BACKGROUND OF THE INVENTION Field of the Invention The present invention provides a selective fluorescent chemosensor, sensitive to nanomolar concentrations of zinc (II), Zn2+, a device containing the same, and methods for measuring the concentration of a metal in a sample using the device.

Discussion of the Backgrounds The selective and quantitative detection of trace amounts of divalent zinc (Zn2+) is commercially desirable fc ~the diagnosis of metal ion induced disease states and in the protection of the environment. Zinc is an essential element which is present in the body at approximately 1 pmole/L. The USDA recommended dietary intake of Zn2+ is only 15 mg/day, this provides an indication of how little Zn2+ is required to maintain the required level of this element in a healthy adult. l Despite this relatively low concentration, Zn2+ plays an essential role in biology and nutrition. Minor perturbations of normal Zn2+ levels have been associated with retarded sexual maturation, stunted growth and skin damage.

Over 99% of Zn2+ in biological tissues and fluids is present in a chemically-combined form, with very little present as free

Zn2+ ion. Traditional methods such as atomic absorption effectively measure total Zon2+ but can not distinguish between the chemically-combined and free forms. The problem of detecting free Zn2+ is compounded because total free Zn2+ is decreased only very slightly (50-100 pmol/106cells) in cases of severe Zn2+ deficiency.

The detection of Zon2+ in the environment is also important, and is presently an intractable problem. 2 For example, interest in Zn2+ concentrations in the ocean stems from its dual role as a required nanonutrient and as a potential toxic agent due to its widespread industrial and marine usage. Zinc exists at natural levels in ocean surface water at a total concentration of ca. 0.1 nM. 3 Dissolved Zn2+ concentrations in seawater have been determined using atomic absorption spectrometry, mass spectrometry and voltammetry.

The equipment for these procedures is shore based, time consuming and expensive. The concentration data are also inaccurate due to interference from other cations naturally present in sea water. A rapid, selective and more sensitive test for Zn2+ concentrations which can be performed at sea is desirable.

The current technologies for selectively detecting free Zn2+ in the presence of the chemically-combined forms have sensitivities in the micromolar range. For example, zinquin4 has a rather low sensitivity with association constants approximately 5 orders of magnitude weaker than many metal

binding peptides and proteins. 5 In addition to its insensitivity, Zinquin cannot be used in the presence of copper.

There remains a need in the art for probes which have excellent sensivity for Zn2+ and high selectivity for this element in the presence of other transition metal cations.

SUMMARY OF THE INVENTION Accordingly, one object of this invention is to provide a selective fluorescent chemosensor, sensitive to nanomolar concentrations of a metal cation.

A second object of the present invention is to provide a device comprising the chemosensor and a detector.

A third object of the present invention is to provide a method for detecting the presence of a metal in a sample using such a device.

In light of the selectivity and avidity with which proteins bind Zn (II), 6 the present inventors used polypeptide architecture as the framework for metal ion recognition.'By manipulating synthetic polypeptides, the present inventors achieved the coordinated integration of fluorescent reporters for signal transduction within a metal binding peptidyl construct and obtained the chemosensor described herein. The present inventors have now developed a chemical probe which can provide direct, quantitative read-out of total free Zn2+ concentrations in a fluid sample, either biological or seawater, similar to existing fiber optic sensors. 8

BRIEF DESCRIPTION OF THE DRAWINGS A more complete appreciation of the invention and many of the attendant advantages thereof will be readily obtained as the same becomes better understood by reference to the following detailed description when considered in connection with the accompanying drawings, wherein: Figure 1. Schematic of the chemosensor design. Metal binding induces peptide folding, shielding a fluorophore from solvent and increasing emission intensity.

Figure 2. Fluorescence emission response of ZNS1 to increasing levels of Zn2+ (M) in the presence of Mg2+ (50 mM), and Co2+ (100 pM) ions; excitation at 333 nm. Spectra were acquired with 1.4 AM ZNS1,50 mM HEPES, pH 7.0, » =0. 5 (NaCl), 3.5 AM EDTA. Addition of EDTA was required to complex adventitious metal and establish the initial baseline prior to metal addition. The relative increase in fluorescence at 475 nm is linear between 0.1 and 1.0 ßM Zn2+.

Figure 3. Ratiometric analysis of the fluorescence emission changes upon the addition of various divalent metals.

Initial conditions are 1.4 UM ZNS1,50 mM HEPES, pH 7.0,3.5 AM EDTA, p=0. 5 (NaCl). Arrows represent the addition of concentrated metal stocks to give the indicated composition.

DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention comprises (i) a synthetic polypeptide template and (ii) a covalently attached fluorescent reporter that is sensitive to metal-induced conformational changes of the supporting framework, yet remote

from the metal binding site. This approach combines the advantages of optical signalling9 with the selective metal binding properties of an appropriate peptide. Furthermore, the modular design enables an approach wherein the fluorescent signalling mechanism is decoupled from the composition of the coordination sphere.

As used herein, the term free Zn refers to Zn ions which are available. The term chemically-combined refers to Zn that is tightly bound, or coordinated, to a competing ligand.

Metal Binding Peptide Templates Suitable metal binding peptide templates can be designed based on any naturally occurring metal binding protein.

Preferred Zn binding peptidyl templates in accordance with the present invention are described in the following Table along with the protein they are components of and their SEQ ID NOs.

Human SP1 QHICHI-QGCGKVYGKTSHLRAHLR--WHTG (SEQ ID NO : 1) PFMCTW-SYCGKRFTRSDELQRHKR--THTG (SEQ ID NO : 2) KFACPE---CPKRFMRSDHLSKHIK--THQN (SEQ ID NO : 3) Drosophila Serendipity b EIPCHI---CGEMFSSQEVLERHIKADTCQK (SEQ ID NO : 4) <BR> <BR> QATCNV---CGLKVKDDEVLDLHMN--LHEG (SEQ ID NO : 5) ELECRY---CDKKFSHKRNVLRHME--VHWD (SEQ ID NO : 6) <BR> <BR> KYQCDK---CGERFSLSWLMYNHLM--RHDA (SEQ ID NO : 7) ALICEV---CHQQFKTKRTYLHHLR--THQT (SEQ ID NO : 8) -YPCPD---CEKSFVDKYTLKVHKR--VHQP (SEQ ID NO : 9) Drosophila Serendipity q KQECTT---CGKVYNSWYQLQKHLS-EEHSK (SEQ ID NO : 10) NHICPI---CGVIRRDEEYLELHMN--LHEG (SEQ ID NO : 11) EKQCRY---CPKSFSRPVNTLRHMR--SHWD (SEQ ID NO:12) KYQCEK---CGLRFSQDNLLYNHRL--RHEA (SEQ ID NO : 13) PIICSI---CNVSFKSRKTFNHHTL--IHKE (SEQ ID NO : 14) -HYCSV---CPKSFTERYTLKMHMK--THEG (SEQ ID NO : 15) SGFCLI---CNTTFENKKELEHHLQ--FHAD (SEQ ID NO : 16) Drosophila Kruppel SFTCKI---CSRSFGYKHVLQNHER--THTG (SEQ ID NO : 17) PFECPE---CDKRTRDHHLKTHMR--LHTG (SEQ ID NO : 18) PYHCSH---CDRQFVQVANLRRHLR--VHTG (SEQ ID NO : 19) PYTCEI---CDGKFSDSNQLKSHML--VHTG (SEQ ID NO : 20) PFECER---CHMKFRRRHHLMNHK----CGI (SEQ ID NO : 21) Drosophila Snail RFKCDE---CQKMYSTSMGLSKHRQ--FHCP (SEQ ID NO : 22) THSCEE---CGKLYTTIGALKMHIR---HTL (SEQ ID NO : 23) PCKCPI---CGKAFSRPWLLQGHIR--THTG (SEQ ID NO : 24) PFQCPD---CPRSFADRSNLRAHQQ--THVD (SEQ ID NO : 25) KYACQV---CHKSFSRMSLLNKHSS-SNCTI (SEQ ID NO : 26) Xenopus Xfin SHHCPH---CKKSFVQRSDFLKHQR--THTG (SEQ ID NO : 27) PYQCVE---CQKKFTERSALVNHQR--THTG (SEQ ID NO:28) PYTCLD---CQKTFNQRSALTKHRR--THTG (SEQ ID NO:29) PYRCSV---CSKSFIQNSDLVKHLR--THTG (SEQ ID NO:30) PYECPL---CVKRFAESSALMKHKR--THST (SEQ ID NO : 31) PFRCSE---CSRSFTHNSDLTAHMR--KHTE (SEQ ID NO : 32) PYSCSK---CRKTFKRWKSFLNHQQ--THSR (SEQ ID NO : 33) PYLCSH---CNKGFIQNSDLVKHFR--THTG (SEQ ID NO : 34) PYQCAE---CHKGFIQKSDLVKHLR--THTG (SEQ ID NO : 35) PFKCSH---CDKKFTERSALAKHQR--THTG (SEQ ID NO:36) <BR> <BR> PYKCSD---CGKEFTQRSNLILHQR--IHTG (SEQ ID NO : 37) PYKCTL---CDRTFIQNSDLVKHQK--VHAN (SEQ ID NO : 38) PHKCSK---CDLTFSHWSTFMKHSK--LHSG (SEQ ID NO : 39) KFQCAE---CKKGFTQKSDLVKHIR--VHTG (SEQ ID NO:40) <BR> <BR> PFK^LL---CKKSFSQNSDLHKHWR--IHTG (SEQ ID NO : 41) PFPCYT---CDKSFTERSALIKHHR--THTG (SEQ ID NO : 42) PHKCSV---CQKGFIQKSALTKHSR--THTG (SEQ ID NO : 43) PYPCTQ---CGKSFIQNSDLVKHQR--IHTG (SEQ ID NO : 44)

PYHCTE---CNKRFTEGSSLVKHRR--THSG (SEQ ID NO : 45) PYRCPQ---CEKTFIQSSDLVKHLV--VHNG (SEQ ID NO:46) PYPCTE---CGKVFHQRPALLKHLR--THKT (SEQ ID NO : 47) RYPCNE---CSKEFFQTSDLVKHLR--THTG (SEQ ID NO : 48) PYHCPE---CNKGFIQNSDLVKHQR--THTG (SEQ ID NO : 49) PYTCSQ---CDKGFIQRSALTKHMR--THTG (SEQ ID NO:50) PYKCEQ---CQKCFIQNSDLVKHQR--IHTG (SEQ ID NO : 51) PYHCPD---CDKRFTEGSSLIKHQR--IHSR (SEQ ID NO:52) PYPCGV---CGKSFSQSSNLLKHLK--CHSE (SEQ ID NO : 53) SFKCND---CGKCFAHRSVLIKHVR--IHTG (SEQ ID NO : 54) PYKCGL---CERSFVEKSALSRHQR--VHKN (SEQ ID NO : 55) RYSCSE---CGKCFTHRSVFLKHWR--MHTG (SEQ ID NO:56) PYTCKE---CGKSFSQSSALVKHVR--IHTG (SEQ ID NO : 57) PYACST---CGKSFIQKSDLAKHQR--IHTG (SEQ ID NO : 58) PYTCTV---CGKKFIDRSSVVKHSR--THTG (SEQ ID NO : 59) PYKCNE---CTKGFVQKSDLVKHMR--THTG (SEQ ID NO : 60) PYGCNC---CDRSFSTHSASVRHQR--MCNT (SEQ ID NO : 61) Drosophila Terminus DLHCRR---CRTQFSRRSKLHIHQK-LRCGQ (SEQ ID NO : 62) Yeast SW15 TFECLF-PGCTKTFKRRYNIRSHIQ--THLE (SEQ ID NO : 63) PYSCDH-PGCDKAFVRNHDLIRHKK--SHQE (SEQ ID NO : 64) -YACP----CGKKFNREDALVVHRSDRMICSG (SEQ ID NO : 65) Xenopus transcription factor IIIA RYICSF-ADCGAAYNKNWKLQAHLC--KHTG (SEQ ID NO : 66) PFPCKE-EGCEKGFTSLHHTLRHSL--THTG (SEQ ID NO : 67) <BR> <BR> NFTCDS-DGCDLRFTTKANMKKHFN-RFHNI (SEQ ID NO : 68) VYVCHF-ENCGKAFKKHNQLKVHQF--SHTQ (SEQ ID NO : 69) PYECPH-EGCDKRFSLPSRLKRHEK--VHAG (SEQ ID NO:70) -YPCKKDDSCSFVGKTWTLYLKHVA-ECHQD (SEQ ID NO : 71) -AVCDV---CNRKFRHKDYLRDHQK--THEK (SEQ ID NO : 72) VYLCPR-DGCDRSYTTAFNLRSHIQ-SFHEE (SEQ ID NO:73) PFVCEH-AGCGKCFAMKKSLERHSV--VHDP (SEQ ID NO:74) Yeast ADR1 SFVCEV---CTRAFARQEHLKRHYR--SHTN (SEQ ID NO : 75) PYPCGL---CNRCFTRRDLLIRHAQ-KIHSG (SEQ ID NO : 76) Drosophila KRH TYRCSE---CQREFELLAGLKKHLK--THRT (SEQ ID NO : 77) KYQCDI---CGQKFVQKINLTHHAR--IHSS (SEQ ID NO:78) <BR> <BR> PYECPE---CQKRFQERSHLQRHQK--YHAQ (SEQ ID NO : 79) SYRCEK---CGKMYKTERCLKVHNL--VHLE (SEQ ID NO : 80) PFACTV---CDKSFISNSKLKQHSN--IHTG (SEQ ID NO : 81) PFKCNY---CPRDFTNFPNWLKHTR-RRHKV (SEQ ID NO:82)

Mouse MKR1 PFVCNY---CDKTFSFKSLLVSHKR--IHTG (SEQ ID NO : 83) PYECDV---CQKTFSHKANLIKHQR--IHTG (SEQ ID NO : 84) PFECPE---CGKAFTHQSNLIVHQR--AHME (SEQ ID NO : 85) PYGCSE---CGKAFTHQSNLIVHQR--IHTG (SEQ ID NO : 86) PYECNE---CAKTRRKKSNLIIHQK--IHTG (SEQ ID NO : 87) RYECNE---CGKSFIQNSQLIIHRR--THTG (SEQ ID NO : 88) PYECTE---CGKTFSQRSTLRLHLR--IHTG (SEQ ID NO : 89) Mouse MKR2 VYGCDE---CGKTFRQSSILLKHQR--IHTG (SEQ ID NO : 90) PYTCNV---CDKHFIERSSLTVHQR--THTG (SEQ ID NO : 91) PYKCHE---CGKAFSQSMNLTVHQR--THTG (SEQ ID NO : 92) PYQCKE---CGKAFRKNSSLIQHER--IHTG (SEQ ID NO : 93) PYKCHD---CEKAFSKNSSLTQHRR--IHTG (SEQ ID NO : 94) PYECMI---CGKHFTGRSSLTVHQV--IHTG (SEQ ID NO : 95) PYECTE---CGKAFSQSAYLIEHRR--IHTG (SEQ ID NO : 96) PYECDQ---CGKAFIKNSSLIVHQR--IHTG (SEQ ID NO : 97) PYQCNE---CGKPFSRSTNLTRHQR--THT- (SEQ ID NO : 98) Drosophile Hunchback NYKCKT---CGWAITKVDFWAHTR--THMK (SEQ ID NO : 99) ILQCPK---CPFVTERKHHLEYHIR--KHKN (SEQ ID NO : 100) PFQCDK---CSYTCVNKSMLNSHRK--SHSS (SEQ ID NO : 101) QYRCAD---CDYATKYCHSFKLHLRHYGHKP (SEQ ID NO : 102) IYECKY---CDIFFKDAVLYTIHMG--YHSC (SEQ ID NO : 103) VFKCNM---CGEKCDGPVGLFVHMARNAHS- (SEQ ID NO : 104) Trypanosome TRS-1 PTKCTE---CDATYQCRSSAVTHMV-NKHGF (SEQ ID NO : 105) VLHCTI---CASKFAVPGRLLHHLR-TIHGI (SEQ ID NO : 106) PFQCDL---CEASFGTHSSLSLHKK-LKHKS (SEQ ID NO : 107) EVQCGV---CQKVLSCRDSLIRHCK-AFHKG (SEQ ID NO : 108) MLV-T---CGRQCASKTGLTLHQK-KMHGM (SEQ ID NO : 109) Mouse NGFI-A PYACPV-ESCDRRFSRSDELTRHIR--IHTG (SEQ ID NO : 110) PFQCRI---CMRNFSRSDHLTTHIR--THTG (SEQ ID NO : 111) PFACDI---CGRFARSDERKKRHTK--IKLR (SEQ ID NO : 112) Human ZFY VYPCMI---CGKKFKSRGFLKRHMK--NHPE (SEQ ID NO : 113) KYHCTD---CDYTTNKKISLHNHLE--SHKL (SEQ ID NO : 114) AIECDE---CGKHFSHAGALFTHKM--VHKE (SEQ ID NO : 115) MHKCKF---CEYETAEQGLLNRHLL-AVHSK (SEQ ID NO : 116) PHICVE---CGKGFRHPSELRKHMR--IHTG (SEQ ID NO : 117) PYQCQY---CEYRSADSSNLKTHIK-TKHSK (SEQ ID NO : 118) PFKCDI---CLLTFSDTKEVQQHTL--VHQE (SEQ ID NO : 119) THQCLH---CDHKSSNSSDLKRHVI-SVHTK (SEQ ID NO : 120)

PHDCEM---CEKGFHRPSELKKHVA--VHKG (SEQ ID NO : 121) MHQCRH---CDFKIADPFVLSRHIL-SVHTK (SEQ ID NO : 122) PFRCKR---CRKGFRQQNELKKHMK--THSG (SEQ ID NO : 123) VYQCEY---CEYSTTDASGFKRHVI-SIHTK (SEQ ID NO : 124) PHRCEY---CKKGFRRPSEKNQHIM--RHHK (SEQ ID NO : 125) Xenopus P43 5S RNA binding protein LLRCPA-AGCKAFYRKEGKLQDHMA--GHSE (SEQ ID NO : 126) PWKCGI-KDCDKVFARKRQILKHVK--RHLA (SEQ ID NO : 127) KLSCPT-AGCKMTFSTKKSLSRHKL-YKHGE (SEQ ID NO : 128) PLKCFV-PGCKRSFRKKRALRRHLS--VHSN (SEQ ID NO : 129) LSVCDV-PGCSWKSSSVAKLVAHQK--RHRG (SEQ ID NO : 130) -YRCSY-EGCQTVSPTWTALQTHVK--KHPL (SEQ ID NO : 131) -LQCAA---CKKPFKKASALRRHKA--THAK (SEQ ID NO : 132) QLPCPR-QDCDKTFSSVFNLTHHVARKLHLC (SEQ ID NO : 133) THRCPH-SGCTRSFAMRESLLRHLV--VHDP (SEQ ID NO : 134) All amino acid sequences shown in the above table and throughout the specification are represented by the standard one-letter code for amino acids. A, alanine; C, cysteine; D, aspartic acid; E, glutamic acid; F, phenylalanine; G, glycine; H, histidine; I, isoleucine; K, lysine; L, leucine; M, methionine; N, asparagine; P, proline; Q, glutamine; R, arginine; S, serine; T, threonine; V, valine, W, tryptophan; Y, tyrosine. Dashes are present merely to allow alignment of the sequences.

The above Zn peptidyl templates can be designed based on known metal binding domains in zinc-fingers. A database with the sequence for 135 such domains appears in"A Consensus Zinc Finger Peptide: Design, High-Affinity Metal Binding, a pH- Dependent Structure, and a His to Cys Sequence Variant," Krizek et al., J. Am. Chem. Soc. 1991,113,4518; incorporated herein by reference. A database with 1340 zinc finger domains appears in GH Jacobs, EMBO J., 1992,11,4507-4517.

The metal binding domain of the zinc finger proteins shown above and described throughout the literature can be represented by the consensus sequence : 1Oll (Rl)-X-C-X24-C-X3-R2-X5-R3-X2-H-X35-H-X2-6 Where X is any a-amino acid (both natural and synthetic) and the subscript refers to the number of residues at each location. The residues R1-R3 shown in bold in the above consensus sequence are known to form a hydrophobic core which contributes to the proper folding of the domain. These residues are preferably maintained as hydrophobic residues for high affinity binding. Suitable hydrophobic residues include Phe, Tyr, Leu, His, Ile, Val, etc. Preferably, R1 is Phe or Tyr; R2 is Phe and R3 is Leu. However, as mentioned above, these residues can also be replaced with any a-amino acid. It is well established that the R1, R2, R3 and X residues can be any a-amino acid; for example, each residue was replaced with an Ala and the domain of a zinc finger protein still folded properly and bound zinc. 12 The peptidyl template can either consist of the whole domain shown above or a portion of those domains which bind Zn (herein referred to as the"Zn binding domain"). A Zn binding domain comprises at least 15 residues, preferably at least 20 residues, particularly preferably at least 25 residues, and may include the entire amino acid sequence of the Zn binding domain or, alternatively, a sequence comprised of at least those amino acid residues known to coordinate to Zn in their relative spatial array (the four residues in each sequence shown in bold text in the above Table).

That is, the Zn binding peptidyl template which is comprised of a portion of or the whole metal binding domain may contain amino acid substitutions within the natural sequence, so long as those substitutions do not totally destroy the Zn binding capacity of the domain. The selectivity and sensivity of the probe, that is its ability to detect free Zn versus chemically-combined Zn, can be conveniently varied by substituting non-coordinating amino acid residues.

The metal coordinating amino acids can be replaced with other metal coordinating amino acids. This provides a means for conveniently tuning the selectivity and sensitivity of the probe. Suitable metal coordinating amino acids include natural amino acids such asyeysteine, histidine, aspartic acid, and glutamic acid. Alternatively, unnatural amino acid residues which bind metals can be used. A variety of such residues have recently been described in the literature, including bipyridal-alanine, and phenanthrolyl-alanine. 13 The Zn binding template may also comprise, in addition to a portion of, or the whole metal binding domain, attachments on the N-and/or C-termini. For example, the metal binding domain may be extended with either additional amino acids or organic blocking groups. Any number of amino acids can be used to extend the termini so long as the extension does not totally destroy the Zn binding capacity of the Zn binding domain. Preferably, the extensions at the N-and/or C-termini include the natural sequence of the metal binding protein surrounding the Zn binding domain. The primary sequences for

each of the Zn binding domains in the above Table are known in the literature. The Zn binding template includes any fragment of the natural occurring protein which includes at least the Zn binding domain shown in the above Table.

Suitable organic groups useful for extending the N-and C-terminus include organic blocking groups which are known in the art and include, for example, amines, carboxylic acids, etc. Alternatively, the organic group can be a polyfunctional compound useful for linking the metal binding domain to a support or the support itself. Suitable solid supports and polylinkers are well described as are the means for attaching these to peptides. 14 Suitable solid supports include, for example, silicon, gold, glass and quartz.

Fluorescent Reporter Suitable fluorescent reporter groups include any fluorescent group with photophysical properties that are sensitive to environment, such as:

The above residues can be synthesized as described in the art. lys Fluorophores are known to be polarity dependent as described in the literature. 16 In the absence of divalent zinc, the zinc finger domains tend to be disordered. In contrast, when appropriate metal cations are present, coordination to the metal binding residues nucleates the formation of a defined tertiary structure including a cluster of conserved hydrophobic residues. To detect binding of a metal, the fluorophore is suitably placed within the sequence of the metal binding domain or at one of the ends. There is no limit to the distance between the fluorophore and the metal binding site provided that the metal binding event induces a significant change in the conformation of the polypeptide and causes the fluorophore to experience a new environment.

The fluorophore can be incorporated into the peptidyl template using conventional peptide synthesis techniques (see for example, M. Bodansky, Principles of Peptide Synthesis, Springer-Verlag : Berlin, 1984; Stewart & Young, Solid Phase Pepzide Synthesis, Pierce: Rockford IL, 1984; incorporated herein by reference). In general, residues containing a carboxylic acid, aryl chloride or a sulfonyl chloride group can be incorporated onto the N-terminus of the peptidyl template or, preferably, into any residue in the peptidyl template. For example, these fluorphores can be incorporated onto the side chain of a residue (for example, onto the amino group of a 0-amino alanyl group).

As mentioned above, in the first embodiment of the invention, the present inventors examined Zn2+ binding proteins. In the absence of metal ions, some Zn2+ binding proteins are structurally disordered. 17 When Zn2+ ions are present, coordination to cysteine and histidine residues nucleates the formation of a defined tertiary structure including a cluster of conserved hydrophobic residues (underlined in the above consensus sequence (I)).

For example in the first embodiment of their invention, the present inventors replaced one of the hydrophobic residues with a microenvironment-sensitive fluorophore to allow the monitoring of the metal binding event."The fluorophore used was chosen to allow the selective elaboration of a side chain amine at variable positions within a synthetic polypeptide.

That is, a differentially protected 0-amino alanine (Baa) derivative, (S)-2,3-diamino-Na-9-fluorenylmethyl-oxycarbonyl- N-allyloxycarbonyl propanoic acid (Fmoc-L-Baa (alloc)-OH), was utilized. Compatibility of this residue with Fmoc solid phase peptide synthesis allows for orthogonal deprotection with a palladium catalyst. l9 The newly liberated amine is then available for fluorophore attachment, while leaving the remainder of the protected, peptide intact until the final resin deprotection.

As a first embodiment, the zinc finger domain"ZFY- swap"20 was chosen as a template, and Phelo of that sequence was chosen as the site for replacement with the dansylated Baa amino acid derivative. As described below, the peptide ZNS1

was synthesized21 (Scheme I) and assayed for the ability to selectively bind and report divalent zinc. The amino acid sequence of ZNS1 is (SEQ ID NO. 135): Acetyl-Y Q CQYCEKRFADSSNLKTHIKTKHS-NH2 The fluorophore was coralently attached to the Baa residue via an amide bond.

A solid phase device A solid phase device in accordance with the present invention comprises the chemosensor described above immobilized onto a solid support and a means for detecting fluorescence.

The peptidyl template with the incorporated fluorophore can be assembled by standard solid phase synthesis methods onto a solid phase as described above. A quartz surface is preferably used.

The solid phase immobilized chemosensor can then be integrated with a means for detecting fluorescence. A preferred means is a charge coupled device (CCD). This device can then be used to obtain direct read out of fluorescent emission from the chemosensor following selective irradiation at the appropriate wavelength of the fluorophore (usually in the range of 300-350 nm). Selective irradiation of the probe can be applied using fiber optics. 22

Methods for detecting Znfll) Methods for detecting the presence of Zn in a sample will depend on the nature of the chemosensor.

If the chemosensor is in solution, the method comprises: contacting a sample with the chemosensor to form a test solution, irradiating the test solution at a wavelength sufficient to excite the fluorophore, measuring the fluorescence emission of the test solution, and determining the concentration of metal in said sample based on said fluorescence emission.

The amount of Zn in the sample can be quantified by comparing the measured fluorescence against a standard sample with known Zn (II) concentration.

Suitable samples include any aqueous solution including seawater, biological fluids such as plasma, blood, urine, etc.

In this embodiment of the method, the means for irradiating the sample and the means for measuring the fluorescence emission of the test solution are ideally performed by the same means-preferably a fluorometer. A wide range of fluorometers are commercially available.

If the chemosensor is immobilized onto a solid support as described above, the method of the present invention comprises : contacting the immobilized chemosensor with a sample, irradiating and measuring the fluorescent emission of said sample, and

determining the concentration of metal in said sample based on said fluorescent emission.

Suitable samples are described above.

In this embodiment, the device provides the means for irradiating the sample and measuring the Zn (II) concentration in the sample.

In the first embodiment, the fluorescence emission properties of ZNS1 were sensitive to the presence of nanomolar concentrations of Zn2+ (Figure 2). The addition of Zn2+ results in increased emission upon excitation of 333 nm, and the response (at 475 mn) is linear between 0.1 and 1 ßM Zn2+. 23 The chemosensor was selective for Zn (II) by a factor of 107 over Na+, 106 over Mg2+ and 103 over Co2+.

As one important application of the zinc sensor is for the analysis of sea water, preliminary studies were also carried out in the presence of metal cations including 0.5 M Na+, 50 mM Mg2+ and 100 ßM Co2+. The chemosensor is compatible with these competing species. 2a Due to a blue-shift which occurs upon metal binding, dual-wavelength ratiometric analysis of the fluorescence emission using 525 nm and 650 nm readings were used for Zn2+ quantitation in solution. 2' An example of the response obtained by this method and the minimal effect of competing ions is shown in Figure 3. The addition of EDTA to the complexed material demonstrates that the signalling mechanism of this peptide is readily reversible. The presence of two cysteine residues in ZNS1 makes it incompatible with ions such as Cu2+, due to complications from dithiol oxidation. However, the modular

nature of the design and the ready access to a variety of metal coordinating amino acids suggest strategies for remediating this limitation.

The first embodiment of the chemosensor is described in detail below and demonstrates the feasibility of exploiting metal-dependent structure modulation as the basis for trace analyte detection. ZNS1 shows excellent affinity for Zn2+ relative to the chemical probe Zinquin. 26

1. W Kain and B Schwederski in Bioinorganic Chemistry: Inorganic Elements in the Chemistry of Life Wiley, 1994.

2. RB Thompson and ER Jones, Anal. Chem. 1993,65: 730-734.

3. JL Nowicki et al, Anal. Chem. 1994,66: 2732-2738.

4. PD Zalewski et al., Biochem. J., 1993,296: 403-408.

5. PD Zalewski et al., Chemistry and Biology, 1994,1: 153-161.

6. Fracesto da Silva et al., The Biological Chemistry of the Elements, Clarendon Press; New York, p. 561 (1993).

7. For a protein-based sensor that uses a diffusible fluorophore to report divalent zinc see (a) Thompson et al., Anal. Biochem., vol. 227, pp. 123-128 (1995); and (b) Thompson, Anal. Chem., vol. 65, pp. 730-734 (1993).

8. RB Thompson, Circuits and Devices, 1994: 14-21.

9. Czarnik in Fluorescent Chemosensors for Ion and Molecule Recognition, ACS, Washington, D. C., pp. 1-9 (1993).

10. Berg, Acc. Chem. Res., vol. 28, pp. 14-19 (1995); Klug et al., FASEB J., vol. 9, pp. 597-604 (1995).

11. An alternate consensus sequence is observed for a subset of the zinc finger domains. M. A. Weiss et al, Biochemistry, vol. E29, pp. 5660-5664 (1990).

12. Michael et al., Proc. Natl. Acad. Sci. USA, 1992,89: 4796- 4800.

13. Include. papers which describe the synthesis and incorporation of these amino acids-at all positions of substitution; incorporated herein by reference.

14. For a review see, MA Gallop et al., J. Med. Chem., 1994, 37: 1233-1251.

15. For a review see, RP Haugland,"Covalent Fluorescent Probes", In Excited States of Biopolymers, RF Steiner, Ed., Plenum Press: NY, 1983; incorporated herein by reference.

16. JR Lakowicz, Effects of Solvents on Fluorescence Emission, In"Principles of Fluorescence Spectroscopy", Plenum Press, New York, 1983; pp 189.

17. Frankel et al., Proc. Natl. Acad. Sci. U. S. A., vol. 84, pp.

4841-4845 (1987); Eis et al., Biochemistry, vol. 32, p.

7981 (1993).

18. Lakowicz in Principles of Fluorescence Spectroscopy, Plenum Press, New York, pp. 189-218 (1983).

19. Kates et al., Anal. Biochem., vol. 2/2, pp. 303-310 (1993).

20. Zinc finger domains with single residue replacements have been structurally characterized, showing the hydrophobic cluster is tolerant of substitution at position 10 (i. e., the second underlined residue in the consensus sequence).

Weiss et al., Biochemistry, vol. 29, pp. 9808-9813 (1990).

21. The peptide was purified by reversed phase (C18) HPLC. The identity of ZNS1 was confirmed by electrospray mass spec.

22. RB Thompson, Circuits & Devices, May 1994,14-21.

23. CD and W-vis metal titration experiments verify that ZNS1 binds C02, with high affinity (Ko 1-5 ßM). Competition analysis in the presence of 1.5 ßM Zn2+ AND 100 UM Co2+ reveals less than 5% change in the zinc-bound species, therefore the KD for Zn2+ is <1 nM.

24. Bruland in Trace Elements in Sea Water; J. P. ERilet et al; Academic Press, London, pp. 157-220 (1975).

25. In the absence of metal ions the emission maximum is at 560nm. The Zn2+ complex has an emission maximum at 525 nm.

Since these maxima are not widely separated, ratiometric analysis is carried out at 525 and 650 nm. This analysis normalizes data against bias from sources such as varying instrument response.

26. Zalewski et al., Chem. Bio., vol. 1, pp. 153-161 (1994).

Having generally described this invention, a further understanding can be obtained by reference to certain specific examples which are provided herein for purposes of illustration only and are not intended to be limiting unless otherwise specified.

EXAMPLES Reagents used were purchased from Aldrich Chemical Co. or Fluka. Protected amino acid derivatives and reagents were obtained from Milligen (Perseptive) Biosearch, except FmocL- Baa (alloc)-OH which was synthesized as outlined below.

Fluorescence experiments were performed on a SLM-Aminco SPF 500c spectrofluorometer at ambient temperature. W-Vis spectra were obtained, on a Shimadzu UV160U, or a Beckman DU 7500 recording spectrophotometer.

All solutions and buffers employed in assays-. :-ere composed of high-purity (18 Mn cm-1) water obtained from a MILLI-Q (available from Millipore) deionization purification system, and were handled in acid-washed HDPE containers. To prevent metal catalyzed-and auto-oxidation of ZNS-1, buffer solutions were sparged with argon for one half hour then hydrogen for fifteen minutes immediately prior to use.

Metal titrations were performed by the addition of aliquots of concentrated stock solutions. The concentration of these stocks were determined by complexometric titrations performed in triplicate with standard EDTA solutions (Aldrich) and an appropriate colorometric indicator. 1 Synthesis of Fmoc-L-Baa (alloc)-OH The amino acid derivative (4) was synthesized from the precursor N «-Boc-L-ß-amino alanine33 via the route outlined in Scheme I. *w*r onoos o ots DIEA NH TFA TFA 1 2 3 2 3 1 2 3 0 0 Fmoc-ONSu DIEA FmocN C O O H THF/H2O H 4

Scheme 1. Strategy for the synthesis of Fmoc-Baa (alloc)-OH.

Boc-L-Baa (alloc)-OH (2) A solution of the Boc-amine (1) was prepared by dissolving 4.33 g (21.2 mmol) in 100 mL of 50% v/v THF in deionized water, assisted by sonication at 50 °C.

Sodium carbonate (2.46 g, 23.2 mmol) in 7 mL of water was added, and the resultant slurry was cooled to 0°C with constant stirring. Allyl chloroformate (2.25 mL, 21.2 mmol) and diisopropyl-ethylamine (3.66 mL, 21.2 mmol) were added in alternating portions over a period of 20 minutes, after which the ice bath was removed and the reaction vessel let stir an additional 1.5 hr. The organic solvent was removed under reduced pressure, and additional deionized water added to increase the volume to 250 mL. Citric acid was added until pH 3.2, and the solution extracted with CH2Cl2 (5 x 200 mL). The organic layers were combined, dried over MgSO4, and the solvent

removed under reduced pressure to give a clear oil. Traces of contaminating lactam were separated by chromatography on silica gel (95: 4: 1, CHCl3/MeOH/AcOH) to afford 4.60 g (75% yield) of a clear, colorless oil.

L-Baa (alloc)-OH TFA (3) A 100-mL round bottom flask was charged with a stir bar, Boc-L-Baa (alloc)-OH (2,4.55 g, 15.8 mmol), and CH2C12 (15 ML), then chilled to 0 °C with constant stirring. Trifluoroacetic acid (TFA, 10 mL, 130 mmol) was added drop wise over 15 minutes. After an additional 15 minutes the reaction was allowed to come up to room temperature and let stir for 2 hours. Excess TFA was removed under a stream of nitrogen, then the solvent was removed in vacuo to give a clear gum. Lyophilization of this residue from benzene gave 3 as a clear oil (4.63g, 97% yield).

Fmoc-L-Baa (alloc)-OH (4) The carbamate amine 3 (370 mg, 1.21 mmol) was dissolved in 2: 1 THF/H2O (12 mL), requiring the assistance of vigorous stirring and sonication.

Diisopropylethylamine (DIEA, 465 ßL, 2.66 mmol) was added until the solution was slightly basic to litmus. Under vigorous stirring, N- (9-Fluorenylmethoxy-carbonyl) succinimide (FmocONSu, 816 mg, 2.42 mmol) was added in alternating portions with DIEA (420 ßL, 2.42 mmol) over 30 minutes. The reaction was allowed to stir an additional four hours. The organic phase was removed in vacuo, and the remaining aqueous solution was increased in volume to-100 mL, and acidified with citric acid to pH 3. This was extracted with diet-yl

ether (2 x 50 mL), and the organic phase discarded. The aqueous phase was further extracted with CH2Cl2 (4 x 100 mL), the organic phases combined, dried over anhydrous Na2SO4, and concentrated under reduced pressure to afford a clear oil.

Purification by silica gel chromatography (95: 5: 3 CHCl3/MEOH/AcOH), followed by lyophilization from benzene gave 4 (465 mg, 96% yield) as a white powder.

Peptide synthesis General. Peptides were synthesized on a Milligen 9050 peptide synthesizer at a 0.125 mmol scale. The support used was in all cases PAL-PEG-PS (Millipore) thus providing carboxyterminal primary amides. Coupling were performed at a concentration of 0.3 M acylating reagent and HOBT, in a volume sufficient to achieve a three-fold excess of amino acid to resin-bound acne. Pentafluorophenyl ester/1- hydroxybenzotriazole (Pfp/HOBT) chemistry was employed for all residues except Fmoc-L-Baa (alloc)-OH which was coupled by in situ active ester generation using HOBT/N, N'-diisopropyl- carbodiimide (HOBT/DIPCDI) activation. All coupling reactions were allowed to proceed for greater than 45 minutes, and were monitored for completion with the Enhanced Monitoring Opfion (EMO, Milligen). A coupling judged to be incomplete (< 97%) was double coupled for an additional 45 minutes. After successful coupling cycles, any residual amines were acylated by a 10 min wash with 0.3 M acetic anhydride/HOBT solution in 9: 1 DMF/dichloromethane. Removal of the Fmoc group was performed with piperidine (20% v/v in DMF) with a standard

wash duration of 7 minutes. This time was extended if an inefficient deblock was detected by EMO. After addition of the final residue, the amino-terminus was acetyl-capped (DMF/acetic anhydride/triethylamine, 4 mL: 63 ßL : 94 ßL) for 0.5 h then washed with DMF (5 x 10 mL) and MEOH (5 x 10 m).

Residual solvent was removed under reduced pressure.

Alloc removal. The method of Kates et al. 4 was employed with some minor modifications. A typical procedure for removal of the alloc group was as follows. Under a blanket of nitrogen, a 20-mL plastic, stoppered vial was charged with resin from the completed peptide synthesis (300 mg, 0.21 meq/g, 63 Mmol), and 5 mL of a solvent cocktail (CHCl3 morpholine/AcOH, 90: 5: 5). The resin was allowed to swell 10 min, to which tetrakis (triphenyl-phosphine) palladium (200 mg, 173 ßmol) was added under a blanket of nitrogen. The vial was capped, shielded from light, and placed ori a wrist-action shaker at low speed for 2 h. The resin was filtered, washed with CHC13 (5 x 10 mL), and a palladiumchelating cocktail (DMF/diethyldithiocarbamic acid 3H2O/triethylamine, 25 mL/ 225 mg/250 gel). Traces of this solution were removed with a basic wash (0.5% v/v triethylamine in DMF), and a final wash with methanol. The resin was transferred to a clean plastic vial, and the residual solvent removed under reduced pressure.

Dansyl coupling. Lyophilized resin taken directly from the alloc-removal procedure (280 mg, 0.21 meq/g, 59 Umol), was allowed to swell in DMF (5 mL). Dansyl chloride (159 mg, 590

Umol) was added, followed by triethylamine (82 mL, 590 Umol).

The vial was capped under a blanket of nitrogen, and placed on a wrist-action shaker at low speed for 2 h. The resin was filtered, washed with a basic wash (5 x 10 mL, see above), washed with DMF (5 x 10 mL), then finally washed with MeOH (5 x 10 mL). The resin was transferred to a clean 20 mL polyethylene vial, and residual solvent was removed under reduced pressure prior to peptide cleavage.

Peptide cleavage acid purification. Peptides were cleaved after fluorophore coupling using 10 mL of Reagent Ks (trifluoroacetic acid/H20/ethanedithiol/thioanisole/ phenol, 82.5/5/5/5/2.5) with a 2 h incubation period. The resin was filtered, concentrated to about 2 mL volume, and precipitated with ether/hexane 2: 1 at-20°C for 30 min. The supernatant was decanted, and the peptides triturated with ether/hexane 2/1 (5 x 50 mL). The resultant solid was resuspended in water (20 mL), lyophilized, and then purified by reverse phase (C18) high performance liquid chromatography (HPLC), gradient elution with 0. 1% v/v TFA in acetonitrile (solvent B) added to 0.1% v/v TFA in water (solvent A). The reduced (dithiol) and oxidized (intramolecular disulfide) forms of the peptide are separable by HPLC. Both forms of the peptide were collected and characterized, then combined and submitted together to a reduction protocol.

Peptide reduction The reduction was performed by dissolving the combined HPLC purified peptides in 0.1 M phosphate buffer, pH 8.0,50 MM dithiothreitol (DTT), lOmM EDTA, 6 M guanidine, and incubating at 25 °C, 1 h. The reduction mixture was loaded onto a Cl, Sep-Pak (Millipore), and buffer salts and DTT were eluted with 10 mL water. The reduced peptide was eluted with 30% acetonitrile in water.

Acetonitrile was removed from this solution under a stream of nitrogen until the sample could be frozen at 0°C, whence it was lyophilized. This was resuspended in high-purity water that had been sparged with argon 0.5 h then hydrogen 0.5 h. to delay oxidation. The resulting stock solution was periodically analyzed by HPLC and no significant (> 2%) disulfide formation was detected.

The concentration of stock solutions ZNS-1 was determined by reaction with Eliman's reagent, 2,5'-dithiobis (2- nitrobenzoic acid) (DTNB) 6. These assays were performed in triplicate, with excellent agreement (< 5% error) between replicate runs. With this knowledge, the extinction coefficient of the dansyl chromophore was determined at 333 nm for ZNS-1 to be 5.3 x 103 M-1 cm-1 ih 50 mM Hepes, pH 7.0. This value was used for all subsequent quantitations of ZNS-1.

Emission fluorescence assay. Assays were performed at pH 7.0 in 50 mM Hepes buffer, 0.5 M NaCl in a stoppered 750 AL (1 cm x 3mm) quartz fluorometer cell. The concentration of ZNS-1 in a given assay was determined by measurement of the unique absorbance of the dansyl chromophore (e 333 nm = 5.3 x 103).

Blank scans were taken against the same buffer immediately prior to ZNS-1 addition. Emission spectra were accumulated at I nm intervals with the following parameters: excitation wavelength = 333 nm, emission wavelength = 400-750 nm, excitation band pass 4 mm, emission band pass = 2 mm, lamp potential = 975 V, gain = 10, filter (time constant) 3. To mask the effects of adventitious metal ions present prior to the initiation of metal titration, small aliquots of EDTA were added (in increments of 0.5 ßM) until overlaying emission spectra were observed, and the wavelength of maximum emission was 560 nm. The EDTA concentration at the initiation of the metal titration was never in excess of 4 MM (< 0.001% of the Mg2+, and < 4% of the co2+ subsequently added). With the concentrations of Zn2+ and CO2+ present over the course of a filtration, and the relative stability constants of EDTA for these species, at most 10% of the Zn2+ would be complexed by EDTA. Divalent metal ions were added to the assay from stock solutions prepared in unbuffered water. These included a 2.83 M solution of MgCl2, a 0.0969 M solution of COCl2, and a 100 PM solution of ZnCl2.

References 1."Vogel's Textbook of Quantitative Inorganic Analysis;" Bassett, J.; Denney, R. C.; Jeffery, G. H. Mendham, J.; William Clowes: London, 1978; pp 223-402.

2. Arnold et al., J. Am. Chem. Soc. 1985,107,7105-9.

3. Kucharczyk et al., Synth. Comm. 1989,19,1603-1609.

4. Kates et al., Anal. Biochem. 1993,212,303-310.

5. King et al., Int. J. Pept. Protein Res. 1990,36,255- 266.

6. Riddles et al., Meth. Enzymol. 1983,91,49-60.

7. O'Sullivan & Smithers, Meth. Enzymol. 1979,63,294.

Obviously, numerous modifications and variations of the present invention are possible in light of the above teachings. It is therefore to be understood that within the scope of the appended claims, the invention may be practiced otherwise than as specifically described herein.