 49:1497-1504 (198g). Ths PCR
product was used to screen a human skin library and a human skin fibroblast library. Five dones wers isolated from the skin fibroblast library (lf1.1, 1f3.1, lf5.1, A~5.4 and Af5.6).
Two clones (~f1.1, 1f3.1) were sequenced and found to be identieal. The predicted a nino acid sequences ot these clones wera found to be 77.4% similar to human CRABP-I and 20 93.5 % similar to the recently cloned mouse CRABP-I. See Giguere, V. et al., PNAS (USAJ
87:6233 6237 (1990). ~Because of the high homology to the mouse CRABP-II, clone lf1.1 wæ designated as human CRABP-II. A third clone Af5.1 was partially sequenced andfound to be a shorter clone, with a sequancs identical to human CRABP-II. No CRA8P-I
clones were isolated from the skin fibroblast library. Two clones were isolated from the 25 skin library and ~ound to contain shorter inserts, one with a s~quence identical to human ORABP-II (As~1) and one with a sequence identical to human CRABP-II (ls3.1).
SEa ID NOS. 1 and 2 In the Sequence i isting represent the cDNA nucleoffde and predict0d arnino acid sequences of human CRABP-II. The ~anslaffon initiaffon sRe was assigned to the flrst methionine codon corresponding to nucleoffdes 9~101. An 30 open-reading frame of 138 amino acids was found, predicting a poiypeptide oi Mr 15,693.
The 3' untranslated region was tound to contain a poiyadenylaffon signai (Al~MA) and a poly(A) tract of 1 (lfl.1) to 25 (Af5.t). SEQ ll:~ NOS. 3 and 4 represent the cDNA
nucleotide and predicted animo acid sequence of CRABP-I resp~ctveiy.

~ 1 3 4 ~O 93/22331 P~/US93/03936 ~
- 14- ~
Comparison of amino ~cJd s~quences of CRABP
The amino acid sequence comparison of CRABP-I and CRABP-llis presented in Figure 2. Panel A is a comparison of the amino acid sequenees for human th) and mouse ~m) CRA8P. The predictsd amino acid seguences of mouse and human CRABP-I and 5 mouse CRABP-II were aligned with human CRABP-II. Dashes (--) represent identity to human CRABP-II. One gap as indicated by an asterisk (~) was introduced in the human and mouse C:RABP-I sequences for ma~dmum alignment. Panel B o~ Figure 2 is the sequence comparison of the Nl 12-terminal ends of human (h), mouse (m), rat (r), CRABP-I
and CRABP-II. Residuesdissimilarto human CRABP-II are boxed.
The amino acid sequsnce comparison presented in Figur~ 2, Panel A, reveals a 73.7% overall degree of identity between mouse CRABP-I (Stoner, C.M. et al., Cancer P~es.
49:1497-1504 (1989); Giguere, V. et al., PNAS (USA) 87:6233 6237 (1990)) and between human CRA8P-I and human CRABP-II. Human CRABP-I and mouse CRABP-I displayed an overall identny of 99.3%, with a single amino acid substit~ion (amino acid residu0 86, 15 Ala instead of Pro), while human CRABP-II and mouse CRABP-II were 93.5% identical.
Human and mouse CRABP-II displayed 9 amino acid dmerences with the following amino acids in the human sequence: residue 19 ~ Leu, 2~ - Val, 27 - Val, 29 ~ Leu, 48 - Gly, 68 - Val, 91 - Glu, 99 - Lys and 111 - Thr. In 6 of the 9 amino acid dfflerences seen bet Neen human CRABP-II and mouse CRABP-II (residues 19, 29, 48, 68, 91 and 111), the 20 human ssquence was idantical to CRABP-I.
The NH2-terminal amino acid sequences of twoCRABPs from neonatal rat (Bailey, J.S. et al., J. Blol. Chem. 263:932~9332 (1988)) and chicken embryos (Kitamoto, T. et al., Biochem Bioptrys. Res. Comm. 157:1302-1308 (1988) have been reported and were also compared to human CRABP-I and CRABP-II. As shown in Figure 2, Panel B, there were 25 no amino acid dfflarences between the NH~terminal regions of rat, chicken, mouse and human CRABP-I, whereas several differences appeared between the CRABP-lls.
SPECIFIC EXAMPiLE 2. iExpre~sion of Human CRABP
Exprcsslon of human CR~4~P In human skln In v~o Volunteers were treated with 0.1% RA cream or control treatments under occlusion30 for various time intervals. Keratome biopsies consisting mostly of epldermis (see Voorhees, J.~l. et al., Arch Dermatol. 105:695-701 (1972)) were obtained, and used to prepare total RNA which was subsequently analyzed by blot hybridkation. The autoradiograms of the RNA blots derived from thres individuais are shown in Figure 3.
Total RNA (40 ~g per lane) was hybridized against the human CRABP-II or cyclophilin 35 cDiYA probes as described above. Each volunteer was treated topically for 16 hours prior 2134550 ~ ;
~- WO 93~22331 ' PCr/US93/03936 ~ -- -15- f to biopsy as follows: ~a) no treatment, (b~ RA vehicle or (c) 0.1% RA cream in RA vehicle under occlus~on with plastic wrap. Mobilities of ribosomal RNAs are indicated to the left of the blots.
CRABP-II transcripts were detectable in untreated skin as well as skin traated with 5 vehicle, a shown in Figure 3. CRABP-II transcripts were marked~ (16.1-~old) and significantly (p<0.004, n=4) induced in RAtreated relative to untreatad skin. Similar, albeit less markad, inductions were observed after 4 days of RA treatment (8.3 ~ 2.~fold, n =
6). Induction did not occur after 4 hoursj but was evident after 12 hours of treatment. I
Consistent with our ability to ampl~y CRABP-I from human skin F~NA, faint CRABP-I probe 10 hybridization was observed in some but not ali blots ot human skin RNA samples alter prolonged autoradiographic exposure (data not presented in Figures). Howevar, CRABP-I
transcripts were usually undetectable under exposure conditions sufficiently sensitNa to detect single copy DNA sequences *om 10 ~^~g human genomic DNA. CRABP-I and CRABP-II cDNA probes datect~d distinct band pattems using genomic DNA digested with 15 BamHI, EcoRI, Hindlll and Pstl, demonstrating the specHicity of these probes under our hybridizaUon condiUons (data not presented in Figures).
Expresshn of CR4BP ln human fblobl2sts Treatment of human de~nal fibroblasts with RA resulted in a marked (approximately 1~fold) and significant (P ~0.04) induction of human CRABP-II mRNA after treatment of 20 five independent dermal fibroblast stralns with 3 x 1~7 or 3 x 10~ M RA for 24 or 48 hours as shown in Figure 4. Resuits shown are derived from nine independent experiments invo~ving five dermai fibroblast lines prepared from three individuais as described above and three diploid human lung fibroblast lines (LL47, CCD-1 8Lu, and CC~1 6Lu). RNA blot hybridizations (20 1~9 total R~lane) were quantitated by laser densitometry and 25 normalized to the control gene, cy_~ophilin, as described in Elder, J. T. et al., J. Invest.
Dermatol. 94:1~25 (1990). At confluence, medium was changed and cells were treated with RA dissoived in dimethyl suifoxide at the concentrations and for the times indicated beneath the Figure. Data are expressed as fold induction ~ SEM, relative to the average , of duplicate dishes treated with dimethyl suifoxide aione for 4 hours *p~0.06, **p~0.005.
The relevant autoradiographic bands from two representative experiments comparing ~ermal and lung fibroblasts are, shown in the inset. ~
As shown in Figure 4, CRABP-II mRNA was not slgnificantiy induced by PA in three~trains of human lung fibroblasts, suggesUng that ~Is response may be Ussua specific.
Induc~ion of human CRABP-II transcripts by RA was dos~dependent over the range of 3 36 x 10-1 to 3 x 1~7 M RA (data not presented in Figures). In contrast, CRAPP-I transcripts ~1~453~
WO 93/22331 ~ . PCr/US93/03936 ~--~
` -16- ~;
were undetectable in dermal and lung fibroblasts and were not induced by RA, whereas genomic DNA blots hybridked in parallel were positive.
Expression of CRd~P In human keratinocytes As shown in Figure 5, ::RABP-II mRNA was markedly and significantly inducad in 5 cultured adult human keratinocytes when cells reached confluence, and remained elevatsd in the postconfluent ~tate. Figure 5A summarizes the r~suits obtain~d for seven independent keratinocyte strains. Third passage norrnal aduit human keratinoc~tes were grown in KGM containing 0.15 mM CaCI2, and medium was chang~d every other day. At the indicatsd number of days pr~ or post~o~uence, total RNA was prepared and 10 analyzed for human CRA~P-II mRNA by blot hybridization and densitometry. Data are expressed as percent ma~amal expression for any given strain ot keratinocytes, as ths absolute level of human Ci~ABP-II mRNA was variable from strain to strain. Error bars represent ~ SEM. **p<.005, ***p<0.0005 relative to cells at 2 days preconFluence.
CRABP-II mRNA was also markedly induced in subconfluent cultures by raising the 15 caleium concentration in the medium from 0.15 mM to 2 mM, as illustrated ~or a r~presentative strain of keratinocytes in Figure 5B. At 2~30% confluence, the medium was changed to KGM or KGM containing 2 mM CaCI2 in the presence or absence of 3 x 10-9 M RA and maintained for the Indicatad number of days, with medium change every other day. Mobilities of 28S and 1 8S ribosomal RNAs are indicated to the left. Figure 5C shows ~0 the effects of prolonged treatment with high concentrations of RA on CRABP-II mRNA
levels. The same experiment described for Figure 5B was conducted above, except that cells were treated with or wlthout 3 x 1 o~6 M RA for 2 to 5 days, as indicaSed above the autoradiograrns. Mobilities of 28 and 18S ribosomal RNAs are indicated to the left.
Figures 5A and B show the results representative of four independent experiments.
2~ As shown in Figure 5B, treatment of keratinocytes for prolonged periods of time with low concentrations o~ RA (3 x 10-9 M) had no detectabls effect on CRABP-II mRNA
levels. Howevef, as shown in Figure 5C, prolonged treatment of subconfluent keratinocytes with high concentrations of ~A (3 x 10~ M) reducsd CRABP-II mRNA to undet~ctable l~v~ls. Hybridkation of the same blot against CRABP-I prior to its 30 hybridization against CRABP-II failed to detect CRABP-I transcripts, whereas genomic DNA
blots hybridized in parallel were positive (data not presented in Figures).
SPECIFIC EXAMPLE 3. Alternate Clonin~ Scheme The CRABP-II cDNA probe was cloned by PCR from retinoic acid (RA) treated human skin using the CRABP-II degenerate primers shown in Figure 1B, derNed frem35 mouse CRABP-II mRNA sequ~ncs. See Giguere, V. 0t al., PNAS ~VSAJ 87:6233 6237 - W09 . 21345~0 3/22331 PCI`/US93/03936 - 17 - ;
(1990). BamHI restriction sites were included in the forward and r~verss primers to aid in subsequent cloning.
RA-treated human skin was used asthe source of RNAfor r~verse transcri~tion and PCR amplification. Total RNA was extracted from epidermal ksratom~s as describ~d in 5 Elder, J.T. et al., J. Invest. Dermatol. 94:1~25 (1990), and reverse transcribed as described in Maniatis, T. et al., Molecular Cloning: A Labo~atory Manual, Cold Spring Harbor Laboratory, Cold Spring Harbor, NY (1982). During each PCR cycl~, samples were heated to 94C in 1 minute and maintained for 30 seconds, cooled to 5~C for primer annealing, and then heated in 1 minute duration to 72C and kept at th~t temperature for a further 30 10 seconds to allow template extension. The PCR product was further purified (Geneclean, BIO 101 Inc.), digested with BamHI and cloned into BamHJ digested PGEM 3Z plasmid (Promega inc., Madison, Wl). The PCR ~ragment was iniffally identified as CRABP-II by digestion with Pstl restriction enzyme that cleaved the DNA at the same position as that described in the mouse CRABP-II mRNA. The clone was s~quenced and found to 15 represent the major por~ion of the mature human analogue of the mouse CRABP-II. The cDNA probe was hybridized to human P~NA prepared from skin of human volun~eers topically ~reated with RA for 4 and 12 hours. Indwtion of CRABP-II mRNA was observed in skin which had been RA-treated for 12 hours, but not skin treatad for only 4 hours (data not illustrated in Figures).
20 SPECIFIC EXAMPLE 4. CRABP Mamn alian Expres~ion Vectors and Con~truc~
CRABP-I was excised from the Bluescript phagemid (Stratagene) cloning vector andinserted between the Xbal and BamHI sites of the eukalyotic expression vector pSVL
(Pharmacia) in forward ori~ntation (pSVLCRABP-I). CRABP-II was excised from the BiuescrTpt phagemid (Stratagene) cloning vector and inserted into the EcoRI site of the 25 eukaryotic expression vector pSG5 (Stratagene) in forward (pS~;5CRABP-II) or reverse oriontation (pSG5CRABP-llAS). The orien~ation of the insert was datermincd by restriction analysis.
CV-1 cells are grown in Dulbecco's modified Eagles medium (DMEM) containing 10% fetal caff serum. CV-1 are monkey kidney cells derived from the CV-1 cell line which 30 do not express T cell antigen. Th8 day beforfl transfection, cells are seeded on tissus culture dishes in DMEM containing 10% charcoal treated fetal calt serum (ChFCS). Cells are contransfected using the calcium phosphate co precipitation tachnique essentially as described in Rosenthal, N., Mc~7. Enzymol. 152:704 720 (1987) with CRABP-I (pSVLCRABP-I) or CRABP-II (pSGSCRABP-II) expression vector. 24 hours after transfection, cells are 35 t~ypsinked and suspended in medium, pelleted and washed once with 40mM Tris-C1, pH

~,~ . . . -213~ C~SJ
WQ 93/22331 ' pcr~lJss3/o3~36 l 18 - ~-7.6 containing 150 mM NaCI and 1 mM EDTA. A cytosolic fraction is prepared on cell Iysates by centrifugation at 100,000 x 9 for I hour.
SPECIFIC EXAMPLE 5. Bacterlal Production of Human CR~4BP
For CRABP-I, the nucleotide sequence of CRABP-I is changed at position 9 (T to 5 G) and 13 (C to T) ~o create a Stul site, using synth~tic oiigonucleotides and the polymerase chain r~action. For CRABP-II, the nucleotid~ sequ~nce of C:RABP-II ischanged at position 100 (T to G) and 104 (C to T) to create a Stul site, using syntheUc oligonucleotides and the polymerase chain reaction. The mutated CRABP cDNAs are then cut with Stul and ligated into the Stul and EcoRI sites of the bacterial expression vactor 10 pMAL-c (New Engiand Biolabs). Bacteria is transformed and th~ maltosa-binding protein (MBP)-C~ABP fusion protains are expressed in large guantities. The MBP-Cf~ABP fusion protein is purffied by amnity chromatography on any amylos~ column (New England Biolabs). CRABP lacking the first methionine is released from MBP by digestion with factor X~ (New England Biolabs) and purified from MBP by a second passage over an amylose 1 5 column.
Spectrofluorimatric methods are used to study the amnities and binding stoiclliometries ~f purified human CRABP-I and CRABP-II for a variety of iigands. The capacity of ligands to bind to CRABP-I or CRABP-II is assessed by monitoring their ability to queneh the native fluorescence of this protein.
2Q SPECIFIC EXAMPLE 6. Production of Antibodle~
Peptides correspondin~ to amino acids 94 to 104 in the CRABP-I and CRABP-II
pro~eins have been synthesized using the multiple antigenic peptide method as generally desbribed in Posnett, D.N. et al., Met~7. En~ymol. 178:73~746 (1989), eliminating the need for conjugation to a carrier protein. This region of the ~vo CRABPs has a low homology 25 and is most probably situated on the outside of the proteins based on hydrophobicity and surface probability calculations.
The peptides were injectesi ~to chickens for production of egg IgY. llle resulting antibodies, if not specfflc, are adsorbed to peptides (CRABP-I antibodies to CRABP-II
peptides and the reverse) irnmobilked on a sapharose gel to enhance specificity. Once 30 monospecific antibodies are obtained, the expression and regulation of CRABP-I and Ci~ABP-II proteins In human skin and skin cells is examined quanUtativeiy by Western blot anaiysis and seml~uantitativeiy by immunocytochemistry. The pattem o~ expression and regu!ation of Ci~ABP-II (and CPABP-I) by PA in normal and psoriatic skin will provide insight into the function of CRABPs.

WO 93/22331 2 1 ~ 4 5 S ~ P~/US93/03936 Antibodies to ::RABP-I and CRABP-II, or binding fragments (e.g. F~(ab)) thereof,and the purified proteins obtained as deseribed above can be us~d to monitor levels of CRABP-I and CRABP-II in normal and pathologicai states by using immunological techniques known to those skilled in the art. For exampls, patient skin tissue can be assayed with antibody specific for human C:~ABP-I or CRABP-II to determine the presence and levels of CRABP by ELISA, Western blot analysis or immunochemistly essentially as desc!ibed in Busch, C. et al., Meth. Enzymol. 189:316 324 (1990).
SPECIFIC EXAMPLE 7. Report~r As~ay Sy~tem CV-1 cells ar~ grown in Dulbecco's mod~ied eagies medium ~DMEM) containing 10% fetal calf serum. The day before transfection, cells are seeded on tissue cultured dishes in DMEM containing 10% charcoal treated fetal calf serum (CHFCS). Cells are cotransfectsd using the calcium phospha~e co-precipitation technique with 0.6 mg of human retinoic acid receptor (hRAR) expression vectors (hRARaO, hRAR~O or hRARAO), a reporter plasmid and a ~alactosidase expression vec~or (pcH110, Pharrnacia) used as an internal control to normalize tor variaHons in transfection emciency essentially as dascribed in Astrom, A. et al., B~ochem. Biophys. Res, Comm. 173:33~345 (1990). Cells are also cotransfscted wiUl CRABP-i (pSVLCRABP-I) or CRABP-II (pSG~CRABP-II) expression vectors of pSVL rPharmacia) as a control. The reporter plasmid (TRE)~-CAT
is constructed by ligating synthetic oli~onucleotides encoding three palindromic thyroid hormone rssponsiva elements (TRE) ((TCAGGTCATGACCTGA)3) flanked by Hindlll and BamHI sites on the 5' and 3' ends respectively and cloned into the Hindlll - BamHI cloning sites of the plasmid pBLCATæ 24 hours after transfection, cells are washed once with DMEM, and medium (DMEM, 10%ChFCS) containing different concentrations of ligandsars added to the cells~ 24 hours later the cells are 'aypsinized and sucpended in medium, pelleted and washed once with 40mM Tris-CI, pH 7.6 containing 1 50mM NaCI and 1 mM
EDTA. C:ell Iysates are prepared by Ulree consecutive freeze-thaw cycles and ~-galactosidase and CAT-activities are determined by a xylene extraction method. The effect of coexpression of CRAi3P-I or CRABP-II on RAR transcriptional activaUon for dfflerent ligands can now be deterrnined as described below.
An aiternative reporter assay in which a recombinant adenovirus system is used to coinfect cells in Gulture may aiso be employed to measure transcripUonal activation by retinoids essentially as described in Shih, E~ et ai., MoL Endocrinol. 5:300 309 (1991). In - such a system, two mutually dependent viruses, one containing a receptor transcripUon unit and the second containing a gene responsive element are coinfected into receptor-35 deficient cells such as CV-1. For example, hvo mutualiy dependent adenoviruses, one w0~3~2 ~1345~0 2331 ` : PCI'/US93~03~36 ~^
- ` - 20 -containing a human glucocorticoid receptor transcription unit and the other a glucocorticoid responsive element linked to the firefly luciferase gane, or one con~aining rat thyroid hormone receptor a and the other the luciferase gene, can be utilized in the practice of the assay. Hormons-induc2d transcription is then quantitated a~ter infection 5 from cells coinfected with the complementary virus pair.
SPECIFIC EXAMPLE 8. Llgand Bind~n~A~say~ ;
Since the amino-acid homology between CRABP-I and CRAB~-II Ts on~ 77%, it is important to determine the ligand binding properties of thas~ two protelns. There are ssveral reports on the lack of correlation betwe0n biological activity and affini~y of retinoids 10 to CRABP. Darmon, M. et al., Skin Pharmacol. 1:161-175 (1988). Most of these studies have used extracts from rat testis containing mostly rat CRABP-I as the source of CRABP.
In contrast, we are expressing the CRABP-I and CRABP-II cDNAs can now be expressed in mammalian cells.
Availability of cloned human CRABP-I and 11 thus improves previous systems for 15 identification of tissue-specific ligands. This also makes identification of ligands interacting with CRABP possible, in addition to nuclear receptors. Ugands for testing include, but are not limited to didehydr~RA, RA and its metabolites, 4hydroxy-RA, 4 oxo-RA, and 5,~
epoxy-RA æ well as compounds previous reported not to bind to CRABP, e.g. CD 394.
Darmon, M. et al., Shn Phsn77acol. 1:161-175 (1988). Availabilty o~ CRA8P amin~acid 20 sequences or proteins provide the msans to select more tissue-specific drugs for repair of photoaging skin, psoriasis, acne, skin cancer, leukemia, diseases of keratinization, osteoperosis, rheumatoid sclerosis, and other conditions.
Soluble protein from CV-1 cells transfected with CRABP-I (pSVLCRABP-I) or CRABP-II (pSG5CRABP-II) expression vectors obtained as described above is incubated 25 with [3H3 retinoic acid at 4C overnight. Free ligand is separat~d from bound ligand by size fractionation on a GF 250 column (DuPont Pharrnaceuticals) connected to a FPLC
system (i'harmacia). The amount of specific binding is determined by incubating samples with an excess of cold retinoic acid. The amnity of CRABP-II for retinoic acid can be determined by titrating with increasing amounts of i~3H~ Retinoic acid.
The relative amnity for other ligands is deterrnined in competition assays in the same system. A fixed amount of CRABP-I or i~H] reffnoic acid is Incubated with soluble proteln from CV-1 cells expresslng CRABP-II together with increasing amounts of othsr posslble ligands. -WO 93/22331 2~ 1 ~ 4 ~ ~ O PCI~US93/0393~ i `
SPEC:IFIC EXAMPLE 9. Hybridkation Assay Availability of cloned human CRABP-I and CRABP-II cDNAs makes it possible to determine the relative expression of Ulese two genes in norrnal and pathological states.
For exampla, RNA i5 isolated from tissue biopsies and cultured cells by guanidinium 5 isothiocyanate Iysis and ultrac~ntri~gation as previously described in Elder, et al., J.
Invest. Dermatol. 94:1~25 t1990). Equal quantities of total RNA can be separated on 1%
formaldehyde-agarose gels and transferred to nylon membranes. Aiter baking for 2 hours at 80C, filters can be hybridized to agarose-purffled t::RABP-I or CRABP-II cDNAs labeled by random priming. The amount of CRABP mRNA in a tissue can be detsrmined after 10 quanfftation of autoradiograms by laser-densitometry as described (Elder, et al., J. Invest.
Derma~ol. 94:19-25). It will be appreciated that CRABP^I and ll oligonucleotides (typically ~15 residues long) may also be utilized as probes in hybridization assays if o~ a sufficient length to bind complementary sequences.
SPECIFIC EXAMPLE 10. Examination of Function of Human CRAE~P
15 Receptor assay The most important and difficult issue to resolve regarding the CRABPs is their function. It has been suggested that CPABP transports RA to the nucleus ~rakase, S. et al.,Arch Biochem. f~iophys. 247:328 334 (1986)) or that CRABP remains in the cytoplasm, thereby prevenUng RA from movinQ to the nucleus. Maden, M. et al., Nature 335:733 7~5 20 (1988). One way of addressing tnis issue is to express increasing amounts of CRABP-I
and CRABP-II in CV-1 cells tog~ther with the RARs and a reporter gene containing a retinoic acid responsive element, as described above. if CRABP transports RA to the nucleus an enhancement ot reporter gene activity at low concentrations of RA may be seen. On the other hand, H CRABP sequesters RA a decrease in response will be seen.
25 Overexpression CRABP-I and CRABP-llis overexpressed in fibroblasts and keratinocytes. Cells will be transfected with CRABP expression vectors (pSVLCRABP-I or pSG5CRABP-II) as described above, using Upofectin ~BRL) essentialiy as described by the manufacturer In fibroblasts the effect of CRABP overèxpression on induction of til~ RAR-~ gen~ by i-~ is 30 studied. It has been shown that RAR-i3 mRNA is induced by RA in skin fibroblasts, and it is known that this gene Is directly regulated by the RARs. See DeThe, H. et ai., NahJre 343:177-180 (19gO). The i~AR-~ gene acts as an ~endogenous reporter gene~ for RAR
induction. In prslHerating keratinocytes the effect of overexpr3ssion of CRABPs on markers of dmerentiation is studied.

~,, .. ,. ~ . .. . . . . . . . . . . . .

wo g3/Z23312 1 3 4 5 S '~ ` PCI`/US93tO3936,~

TranslaUon blocking Translation of CRABP-II is blocked in fibroblasts and keratinocytes by transfecting the cells with an axpression vector construct having th~ CRABP-II cDNA in a reverse orientation. This produc~s an anti-sense mRNA that jQ able to hybridize to the 5 endogenous CRABP-II mP~NA, thareby blo~king translation. The effect of antisense expression in fibroblasts and keratinoc~tes is studied the same way as describad for overexpression.
Effsct of CRAE~P on RA metabolJsm The involvement of the cellular retinol binding protein (CRBP) in delivering retinol 10 to the appropriate m3tabolic enzyme in rat liver microsomes has been described. Ong, D.E. et al., J. Biol. Chem. 263:578~5796 (1988). Since th~ CRABPs belong to the same family of binding protein as CP~BP, it is possible that they also are involved in the metabolism of their ligands. RA Ts metabolized by the cytochrome P~50 monooxygenase system present in the endoplasmic reticulum. Bossche, H. et al., Skin Pharrnacol. 1:17~
15 185 (1988). CRABP may function by increasing the interaction of RA with cytochrome P-450. Another possibility is that ~ is not the natural ligand for CRABP-II, but rather one of the metabolites. If CRABP-II for example, bind-~ ~hydroxy-RA (as deterrnined by ligand-binding studies) with a higher affinity than RA, it is possible that CRABP-II increases cytochrome P~50 mediated metabolism by decreasing produc~ concentraUon. Whether 20 the CRABPS are involved in RA metabolism or not is tested by incubating skin microsomes with increasin3 amounts of RA and of expressed and purified CRABP-I or CRABP-II. The effects of CRABPs on RA metabolism is assayed by HPLC.
SPEC1FiC EXAMPLE ~1. Clonlng and Sequencing of Human CRABP-II Gene and Transcription Studies MATERIALS AND METHODS
ClonJng of U~e human CRA~P-~I gene To isolate the CRABP-II gene, approximateiy 6 x 105 recombinants lrom a human placenta genomic DNA libraly in ~ ambda FiX ll (Stratagene Inc., LaJolla, CA) was screened using the human CRABP-10 cDNA as a probe. Astrom, A. et al., J. Biol. Chem. 266:17662-17666 (1991). One positive clone was isolated, purified and restriction mapp~d. Parts ot the insert were subcloned into Bluescript phagemid SK (Stratagene Inc., L~Jol.a, CA). The sequence of the gene was determined on both strands after subcloning into M13 vectors by dideoxy chain terrninaUon using modified T7 poiymerase (Sequenase, U.S. Biochemlcal Corp.) and synthetic oligonucleoUdes. Sanger, F. et al., P~S ~US4J 74:5463 5467 (1~77).
35 Exon positions were determined by restriction mapping and sequencing.

- wo93/22331 2134550 :~cr/us93/03936 t Cell culture Human dermal fibroblast cultures were prepared from punch biopsies of buttock skin and propagated in Dulbecco's modffled Eagle's medium containing 10% calf serum.
Harper, R.A. et al., Science 204:52~527 (1979).
5 Northem analysls of mRNA
RNA was isolated from cuitured human skin fibroblasts by guanidinium isothiocyanata Iysis and uitracentrifugation as previously described. Elder, J.T. ~t al., J.
Invest. Dermatol 94:1~25 ~1990). RNA concentrations were determined by absorbance at 260 nm and equal quantities of total RNA were electrophoretically separatad in 1%
10 formaldehyd~agaross gels containing 0.5~g/ml ethidium bromide. The RNA was transferred to nylon membranes (Zeta-Probe, BioRad, Richmond, CA) as described. Elder, J.T. et al., J. Inv~st. Dermatol 94:1~25 (1990); Maniatis, T. et al., Molecular Cloning: A
LaboratoryMamJal (Cold Spring Harbor Laboratory, Cold Spring Harbor, NY) (1982). Filters were basked 2 hours at 80C in vacuo, then prehybridized for 24 hours at 42C in 50%
15 formamide, 5 x SSC (1 x SSC = 150mM NaCI, 15 Mm sodium citra~e), 50 mM sodiumphosphate, pH 7.0, 1 x Denhardt's solution, 250~g/ml yeast tRNA, 100~g/ml sonicated herring sperm DNA and 1% SDS. Hybridization was carried out for 1~24 hours at 42C
in the same bu~fen Filters were washec once in 0.2 x SSC, 0.1% SDS at room temperature, then twice for 20 min in 0.2 x SSC, 0.1% SDS at 56C, and final~ once for 20 20 min in 0.1 x SSC, 0.1% SDS at 56C. Autoradiography was performed using intensifying screens at -70C. Filters were stripped by boiling 2 x 10 min in 0.1 x SSC, 0.1% SDS.
Hybridization probes were prepared by random primTng (Boeringsr Mannheim) ot purified insert fragments from human CRA8P-II cDNA tlfl .1 ) rat cyclophilin and RAR-y, Astrom, A.
et al., J. E~lol. Chem. 266:17662-17666 (1991); Dani~lson, P.E. et al., DNA 7:261-267; Elder, 25 J.T. et al., J. Invest Derrnatol 96:425 433 (1991). Quantitation of mRNA levels were performed using a phosphorimagar (Molecular Dynamics).
Primer ~xtensJon analy~is RNA was isolated from untreated and RA treated skin fibroblasts as described.
Astrom, A. et ai., J. Biol. Chem. 266:17662-17666 (1991). A synthetic oligonucleotide 5' 30 CTAGGCTGGAGCACTGGACACTGTC 3' complementary to position 80-104 in the gene was used as an ext~nsion primer. 1 O~g of totai RNA was heated for 10 minutes at 70C
togsther w~th 32p 5~ end-labeled primer. The mixture was allow~d to cool to 30C over - 30 minutes and then kept at 30C for an additional 30 minutes. To the mixtures were added (final concentration) 50mM Tris-HCI (pH 8.3), 75 mM KCI and 3mM MgCI2, 5mM35 DDT and 250 mM of dATP, dCTP, dGTP, drrP in 40~1 total volume. The reactions were 21345S~
WO 93/22331 . ~ ` PCI'/US93/03936 .---- 24 - t, _, started by the addition of 200 u Superscript RNase H (GIBCO, Bethesda Research Laboratories) and incubated for 90 minutes at 42C. Samples were phenol/chloroform-extracted, ethanol-precipitated, resuspended in formamide dye and after hsating to 75C
~or 3 minutes, separated on a 6% saquencing gel.
5 Transcr}ptJonal analysJs Cuitured human skin hbroblasts were grown to confluency on 150 mm tissue cuiture dishes as described. Astrom, A. et al., J. Biol. Chem. 266:17662-17666 (1991).
Cslls wsre then treated for various time-points with 1 ~i RA in pr~warm0d and equilibrat~d Dulbecco's modified Eagle's medium containing 10% fstal calf serum. Nuciei for each 10 time-point were isolated from four dishes after incubation in Iysis buffer containing 0.596 NP40 as described. Greenberg, M.E. et al., Natur~ 311:433 438 (19~4). Nuclear run-on experiments were performed with la-32P] UTP (DuPont-New England Nuclear, 800 Ci/mmol) as described. Antras, J. et al., J. Biol. Chem. 266:1157-1161 ~1991). Equal amounts of radioactivity (0.5-1 x 107 cpm) were hybridized to nHrocellulose fiiters 15 containing 5 ~9 of each plasmid. After hybridization for 72 hours at 42C, the fiiters were washed twice with 2 x SSC at 37C for 15 minutes and treated for 30 minutes at 3PC in 2 x SSC containing 5 ~g/ml RNase A. The fiiters were then washed twi,ce for 15 minutes in 2 x SSC, 0.5% SDS at 42C and once for 30 minutes in 0.5 x SSC, 0.5% SDS at 42C.
A final wash was carried out in 0.1 x SSC, 0.1% SDS for 30 rninutes at 55C. The amount 20 of radioactivity present in each slot was determined using a phosphorimager (Molecular Dynamics) after over-night exposure and autoradiograms were exposed for ~5 days at -70C with intensifying screens.
RESULTS
Clon~ng and char~cter~zaUon of th~ human CR4BP-II gen~
Using the human CRABP-II cDNA as a probe, one bacteriophage lambda clonewas isolated (Aæ1) from a human placenta genomic library. Restriction mapping and sequencing ravealed that this clone contained the entire CRABP-II gene. A restriction map of 12.1 is shown in Figure 7. The gene is composed of four exons, interrupted by one . . , !
large and two small introns. The overall ske of the gene is approximately 6 kb.
30 DNA SQquenc~ of the human CR~4BP-II gene To characterke the CRABP-II g~ne, fragments spanning the ~ntire gene, except forthe first intron were subcloned into M13 vectors for sequence analysis. The nucleotide sequence and the deduced amino-acid sequence is presented in Figure 8. The first three -sxons are small and range in size from 117 to 207 bp, with the fourth exon being the 35 largest at 466 bp as d~monstrated in Figurs 8. The exon sequences were found to be _, . .. .... .. . . . ... . . . . . .. . . . .... . . . ..

- - wo 93~22331 , 2 1 3 ~ 5 ~ O PCr/US93/03936 identical to the published cDNA sequence. Astrom, A. et al., J. Biol. Chem. 266:17662-17666 (1931). All splice junctions contained the expected GT splice donor and AT splice acceptor.
Analysfs of the 5 ' end of U~e human CRABP II gene .
The 5~ bounda~ ot the first exon was determined by p~mer extension ana~sis.
Using an oligonucleotide primer complimentary to position 8~104, a predorninant reaction product was ident fied when mRNA from untrea~ed cultured human skin fibroblasts were used, see Figure 9. Comparison of the ex~ension product with sequencing reactions 7 originating from the same prim~r indicated that the major transcription initiation s~e should be assigned to the A r~sidue 137 bp upsteam of ~he ATG. When ~broblasts were treated with 1 ~M RA for 24 hours an increase in initiation at the A residu3 could ~e seen. In addition a second initiation product with the same intensi~y could be seen at ~
Sequence analysis of the upstream region revealed a TATA box (TAT,MA) at -31 and severai potential regulatory elements including ~wo potential AP2 binding si~es at -631 and ~02, and one potential SP1 site at ~9 as demonstratsd in Figure 8. Mitchell, P.J. et al., Cell 50:847-861 (1987);, Kadonaga, J. T. et al., Jrends 3ichem. Sci. 11:2~23 (1986).
In add~ion hvo sequences can be found that exactly match a binding site for the early growth response gene ~ox-24 (Egr-1, zi~26B or NGFI-A) at -579 and -116. L0maire, P. et al., Mol. CelL Biol. 10:3456~67 (1~90). In addition a direct repeat spaced by one bp (AGl~CAgGGrrCA) can be found in the upstream region of the gene at 454, with homology to the retinoic acid responsive element found in the RAR-~2 gene. l~mesono, ~ et al., Cell 65:12551266.
Effe~ts of RA on c~q~P-n tr~nscrfpUon rat~s and mRNA levels To determine whsther RA induction of CRABP-II mRNA seen in cultured human skin 25 fibroblasts was dus to inaeased transcription, nuclsar run~n assays were performed using nuclei isolated *om cultured human skin fibroblasts traated wKh RA for various periods of Ume. A~ seen in Figure 10, RA caused a rapid transient increase in CRABP-II
transcription, peaking at 2 hours and being almost back to control levels at 6 hours.
However, there was no change u~ RAR~ and cyclophilin transcripUon, two genes known 30 to be unaffected by RA in skin fibroblasts as demonstrated by Figur~ 10. Edler, J.T. et al., .~. Invest. Dennatol. 96:425433 (1991). QuantRaffon and normallzation to cyclophilin shows U~at CRABP-II transcription is induced approximately ~told aner 2 hours of RA
treatment. See Figure 11.
This increase in transcription can account tor the ~fold induction of CRABP-II
35 mRNA after 24 hours of treatment, as seen in Figure 11. Induction of CRABP-II mRNA can i S f~
WO 93/~2331 ` PCI`/US93/03936 ~
- 26 - ~ , be detected as early as 2 hours a~ter addition of RA to cGlls, leveling out between 6 to 24 hours of treatment. However, a smaller increase can be ssen between 24 and 48 hours.
To determine whether RA is required to be present in the medium for maintaining the induced CRABP-II message levels after the transcription has occurred, human skin5 fibroblasts were treated with RA for 12 hours, after which the cells were washed two times and replaced wHh medium without any addition of RA. As can be seen in Figure 11,removal of RA from ths medium causes a decline of the CRABP-II messaga compared to cells traated with RA for the entire experiment.
Effccts of cyclohexlmlde or actlnomycln 1~ on the fnducUon o~ CRABP-U message Experiments were per~ormed to determine whether the addition of inhibitor-e of transcription or translation could inhibit the increase of GRABP-II mRNA produced by RA.
Cycloheximide (10 ~g/ml) or actinomycin D (2 ~uglm!) were added simultaneously with RA
M) and CRABP-II levels were determined a~tsr 2 hours. As can be seen In Figure 12, cycloheximide down-r~gulated basal CRABP-II exprassion and blocked the RA induction.
15 Cycloheximide had no effect on RAR~ expression or RA induction of RAR-,~ ~Figure 12), as previously reported. Nenfl,~. et al., Cell Gr~wth Dffl 1:53~542 (1990); DeThe, H. et al., FM80 J. 8:429 433. Actinomycin D, had no effect on basal levels of CRABP-IImessage, but complebly blocked induction by RA. As controls, RAR~ was found to be down-regulated by actinomycin D, and RA induction of RAR-,B was compl~tely blockad in 20 agreement with previous reports. Nervi, C. et al., Cell Growth Diff. 1:53~542 (19~0);
DeThe, H. et al., EMBO J. 8:429 433`(1989).
Effects of cyclohe)dmlde on CRABP-II transcrlpUon rates Since cyloheximide blocked RA induction of CRABP-II mRNA, run-on experiments were performed to sae if protein synthesis was required for the induction of transcription.
25 As can be seen in Figure 13, addition of Cycloheximide at time 0 completaiy blocked RA
induction of Ci~ABP-II g~ne transcripUon.
The foregoing discussion discloses and describes mereiy exemplary embodiments of the present invention. One skilled in the art will readily recognize trom such discussion, and from the accompanying drawings and claims, that various changes, mod-ffications and 30 variations can be made therein without departing from the spirit and scope of the invention as defined in the tollowing claims.
All publicatlons and applications cited herein are Incorporated by reference.

2134~S~ i - WO 93/22331 PCI`~US~3/03936 SEQUENCE LI5TIING .
~1~ GENERAL INFORMATION:
(i3 APPLICANT: Yoorhees, ~ohn J~
Astrom, Anders .
Pattersson, Ul ri ka Tavakkol, Ami r (ii) TITLE OF INVENTION: HUMAN CRABP-I A~D CRABP-II

(iii) NUMBER OF SEQUENCFS: 6 1:

(iv) CORRESPONDENCE ADDRESS:
tA) ADDRESSEE: Harness, Dickey & Pierce (B3 STREET: PD Box 828 (C) CITY: Bl oomfi el d Hi l l s (D) STATE: Michigan (E) COUNTRY: United States of Anerica (~) ZIP: 48013 ~v) CDMPUTER READABLE FORM:
(A) MEDIUM TYPE: Floppy disk ~B) COMPUTER: IBM PC compatible (C) OPERATING SYSTEM: PC-DOS/MS-DOS
(D) SOFTWARE: PatentIn Release ~1.0, Version ~1.25 (vi) CURRENT APPLICATION DATA:
(A) APPLICATION NUMBER:
(B) FILING DATE:
(C) CLASSIFICATION:

~ J v w o s3/2z331 ~ P ~ /US93/0393 (viii~ ATTORNEY/AGENT INFORMATION:
(A) NAME: Lewak, Anna M.
(B) REGISTRATION NUMBER: 33,006 (C) REFERENGE/DOCKET NUMBER: 211500676POB

(ix) TELECOMMUNICATION INFORMATION:
(A) TELEPHONE: (313) 641-1600 (8) TELFFAX: (313) 641-0270 (C) TELEX: 287637 (2) INFORMATION FOR SEQ ID NO:1:
(i) SEQUENCE CHA~ACTERISTICS:
(A3 LENGTH: 924 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: double (D) TOPOLOGY: linear (ii) MOLECULE TYPE: cDNA to mRNA

(iii) HYPOTHETICAL: NO

(iv~ ANTI~SENSE: NO

(vi) ORIGINAL 50URCE:
~A~ ORGANISM: Homo sapien (F) TISSUE TYPE: skin (G) CELL TYPE: fibroblast (vii) IMMEDIATE SOURCE:
(A3 LIBRARY: HUMAN SKIN FIBROBLAST LAMBDA GT11 (B) CLONE: LAMBDA F1.1 ,~ WO93/~2331 213~SO PCI'/US93/03936 ( i x) FEATURE:
~A) NAME/KEY: CDS
(B) LOCATION: 99..515 (D) OTHER INFORMATION: /codon start~ 99/citation- ( El] ) (ix) FEATURE:
(A) NAME/KEY: polyA_site (B~ LOCATION: 924 ¦
(D) OTHER INFORMATION: /citation= ([1]) (ix) FEATURE: -(A) NAME/KEY: 5'UTR
~8~ LOCATION: 1..98 (D~ OTHER INFORMATION: /citation~

(ix~ FEATURE:
(A) NAME/KEY: 3'UTR
(8~ LOCATION: 516..924 (D) OTHER INFORMATION: /citation- ([1]) (ix) FEATURE:
(A) NAME/KEY: polyA signal (B) LOCATION: 911..916 (D) OTHER INFORMATIO~: /citation~ (~1]) (ix) FEATURE:
(A) NAME/KEY: terminator (B) LOCATION: 513..515 (D) OTHER INFORMATION: /citations (tl]) WO 93/22331 2 13 ~ 3 ~ ~ . P~/US93~03936 k--~x) PUBLICATION INFORMATION:
(A) AUTHORS: Astrom, Anders . .
Tavakkol, Ani r Pettersson, Ul ri ka Cromi e, Matthe~
Elder, J~mes T.
Voorhees, John J.
(B) TITLE: Molecular Cloning of Two Human Cell~lar Reti noi c Aci d- Protei ns ~CRABP) (C) JûURNAL: J. Biol. Chem.
(G) DATE: 1991 (I() RELEVANT RESIDUES IN SEQ ID NO:1: FROM 1 TO 924 (x) PUBLICATION INFORMATION:
(A) AUTHORS: Astrom, Anders Tavakkol, Ani r ~ Elder, James T.
Pettersson, Ulrika Cromie, Matthew ~- Voorhees, John J.
(B) TITLE: Cloning of CRABPII cDNA from Human Skin:
Retinoic Acid Induces Expression of CRABPII but Not CRABPI in Human Skin in Vivo and in Dermal but Not Lung Fibroblasts in Vitro (C) JOURNAL: J. Invest. Dermatol.
(D) VOLUME: 96 (F~ PAGES: 547-547 (G) DATE: April-1991 (xi) SEQUENCE DESCRIPTION: SEQ ID NO:l:

GAGACACm CTACATCMA ACCTCCACCA CCGTGCGCAC CACAGAGATT MCTTCMGG 300 TTGGGGAGGA GmGAGGAG CAGACTGTGG ATGGGAGGCC CTGTMGAGC CTGGTGAAAT 360 213~0 - Wo 93/22331 PCr/VS93/03936 GGGAGAGTGA GMTAMATG GTCT6TGAGC AGMGCTCCT GMGGGAGA~ GGCCCCMGA 420 M GCCCACCA CTG~CCATGC TCACCGCCCT GCTTCACTGC CCCCTCCGTC CCACCCCCTC 600 CAGGGTCTTG C m C m GA CCTCTTCTCT tCTCCCCTAC AGC M CAAA6 AGGAAT6GCT 720 GACTGGCCCT CTAGCTTCTA CCC m ~TCC CTGTAGCCTA TACAG m AG AATArrTATT 900 (2) INFORMATION FOR SEQ ID NO:2:
(i~ SEQUENCE CHARACTERISTICS:
(A) LENGTH: 138 amino acids (B) TYPE: amino acid (C) STRANDEDNESS: si ngl e (D) TOPOLOGY: linear (ii) MOLECULE TYPE: peptide (iii) HYPOTHETICAL: YES

(vi) ORIGINAL SOURCE:
(A) ORGANISM: homo sapien (x) PUBLICATION INFORMATION:
(A) AUTHORS: Astrom, Anders Tavakkol, Amir Pettersson, Ulrika Cromie, Matthew Elder, James T.
Voorhees, John J.

J
wos3/22331 :: : pcrJuss3/o393b (B) TITLE: Mol eeul ar Cl oni ng of Two Human Cel l ul ar Retinoic Acid-Proteins (CRABP~
(C) JOURNAL: J. Biol. Chem.
(G~ DATE: 1991 (K) RELEYANT RESIDUES IN SEQ ID NO:2: FRGM 1 TO 138 ~x) PU~LICATION INFORMATION:
(A) AUTHORS: Astrom, Anders Tavakkol~ Amir Elder, James T.
Pettersson, Ulrika Cromie, Matthew Voorhees, 30hn J.
(B~ TITLE: Cloning of CRABPII eDNA from Human Sktn:
Retinoic Aeid Induces Expression of CF~BPII but Not CRABPI in Human Skin in Vivo and in Dermal but Not Lung Fi brobl asts i n Vi tro (C) JOURNAL: J. Invest. Dermatol.
(D) VOLUME: 96 ~F) PAGES: 547-547 - (G) DATE: April-1991 txi) SEQUENCE DESCRIPTION: SEQ ID NO:2:
M~t Pro Asn Phe Ser Gly Asn Trp Lys Ile Ile Arg Ser Glu Asn Phe Glu Glu Leu Leu Lys Yal Leu l;ly Val Asn Val Met Leu Arg Lys Ile Ala Val Ala Ala Ala Ser Lys Pro Ala Val Glu Ile Lys Gln Glu Gly 35 ~ qO 45 Asp Thr Phe Tyr Ile Lys Thr Ser Thr Thr Val Arg Thr Thr Glu Ile Asn Phe Lys Val Gly Glu Glu Phe Glu Glu Gln Thr Val Asp Gly Arg Pro Cys Lys Ser Leu Val Lys Trp Glu Ser Glu Asn Lys Met Val Cys 8~ ~0 95 Glu Gln Lys Leu Leu Lys Gly Glu Gly Pro Lys Thr Ser Trp Thr Arg 213~Sg~
- ` WO 93~;!2331 PCI/US93/03936 Glu Leu Thr Asn Asp 6ly Glu Leu Ile Leu Thr Met Thr Ala Asp Asp 115 12~ 125 Val Val Cys Thr Arg Val Tyr Val Arg Glu (2) INFDRMATION FOR SEQ ID ~0:3:
(i3 SEQUENCE CHARACTERISTICS: !
(A) LENGTH: 525 base pairs (B) TYP: nucleic acid (C) STRANDEDNESS: double (D) TOPOLOGY: linear (ii) MOLECULE TYPE: cDNA to mRNA

(iii) HYPOTHETICAL: NO

(iv) ANTI-SENSE: NO

(vi) ORIGINAL SOURCE:
(A) ORGANISM: Homo Sapien (F) TISSUE TYPE: skin (G) CELL TYPE: fibroblast :

(vii) IMMEDIATE SOURCE:
(A) LIBRARY: human skin Lambda ZapII
: (8) CLONE: la~bda s3.1 (ix) FEATURE:
(A) NAME/KEY: CDS
(~) LOCATION: 8..418 (D) OTHER INFORMATION: /codon_start~ 8 /citatlon- (tl]) ~13~3~ '{
WO~3/2~331 PCl'/US93/~)3936 ( i X) FEATURE ~
(A) NAME/KEY: 5 ' UTR
(B) LOCATION: 1. . 7 (D) OTHER INFORMATION: /ci tati on~

( i x) FEATURE:
(A) NAME/KEY: 3 ' UTR
~B) LOCATION: 419..525 (D) OTHER INFORMATION: /citation~ (~1]) (i x) FEATURE:
(A) NAME/KEY: termi nator (B) LOCATION: 419 . . 421 (D) OTHER INFORMATION: /citation~

(x) PUBLICATION INFORMATION:
(A) AUTHORS: Astrom, Anders Tavakkol, Amir Pettersson, Ulrika Cromie, Matthew Elder, ~ames T.
Voorhees, John J.
- (B) TITLE: Molecular Cloning of Two Human Cellular Reti noi c Aci d-Protei ns (CRABP) (C) JOURNAL: J. Biol. Chem.
~G~ DATE: 1991 (K) RELEVANT RESIDUES IN SEQ ID NO:3: FROM 1 TO 525 (x) PVBLICATION INFORMATION: ¦
~A) AUTHORS: Astro~, Anders Tavakkol, Amir Elder, James T.
Pettersson, Ul ri ka Cromie, Matthew Voorhees, John J.

w o 93/22331 - ~2 1 3 4 ~ 5 0 P~/uss3/03s36 ~B) TITLE: C.loning of DRABPII cDNA from Human Skin:
Retinoic Acid Induces Expression of CRABPII but Not CRABPI in Human Skin in ViYo and in Dennal but Not Lung Fi brobl asts i n Vi tro (C) JOURNAL: 3. Invest. Denmatol.
(D) VOLUME: 96 (F) PAGES: 547-547 (G) DATE: April-1991 (xi~ SEQUENCE DESCRIPTION: SEQ ID NO:3:
TGCCACCATG CCCMCTTCG CCGGCACCTG 6MGATGCGC AGCAGCGAGA AmCGACGA 60 GCTGCTCMG GBACTGGGTG TGMCGCCAT GCTGAGGMG GT6GCCGTAG cGGclrGcGTc 120 CMGCCGCAC GTGGAGATCC GCCAGGACGG GGATCAGl~C TACATCMGA CATCCACCAC 180 GGTGCGCACC ACTGAGATCA ACTTCMGGT CGGAGMGGC mGAGGAGG AGACCGTGGA 240 CGGACGCMG T6CAEGAGTT TAGCCACTTG 6GA6MTGAG MCMGATCC ACTGI:ACGCA 300 ACTTATCCTG ACGmGGCG CCGATGACGT 6GTCTGCACC AGMmATG TCCGA6AGTG 420 ATAGT~CT6A 6CTGCCAGTG GACCGCCCTT TTCCCCTACC MTAT 525 (2) INFORMATION FOR SEQ ID NO:4 (i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 137 amino acids (B) TYPE- amino acid ~C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: peptide (iii) HYPOTHETICAL: YES

~1~ 4a5 '~7 W(~ 93/22331 PCl`/U!~93~03936 ~v;) ORIGINAL SOURCE:
(A) ORGANISM: homo sapien ~", (x) PUBLICATION INFORMATION:
(A) AUTHORS: Astrom, Anders Tavakkol, Amir Pettersson, Ulrika Cromie, Matthew Elder, James T.
Voorhees, John J.
~B) TITLE: Molecular Cloning of Two Human Cellular Retinoic Acid-Proteins (CRABP) (C) JOURNAL: J. Biol. Chem.
(G) DATE: 1991 ~K) RELEVANT RESIDUES IN SEQ ID NO:4: FROM 1 TO 137 (x) PU8LICATION INFORMATION:
(A) AUTHORS: Astrom, Anders Tavakkol, Amir Elder, James T.
Pettersson, Ulrika Cromie, Matthew Voorhees, John J.
(B) TITLE: Cloning of CRABPII cDNA from Human Skin:
Retinoic Acid Induces Expression of CRABPII but Not CRABPI in Human Skin in Vivo and in Dermal but Not Lung Fibroblasts in Vitro (C) JOURNAL: J. Invest. Denmatol.
(D) VOLUME: 96 (F) PAGES: ~47 547 (G) DATE: April-1991 . .
(xi) SEQUENCE DESCRIPTION: SEQ ID NO:4:
Met Pro Asn Phe Ala Gly Thr Trp Lys Met Arg Ser Ser Glu Asn Phe Asp Glu Leu Leu Lys Ala Leu 61y Val Asn Ala Met Leu Arg Lys Val ~- ~Y0 93/~331 2 1 :~ 1 5 5 ~ PCI /U593/03936 Ala Val Ala Ala Ala Ser Lys Pro His Val Glu Ile Arg Gln Asp 61y

4~ 45 Asp Gln Phe Tyr Ile Lys Thr Ser Thr Thr Val Arg Thr Thr Glu Ile ~0 55 60 Asn Phe Lys Val Gly Glu Gly Phe Glu Glu Glu Thr Val Asp Gly Arg ~0 Lys Cys Arg Ser Leu Ala Thr Trp Glu Asn Glu Asn Lys Ile His Cys Thr Gln Thr Leu Leu Glu Gly Asp Gly Pro Lys Thr Tyr Trp Thr Arg 100 105 11~
Glu Leu Ala Asn Asp Glu Leu Ile Leu Thr Phe Gly Ala Asp Asp Yal Val Cys Thr Arg Ile Tyr Yal Arg Glu (2~ INFORMATION FOR SEQ ID NO:5:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 1322 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: double (D) TOPOLOGY: linear (ii) MOLEGULE TYPE: DNA (genomic) (iii) HYPOTHETICAL: NO

(iv~ ANTI-SENSE:~NO

(vi) ORIGINAL SOURCE:
(A) ORGANISM: Homo sapiens (F) TISSUE TYPE: Placenta '2 13 45;3 `~
W093/22331 ' PCI/US93/03936 (Vi i ) IMMEDIATE SOURCE:
~A) LIBRARY: human placenta genomic library (B) CLONE: lambda 2.1 ,. ;~.
(ix) FEATURE:
(A) NAME/KEY: TATA signal (B~ LOCATION: 1008~.1013 (ix) FEATURE:
(A) NAME/KEY: exon (B) LOCATION: 1039..1245 (îx) FEATURE:
(A~ NAME/KEY: intron - (8) LOCATION: 1246.. 1322 (x) PUBLICATION INFORMATION:
(A) AUTHORS: Astrom, Anders Pettersson, Ulrika .
Voorhees, John J
(B) TITLE: Structure of the human cellular retinoic -acid-binding protein II (CRABP-II) gene: Early transcri pti onal regul ati on by reti noi c aci d (C) JOURNAL: J. Biol. Chem.
(G) DATE: 1992 (K) RELEVANT RESIDUES IN SEQ ID NO:5: FROM 1 TO 1322 (xi) SEQUENCE DESCRIPTION: SEQ ID NO:5:
CT6CAGGAAG CCGTGCCCTC CTCCCACCCT C m GATCTC CCG m CAAA 6CCGCTCTCC 60 AAGGGAGGGG`AGGTCGCTCC TTCCGCCC6T m ACAGCTC A6GATG6T6A CACCTGAGAC 120 -- wo 93/2~331 2 1 3 4 5 ~ () PcrllJS93/03936 CTGTGACCTC TGCCCTCCCA GCCTCGCGCC CTGGGCTCCT GCTTMCCCT TCMTGTCCG 360 ~ -CCGGCACC M GGAGTTG M T GCACCGAGTC AGGTTGGGGA TGG6TGGGGA ACA~GCGAGA 720 CGTGAGGAAC TCGGGTGGGG GACAGCCATA CACGAGCCCT GAGCATCTGC GCCCEiCAGCT 780 AGCTCCCCCC GCCTCTGCGG AGAGCGCGAT TCMGT&CTG GCTTTGCGTC CGCTTCCCCA 840 TCCACl~ACT A6CGCAGGA6 MGGCTATCT CGGTCCCCAG AGAAGCCTGG ACCCACACGC 900 GGAGGCGGGG CCACTTCMT CCTGG6~AGG GGCGGTTCCG TACAGGGTAT AAMGCTGTC 1020 CGCGCGGGAG CCCAGGCCAG CmGGGGTT GTCCCTGGAC TT6TCTTGGT TCCAGMCCT 1080 AGGA6TCTAC GG6GACCGCC TCCCGCGCCG CCACCATGCC CMCl~CTCT GGCAACTGGA 1200 AGGGCCCAGG TGGGGCA~GG GGG6GCTCTG 6AGTCCTCGA AGTTGGGGAT GAGAMGACA 1320 6C ` 1322 (2) INFORMATION FOR SEQ ID NO:S:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 1?17 base pairs (B) TYPE: nucl ei c aci d (C) STRANDEDNESS: doubl e ~D) TOPOLOGY: l ~ near (ii) MOLECULE TYPE: DNA (genomic) 213 ~5 ~ J
WO ~3~22331 ~ PCI`/US93/0393~

- 4~ - :
(i i i ) HYPOTHETICAL: NO

.
(iv) ANTI-SENSE: NO ~

(vi) ORIGINAL SOURCE:
(A) ORGANISM: Homo sapiens (F3 TISSUE TYPE: Placenta (vii) IMMEDIATE SOURCE:
(A) LIBRARY: human placenta genomic (B3 CLONE: lambda 2.1 (ix~ FEATURE:
(A) NAME/KEY: exon (B) LOCATION: 178..356 (ix) FEATURE:
(A) NAME/KEY: intron (B) LOCATION: 3~7..571 (ix) FEATURE:
(A) NAME/KEY: exon (B) LOCATION: 572,.6B8 (ix) FEATURE:
(A) NAME/KEY: intron (B) LOCATION: 699..1152 (ix) FEATURE:
(A) NAME/KEY: exon (B) LOCATION: 1153..1618 ~- W0~3/22331 21345~i0 PCI/US93/03936 ~x) PUBLiCATION INFORMATION:
(A) AUTHORS: Astrom, Anders Pettersson, Ulrika Voorhees, John J
(B) TITLE: StructurR of the human cellular retinoic-acid binding protein (CRABP-II) gene: Early transcrip~ional regulation by retinoic ac~d (C) JOURNAL: J. Biol. Chem.
(6) DATE: 1992 (K3 RELEVANT RESIDUES IN SEQ ID NO:6: FROM 1 TO 1717 ' (xi) SEQUENCE DESCRIPTION: SEQ ID NO:6:
CGCCTACCGT CTCCTTCMG GCACmCTT AGACACCCGG GCACCAGGCA GATGSACCCC 60 CCMCACACC CACCC~MGC AAGTCACA~A TCAGCCTGCT CCMCTGTCT TATGGGGAGG 120 GTGTGAGAGA GGTE. CAAA 6GCCCCTAAA AGGTGAGCCT CTCCTCTCTC CCCACAGGGG 180 AACAGGAGGG AGACACmC TACATCMM CCTCCACCAC CGTGCGCACC ACAGAGAl~A 300 ACTTCAAGGT TGG&GAGGAG mGAGGAGC AGACTGTGGA TGGGAGGCCC TGTMGGTGA 360 GTGGGGGCTA GAMG6GGCT CTCTCCCl~A TCCCTCTCAC TGCATT6CCC CTGCTATGGG 540 ACTGACCAAC GATGGG6AAC TGATCCTGGT AAGTCCTGCC TCCTCCCCAC TMTAGCAAt~ 720 CCATGGtT6G ACTCCGCACC CTGCTGATGG GACTGCTTGA ACAGMCTAA G6TGTCCCTA 840 TCCCATACAG TGCCCTGTGT 6MTTAGAAA TGGTGTTCCT mATGCMG CMAGGGCAT 900 wo 93/2~331 ' Pcr/usg3/o39 CTTCTGGGGG TCACTGGGAT GCCTCTTGCA GGGTCTTGCT TTCmGACC TCTTCTCTCC 1380 M GGGGAGTT GAGGGCCTI;A GCAGGAMGA CTGGCCCTCT AGCllCTACC Cl I IGTCCCT 1560 GTA6C-,TATA CA&~I IAGM TAI I IAmG TTMl~TTAT TAMATGCTT TMMA~ATA 1620 MACCTGTCT CTGGCTCATT GGGCAGGTAG ATMGTCACC TGAGTTCMC CTTGCCrCTG 1680

Claims

WHAT IS CLAIMED IS:
1. The nucleic acid substantially represented by SEQ ID NO. 6, wherein the nucleic acid encodes human CRABP-II and is isolated from a human cell or is not of human in vivo origin.

2. A nucleic acid having a sequence substantially complementary to SEQ ID
NO. 6, which encodes human CRABP-II wherein the nucleic acid is isolated from a human cell or is not of human in vivo origin.

3. The nucleic acid of Claim 1, wherein the residues represented by T are uracil residues.

4. The nucleic acid of Claim 2, wherein the residues represented by T are uracil residues.

5. The nucleic acid substantially represented by a sequence selected from the group consisting of SEQ ID NOS. 1, 5 and 6, wherein the nucleic acid encodes human CRABP-II, I and II, respectively, and is isolated from a human cell or is not of human in vivo origin.

6. A nucleic acid substantially complementary to a nucleic acid encoding human CRABP-II or I and having a sequence selected from the group consisting of SEQ
ID NOS. 1, 5 and 6, wherein the complementary nucleic acid is isolated from a human cell or is not of human in vivo origin.

7. The nucleic acid of Claim 5, wherein the residues represented by T are uracil residues.

8. The nucleic acid of Claim 6, wherein the residues represented by T are uracil residues.

9. The amino acid sequence substantially similar to SEQ ID NO. 2, wherein the amino acid sequence is isolated from a human cell or is not of human in vivo origin.

10. An antibody of binding fragment thereof raised to at least a portion of the amino acid sequence substantially similar to SEQ ID NO. 2.
11. A probe comprising a nucleic acid having a sequence comprising at least a segment of the nucleic acid sequence represented by a sequence selected from the group consisting of SEQ ID NOS. 1, 5 and 6, which encode human CRABP-II, I and II
respective, wherein the segment is of a length sufficient to hybridize to a complementary sequence thereto under stringent conditions, and wherein the residues represented by T
are selected from the group consisting of thymine and uracil.
12. A nucleic acid having a sequence complementary to the probe of Claim 11.13. A method of screening a sample for the expression of human CRABP-II
comprising the steps of:
a) providing a nucleic acid probe complementary to a portion of the nucleic acid sequence represented by SEQ ID NO. 1, wherein the residues represented by T are selected from the group consisting of thymine or uracil and the probe is of a length sufficient to hybridize under stringent conditions with a complementary sequence thereto;
b) contacting the sample to complementary sequences of nucleic acid in the sample under conditions favorable for hybridization of the probe;
c) providing means for detecting hybridization; and d) detecting hybridization, where expression of human CRABP-II is detected.
14. The method of Claim 13, further comprising the steps of:
e) providing means for quantifying hybridization of the probe to complementary sequences; and f) employing the means for quantifying hybridization, wherein expression of human CRABP-II is detected.

15. A method of screening a sample for a presence of CRABP-II comprising the steps of:
a) providing antibody or binding fragment thereof to CRABP-II;
b) containing the sample under conditions favorable for binding of the antibody or binding fragment to CRABP-II in the sample; and c) providing and employing means for detecting binding.
16. The method of Claim 15, further comprising the steps of:
d) providing and employing means for quantifying binding of the antibody or fragment to CRABP-II.
17. A nucleic acid having a sequence substantially represented by SEQ ID NO.3, wherein the nucleic acid encodes human CRABP-I and is isolated from a human cell or is not of human in vivo origin.
18. A nucleic acid substantially complementary to the nucleic acid represented by SEQ ID NO. 3 which encodes human CRABP-I, wherein the nucleic acid is isolated from a human cell or is not of human in vivo origin.
19. The nucleic acid of Claim 18, wherein the residues represented by T are uracil residues.
20. The nucleic acid of Claim 19, wherein the residues represented by T are uracil residues.
21. A nucleic acid probe comprising at least a segment of a nucleic acid selected from the group consisting of the nucleic acids having a sequence substantially represented by SEQ ID NO. 3 which encodes human CRABP-I and the sequence substantially complementary thereto, wherein the residues represented by T are selected from the group consisting of thymine and uracil, and wherein the segment is of a length sufficient to hybridize to a complementary sequence thereto under stringent conditions.
22. An amino acid sequence substantially represented by SEQ ID NO. 4, wherein residue 86 is alanine, and wherein the sequence is isolated from a human cell or is not of human in vivo origin.

23. An antibody or binding fragment thereto raised to at least a portion of the amino acid sequence of Claim 22.
24. A CRABP-II vector comprising a plasmid or virus including the sequence substantially represented by SEQ ID NO. 1 or the sequence substantially complementary thereto or both.
25. A CRABP-I vector comprising a plasmid or virus including the sequence substantially represented by SEQ ID NO. 3 or the sequence substantially complementary thereto or both.
26. A recombinant expression construct comprising a nucleic acid sequence coding for a retinoid-binding protein, the sequence substantially represented by SEQ ID
NO. 1 and operative contained in a cell.
27. The construct of Claim 26, further comprising a reporter element comprising a retinoid responsive element and a reporter gene operatively lined thereto.
28. The construct of Claim 26, further comprising a nucleic acid sequence for at least the DNA binding domain of a second retinoid-binding protein.
29. The construct of Claim 26, wherein the second retinoid-binding protein is an RAR or an RXR.
30. A recombinant expression construct comprising a cell containing the sequence substantially represented by SEQ ID NO. 3 and further including a reporter element containing a retinoid responsive element operatively linked to a reporter gene.
31. A reporter assay system comprising a CV-1 cell costransfected with an expression vector of CRABP-II and an expression vector for an additional retinoid-binding protein, a reporter element in operative configuration with said expression vectors, the reporter element comprising a reporter gene functionally linked to a retinoid responsive element.

32. The reporter assay of Claim 31, wherein the retinoid-binding protein is an RAR.
33. The reporter assay of Claim 31, wherein the retinoid-binding protein is an RXR.
34. A method for assaying binding of a putative ligand to a receptor proteincomprising the steps of:
a) providing a cell for transfection;
b) providing a vector comprising a plasmid or virus including a nucleic acid sequence selected from the group consisting of the sequence substantially represented by SEQ ID NO. 1, the sequence substantially represented by SEQ ID NO. 2, and the sequences substantially complementary thereto;
c) transfecting the cell with the vector of step b);
d) providing a retinoid responsive element;
e) transfecting the cell with the retinoid-responsive element;
f) exposing the cell transfected with the vector and the responsive element tot he putative ligand under conditions favorable for binding to occur; and g) providing and employing means for detecting binding.
35. The method of Claim 34, wherein means for detecting comprises a reportergene functionally linked to the retinoid responsive element and employing means for detecting comprises assaying for the presence of reporter gene product.
36. A nucleic acid probe comprising at least a segment of the nucleic acid sequence represented by SEQ ID NO. 3 which encodes human CRABP-I, wherein the segment is of a length sufficient to hybridize to a complementary sequence thereto under stringent conditions, and wherein the residues represented by T are selected from the group consisting of thymine and uracil.
37. A nucleic acid having a sequence complementary to the probe of Claim 36.38. The cellular retinoic acid binding protein encoded by SEQ ID NO. 6.