CA2205593A1 - Enzyme gene therapy catalysing prodrug extracellular conversion - Google Patents

Abstract

The invention relates to a molecular chimaera for use with a prodrug, the molecular chimaera comprising a transcriptional regulatory DNA sequence capable of being activated in a targetted mammalian cell and a DNA coding sequence operatively linked to the transcriptional regulatory DNA sequence and encoding a secretion signal peptide and a heterologous enzyme, such that on expression of said coding sequence in the targetted cell, the heterologous enzyme is capable of passing through the cell plasma membrane and is capable of catalysing extracellular conversion of the prodrug into a cytotoxic or cytostatic agent. When the molecular chimaera is used in gene or virus directed enzyme prodrug therapy the fact that enzyme is secreted so that conversion of the prodrug to the cytotoxic or cytostatic agent is extracellular increases neighbouring cell kill.

Description

CA 0220~93 1997-0~-16 ENZYME GENE THERAPY CATALYSING PRODRUG EXTRACELLULAR CONVERSION

The present invention relates to a novel targetted enzyme prodrug therapy.

Targetted enzyme prodrug therapies provide a method for restricting the activity of a chemothe,dpeulic agent to a particular target site. This is desirable when the systemic presence of the chemotheld~eu~c agent produce unwanted side effects. Although applicable to any the,dpeuLic regime for which a l~elLed approach is advisable, the technique is clearly 10 particularly applit~hle to the tre~tm~nt of cancer where th~.tpeul;c regimes have previously involved the systemic introduction of highly cytotoxic compounds which exert their effect in a non-selective manner on both healthy and tumourogenic cells.

Research in the area of cancer chemotherapy has produced a variety of ~ntit~-mour agents 15 which have differing degrees of efficacy. Standard clinit ~lly useful agents include adriamycin, actinomycin D, methotlexat~, 5-fluorouracil, cis-pl~tinnm~ vin~ ~ictine and vinblastine. However, these presently available ~ntitllmour agents are known to have various disadvantages such as toxicity to healthy cells and recic~nce of certain tumour types. Other forms of therapy such as surgery, are known. However it is clear that novel approaches and 20 entities for cancer therapies are required if cignific~nt progress in the clinical management of this disease is to be achieved.

Targetted enzyme prodrug therapies may offer cignific~nt improvements in cancer therapy, either alone or in combination with existing tre~tmPnt regimes. One such therapy relates to 25 the use of molPcul~r chim~r~c, which encode a heterologous enzyme and, which are delivered to targetted cells. Intr~c~ r exp,ession of the enzyme allows catalysis of a subsequently ~rlminictered prodrug to its active cytotoxic or cytostatic form. The therapy is known as gene or virus directed enzyme prodrug therapy (GDEPT or VDEPT).

SUBSTITUTE SHEET (RULE 26) CA 0220~93 1997-0~-16 WO-A-90 07936 describes a tre~tmPnt for an infection or a hyperproliferative disorder which is ch~r~rt~ricP~d by the presence, in the ~ffecte~ cells, of a trans-acting factor capable of regulating gene ~cplession by inserting into the cells a polyllucleotide construct having a cis-acting regulatory sequence which is regulated by the trans-acting factor and an effector gene S which renders said cell susceptible to protection or destruction. For PY~mpl~, the cis-acting region may be homologous to the HIV tar region, and the effector gene may encode ricin A
or HSV-1 thymidine kinase. Upon infection with HIV, the HIV tat protein activates the tar region, and induces tr~ncrnrtion and e,~lt;ssion of ricin A, resnlting in cell death, or of HSV-l tk, reslllting in cell death upon tre~tmPnt with dideoxynucleoside agents such as 10 acyclovir and gancyclovir.

EP-A-0 334 301 describes methods for the delivery of vectors using recombinant retrovirus wherein the vector construct directs the e~lession of a protein that activates a compound with little or no cytotoxicity into a toxic product in the presence of a pathogenic agent, 15 thereby effP~tin~ loc~licP~ therapy to the pathogenic agent.

EP-A-0 415 731 describes molP~Cul~r cllim~pr~c for use with prodrugs, comprisingtrancrrirtional regulatory DNA sequences capable of being selectively activated in a m~mm~ n cell, a DNA sequence operatively linked to the transcriptional regulatory DNA
20 sequence and encoding a heterologous enzyme capable of catalysing the conversion of the prodrug into an agent toxic to the cell.

FYicting GDEPT and VDEPr regimes involve the intr~ce-~ r ~ression of the heterologous enzyme and intr~rp~ r catalysis of the prodrug. This limits the efficiency of such therapy 25 as in, for example tumour therapy using retroviral meAi~ted ~c;lling, only between 1 and 10% of cells of the solid tumour may be infected by retrovirus. Hence therapy is limited to those 1 to 10% of infected cells and the neighbouring 90-99% of cells remain untreated.
Clearly in a hyperproliferative condition it is desirable to m~Yimice cell kill and so the ability to kill uninfected neighbouring cells is extremely i.l.~ullant.

SUBSTITUTE SH EET (RULE 26) CA 0220~93 1997-0~-16 An ~lt~rn~tive strategy for Ldrgeued enzyme prodrug therapy which may provide a limited 'neighbouring cell kill' is antibody directed enzyme prodrug therapy (ADEPT) which involves linkage of the heterologous enzyme to a l~ ~tLing antibody which directs and ~tt~rh~s eYtr~ rly the enzyme to a target cell population where eYtrace~ r conversion 5 of a prodrug may occur. Active drug may then diffuse and exert its effect in neighbouring cells. However this approach, described for example in EP-A-0 382 411 and WO-A-88/07378, may result in an immlme reaction to the heterologous enzyme/~ h~g antibody complex and due to the large size of the antibody/enzyme complex, delivers a limited amount of enzyme to the target site which may be incufficient to provide a signific~nt therapeutic 10 effect.

Additionally GDEPT and VDEPT provides some degree of restriction of the resultant chemotherapeutic activity to a target cell population by either the use of retroviruses which ~r~felc:lltially infect dividing cells, or the presence of transcriptional regulatory sequences 15 whose eA~les~ion is rçstri~t~ to certain cell- or cancer-types e.g. liver-specific promoters for use in the tre~tmPnt of hep~tocP~ r carcinoma. However, even this level of restriction only targets the theld~uLic effect to a particular cell-type, i.e. dividing cells, liver cells, etc., whether they be cancerous or healthy. When the delivered chemoth~ )e~ll;c is a cytotoxic agent, it is clearly desirable that the restriction on the cells being targetted be pathology-20 specific and that no cignific~nt ~plcssion of the chemotheldl)euLic agent occurs in non-pathological healthy cells. One object of the present invention is to subst~nti~lly meet this desirable re.luirel,lent by use of a system that is de-cig~ed to exert a therapeutic effect ~re~erellLially in cells with the pathological disorder.

25 The present invention provides a molecular chim~Pra for use with a prodrug, the molecular çhim~Pra comrricing a transcriptional regulatory DNA sequence capable of being activated in a targetted m~mm~ n cell and a DNA coding sequence operatively linked to the tranc~riI)tional regulatory DNA sequence and encoding a signal peptide and a heterologous enzyme, such that on w~l`esaion of said coding sequence in the targetted cell, the SUBSTITUTE SHEET (RULE 26) CA 0220~93 1997-0~-16 heterologous enzyme is capable of passing through the cell plasma membrane and is capable of catalysing extr~cPll~ r conversion of the prodrug into a ~;y~OtOAiC or cytostatic agent.

Thus acco~ g to the invention, mol~p~ r chirn~p~c are introduced into target cells where S CA~ e~ion of the coding sequence which they contain under control of the tr~n~criptional regulatory sequence produces an enzyme having a signal peptide so that the enzyme is exported through the cell mPmbr~ne ~(lmini~tPred prodrug is catalysed to active colllpound PYtr~el~ rly by means of the enzyme and diffuses into neighbouring cells, where it exerts its the~dl)eulic effects According to one emboriimpnt of the invention the heterologous enzyme is obtained from an enzyme ~r~cul~or by cleavage at a proteolytic cleavage site by a pathology ~co,-;~
protease eytracellul~rly to the targetted m~mm~ n cell.

15 Enzyme ~ ;u~OlS which require to be proteolytically cleaved into a mature enzyme before being capable of cAples~ g ~ignific~nt catalytic activity occur naturally in the form of proenzymes or preproenzymes. ~ltPrn~tively they may be ge-nP-ratPA artificially using recombinant DNA te-chnology to produce an enzyme tr~ncl~ted with an inhibitory peptide linked geneti~lly or by recombinant DNA to the enzyme via a pl-Jtease sensitive amino acid 20 or amino acid sequence such that the catalytic activity of the enzyme is inhibited until removal of the inhibiting peptide via cleavage at a proteolytic cleavage site by a pathology-~oci~tP~ protease. All these relatively inert enzymes whether naturally-occurring or artificially created, are enzyme ~ ;U'~ i of use in the present invention and may be ~efel~ed to herein as proenzymes.
In l,rere,lcd embor~impnt~ of the present invention the enzyme precursor comprises the enzyme and a secretion signal sequence which f~rilit~tPs the extr~cPll~ r transport of the enzyme during which process the signal sequence is cleaved to produce extr~Pllul~r mature enzyme.

SUBSTITUTE SHEET (RULE 26) CA 0220~93 1997-0~-16 A pathology-~ccoci~tçd protease is any protease which is either expressed exclusively in association with a cell population as a consequence of a particular cellular pathology or one which is ~;A~,~ssed at elevated levels correlating to a particular cellular pathology. Elevated levels of proteases are ch~r~rt~rictic of a range of tumour types and have been clearly S implic~tPd in the tumourogenic and particularly the mPt~ct~tic process (FASEB 7, 1434-1441 (1993)). FY~mples of such tumour-associated proteases include urokinase, gel~tin~ce (e.g.
72kD) and stromolysin.

According to the invention such pathology-~csoci~ted proteases provide a mech~nicm for 10 elevating the fidelity of targetted enzyme prodrug therapy by exerting control over the post-tr~ncl~tion~l mof~ifiç~tion of the enzyme such that unless and until the enzyme is cleaved by a p~thology-~ccoci~ted protease, it is unable to express its catalytic activity at all or at least to a s--ffi~i~nt degree to obtain theldpeulic effect. Such catalytic activity is necess~ry to convert the prodrug to a CylOtO,~iC or cytostatic compound. Therefore the toxic agent is 15 ~lef~ tially, if not exclusively, generated in association with target cells which are sSing a pathology-associated protease as a consequence of infl~mm~tion, viral infection, tumoulogenesis, etc.

In a cancer patient the met~ct~tic process often produces secondary tumours and it is these 20 secondary tumours which frequently are the eventual cause of death. Studies in vivo have dçmonctr~t~d that the pharmacological inhibition of the activity of the secreted matrix metalloploleases results in prevention of m~t~ct~cic (Reich, et al Cancer Res 40 3307-3312 (1988)).

25 The matrix metalloproteases (MMP) are a family of zinc metallo-enzymes with well cl ~r~cterice~ structural and catalytic pl-~pellies. These enzymes are all secreted as pro-enzymes that undergo proteolytic cleavage of an amino-terminal domain during activation and are further defined by sequence similarities, inhibition by metal chelators, and inhibition by tissue inhibitors of metalloproteases. Well char~ct~rice~ members of the family include SUBSTITUTE SHEET (RULE 26) CA 0220~93 1997-0~-16 WO 96/16179 ~ PCIIGB95/02716 inlcla~iLial collagenase, neuLlopllil collagenase, stromelysin-l, stromelysin-2, matrilysin, gel~tin~ce A and gel~tin~ce B. Studies have clearly demonctr~t~d a positive correlation beLwæn MMP eA~ ;on, invasive behaviour, and m~t~ct~tie potential. Studies of MMPeAl)r~aaion in human lullluula suggest that these proteases are an inl~oll~lt co,l,ponent of the 5 invasive phenotype of many tumours, inclll~in~ breast, pluaLldte~ colon, lung, ovarian, and thyroid cancers. Hence particularly ~ lled heterologous enzymes pleculaola for use acco~ling to the particular embodiment of the present invention include those ~osse,c;nE a protease cleavage site capable of cleavage by a matrix metalloprotease as described herein.
Such protease cleavage sequences are well known to those in the art and may either be 10 present naturally in a suitable proenzyme of a heterologous enzyme suitable for use according to the embo~liment of the invention or may be Pngin~Pred into such a proenzyme using recombinant DNA technology. Tumours particularly suitable for tre~tm~nt according to the invention include those cA~l~asing such matrix-metalloproteases.

15 Particularly p~llcd are pro-enzymes which can be cleaved to mature enzyme by coll~Eton~ce, int~ tlinE those ~o~c~;ng dipeptide protease cleavage sites compricinE one of Gly-Leu, Gly-Val, Gly-Ile, Ala-Leu, Ala-Met, His-Leu, Phe-Leu, Pro-Val, Ala-Val, Pro-Ile, Gln-Phe, Phe-Val, Val-Leu, Met-Leu, Phe-Leu, Pro-Met, and Ala-Leu; those which can be cleaved by stromelysin in~ 1in~ those with sites comprising Glu-Val, His-Phe, Asn-Phe, 20 His-Ile, Ala-Glu, Pro-Met, Ala-Leu, Arg-Ser, Gly-Phe, Gly-Leu, Phe-Tyr and Gly-Phe and those which can be cleaved by gel~tin~C~c/type IV B collagenases inclll-iinE those with sites compricinE Gly-Val, Gly-Leu, Gly-Gly, Gly-Glx, Gly-Phe, Gly-Asn, Gly-Ser and Asn-Tyr.

The term 'heterologous enzyme' as used herein means any enzyme not present naturally in 25 the targetted m~mm~ n cell. This comprises non-m~mm~ n er,zy,nes such as those derived ~rom yeast or b~tPri~ and m~mm~ n enzymes including naturally occl~rring mutant ...~.,....~li~n enzymes or mutant m~mm~ n enzymes which have been generated by recombinant DNA technology.

SUBSTITUTE SHEET (RULE 26) WO 96/16179 PCI~/GB95102716 ~crclled enzymes for use according to the present invention include any capable of passing through the cell membrane and which have a catalytic activity ap~lo~liale to the conversion of a prodrug to a th~ peul ir~lly active co"lpou,ld. Such enzymes include cytosine de~mina~e which converts the prodrug 5-fluorocytosine to toxic 5-fluorouracil, human c~l,o"y~lidase 5 Al which converts the prodrug para-N-bis(2-chloroethyl)-aminobenzoyl gl~-tamic acid into benzoic acid mustard, the enzyme alk~linç phophatase which converts the prodrugsetoposidephosph~tr, doxorubicin phosph~tç and mito"lycin phosphate into the COl~* onding toxic dephosrhorylated metabolite and the enzyme penirillin-V-~mid~e which converts a prodrug which is a phenyl~ret~mi~e derivative of doxorubicin or mrlrhalan into its 10 collc*,onding toxic metabolite. Other plcfellcd non-m~mm~ n enzymes include ~B-l~$t~mace which may be used with prodrugs ~isrlosed in EP-A-0 404 070 and WO-A-94 01137.

The most ~eÇcllt;d enzyme for use in accordance with the present invention is a mutant .. ~.. -alian enzyme as desrribed in WO-A-95 13095. The mutant m~mmalian enzyme is preferably a wild-type m~mm~lian enzyme with one or more mut~tion~ which gçn~rate novel substrate .~perificitPs. This mutation permits a subtle shift in the substrate specificity of the enzyme such that the a~ ,pliate prodrug may be catalysed while being refractory to catalysis by the c~,,cs~ol-ding endogenous wild-type enzyme.
The ec~enti~l char~ctrrictic of a mutant mamm~ n enzyme of the present invention is the presence of a mutant substrate-binding site i"c~ecli~e of the amino acid sequences fl~nking this mutant substrate-binding site. Hence a chim~ric enzyme comprising a mutant substrate-binding site and flanking sequences derived from one or more dir~clent enzymes is included 25 with the dçfinitio~c of a mutant m~mmali~n enzyme of the present invention. Such a çhim~Pric enzyme may be g~nrrat~ by recombinant DNA technology and/or protein enginP~.ring.

Enzymes suitable for directed mutagenesis include any enzymes po~escing a catalytic activity SUBSTITUTE SHEET (RULE 26) CA 0220~93 1997-0~-16 WO 96/16179 P~ 1 I~D5~/027 16 capable of converting a prodrug to a drug. Such catalytic activities include tr~ncf~ce, hydrolase, oxido-reduct~ce~ isomerase, lyase, or ligase. The directed mutagenesis will gpnPr~t~o a novel substrate specificity but preserve the class of catalytic activity involved. For eY~mpl~ a mutant isomerase of the present invention will possess a novel isomerase activity 5 sllffit~i~ntly different from the isomerase from which it was derived to ensure that prodrugs susceptible to activation by the mutant isomerase remain subst~nti~lly stable in the presence of the isomerase from which the mutant enzyme was derived. This rlr~m~tic shift in activity may be achieved by the ~ltPr~tion of as little as one residue at the catalytic site and the ~lt~r~ti~n of the minimmn number of residues neceS$~ry to obtain the required shift in 10 activity ensures continuity of enzyme structure.

A pftre~led mutant enzyme for use in the present invention is mutant m~mm~ n carbo~yy~pLidase A (CPA). This enzyme has the general activity of cleaving carboxy-terminal aromatic and aliphatic amino acid residues and has been çh~r~rtericed in a diversity 15 of species and tissue-specific variants. Particularly plefell~d mutant carbo~yyeplidases include mut~ntc derived from human pancreatic carboxypeptidase Al (Catasus et al Biochem.
J. 287. 299-303, (1992)), human mast cell carboxypeptidase A (Reynolds et al J. Clin.
Invest. 89 273-282, (1992)) and human pancreatic carboxypeptidase A2. It will bea~plcc;ated that the common ch~r~cterictic of these carbo~yyep~idases is the presence of a 20 CPA-like substrate-binding pocket and ~ccoci~ted enzymatic activity irrespective of overall sequence or structure of the enzyme and that any enzyme pocc~cc;llg this CPA-like substrate-binding pocket is amenable to mutation to a mutant c~bo~y~e~Lidase of the present invention.

25 In a particularly prert~ d embodiment the mutant enzyme is mutant human pancreatic carbo~yy~tidase Al or A2 (CPAl or CPA2) and yet more pleÇel~bly, CPAl or CPA2 wherein amino acid substit~lti~nc are generated at one or more of residues 203, 210, 242, 244, 250, 253, 255, 267, 268, 269 and 305, of the amino acid sequences of the wild-type (w.t.) enzymes. Particularly yl~f~led combinations of residue substitution include Gly at SUBSTITUTE SHEET (RULE 26~

CA 0220~93 1997-0~-16 WO 96/16179 PCTt~ 95~2716 residues 250 and 268 when 253 and 255 are w.t.; Gly at 253 and 268 when 250 and 255 are w.t.; Gly at 250 and His at 268 when 253 and 255 are w.t.; Gly at 250 when 253, 255 and 268 are w.t.; Ala at 255 and His at 268 when 250 and 253 are w.t. and His at 268 when 250, 253 and 255 are w.t. The most p-crcll~d m~lt~ntc are carbo~y~cplidase Al or A2 S mut~ntC compriCing a single substitution; Gly at 268.

Human pancreatic carboxypeptidase A is CA~lCSs~ as a preproenzyme which is ~lucessed to a proenzyme and subsc~luently to the mature enzyme. Mutant c~l.o~y~idases for use according to the present invention may be ",~ of the preproenzyme, the proenzyme or 10 of the mature enzyme but are preferably mutants of the mature enzyme. These mllt~ntc may be derived from either preproenzyme, the proenzyme or the mature enzyme.

A mutant enzyme for use in the present invention may be generated from the DNA or RNA
source of any enzyme ~ssess;ng the previously ~iccuscPd activities by metho-lc well known 15 in the art of molecular biology.

The sPl~Pction of the e";~y,.,alic activity and the primary sequence of a mutant enzyme for use in the present invention will clearly depend upon the nature of the relevant prodrug and, if a prodrug of the type which is activated by the removal of a prodrug moiety, on the precise 20 sLluclulc of that moiety.

The ~uLsLl~le-binding or active site of the mutant enzyme must, unlike the COll~ i~onding non-mutant enzyme, be capable of interacting with the prodrug in such a way that enzymatic catalysis is f~-ilit~tPA The ability of the mutant enzyme to perform this activity will depend 25 upon subtle ~ltP~ionC in the 3-dimPncional structure of the active site, which is in turn flepen~Pn~ upon the primary amino acid sequence of this region of the protein. Alterations of the plilll~y sequence of an enzyme by standard techniques of protein enginePrinE will allow the gen~ ;on of an a~lialc mutant enzyme for use according to the invention with a collc~onding prodrug.

SUBSTITUTE SH EET (RULE 26) CA 0220=,=,93 1997-0=,-16 WO 96116179 P~ ~b5slo27l6 To be capable of secretion through the cell membrane, an enzyme should possess a signal sequence at the amino tellninus either because it is a secreted enzyme with a naturally oc~u~ g signal sequence or because the chim~Pra eA~l~s~ih~g the enzyme has been en~in~red such that the ~ ;ssed enzyme has an ~ tion~l amino acid sequence whichS ~o5c-~s~s the plu~ellies of a signal sequence. Such signal sequences are well known in the art and include those intlic~t~d or derivable from Nothwehr et al J. Biol. Chem. 265, 21797-21803 (1990), Nolhwt:hl and Gordon J. Biol. Chem. 265, 17202-17208 (1990) and Kohara et al FEBS Lett. 311, 226-230 (1992) and the references cont~ine~ therein.

10 Prodrugs for use according to the present invention may be any compound which upon catalysis by an enzyme suitable for use in the present invention, will generate a chP~othel~eulic compound. Such prodrugs will generate chemotherapeutic compoundswhich are preferably anti-infl~mm~tory, anti-viral or anti-cancer compounds, will more preferably be cytotoxic colll~ollnds such as nitrogen mustard agents, antifolates, nl-cleoci~e 15 analogs, the vinca ~lk~ s, the anthracyclines, the mitomycins, the bleomycins, the .;~loto,.ic nucl~Qci(les~ the pteridine family of drugs, the podophyophyllotoxins, the sulfonylureas (as described in EP-A-0 222,475) and low-molecular-weight toxins such as the trichothecçnes and the colchi~inPs Particularly including doxorubicin, daunorubicin, aminopterin, methotrexate, taxol, methopterin, dichloromethoL~ ate, mitomycin C,20 porfirmoycin, S-fluorouracil, 6-me~aplopuiine, cytosine arabinoside, podophyllotoxin, etoposide, melphalan, vinblastine, vin~-rictin~, desacetylvinblastine hydrazide, leurosidine, vindecine, leurosine, trichothecene and desacetylcolchicine.

Particularly p,er~lled prodrugs for use in the present invention, and particularly for use with 25 the ~ltÇellcd carboxypeptidase mutant enzymes described herein are described in WO-A-95 13095.
.

A mutant human enzyme as used herein shall be taken to be any human enzyme with a sequence differing by at least one amino acid from the amino acid sequence or sequences of SUBSTITUTE SHEET (RULE 26) that enzyme in the patient to which the therapy is applied. A "chemoth~ c~ c agent"
which may also be referred to herein as a "drug" inrl11des any mo1ecu1e which has activity in human therapy. Such chemoth~d~euLic agents include but are not limited to cyLusLdtic or .;~/LotoAic compounds used in the therapy of cancers or viral infectionc A "functi( n~lly S inactive pl~;Ul~Ol" which may also be referred to herein as a "prodrug" includes any col,lpound which may be converted into a chemotherapeutic agent under the action of an enzyme. Such fi1nctinn~11y inactive precursors may typically be converted to a chemothP ~peul;c agent by the en~y",atic cleavage of the functi~ 1ly inactive pl~;ul~or to yield a chemoth~ e.~lic agent and a "prodrug moiety". Such conversion from f1-nCtiC n~11y 10 inactive plccul~or to chemotheldpc;ulic agent may also occur by enzym~tit~lly me~ ted isom~ric~ti~

A function~lly inactive plcc;ul~or will generally not exhibit clinic~11y ~ignific~nt levels of the th~ u1 ic activity possç~ed by the chemotherapeutic agent into which it may be 15 e"~ 11y converted for example as a result of having been chemie~lly derivitized to decrease its normal pharmacological activity. The functionally inactive p~c~ui~or of a ~Lùtu~ic chemoth~ peulic agent will not itself exhibit clinically ~ignific~nt cytotoxicity and will be snfficiently stable in vivo such that during therapy, cliniç~lly .~ignifi~nt levels of ~;y~Otu~iCity are largely only generated at the site of conversion of functionally inactive 20 ~ uul~or to chemotherapeutic agent i.e. at the site of the targetted m~mm~ n cell.

In the mole,cu1~r chim~ra according to the invention, the coding sequence is under the control of a tr~n~rriptional regulatory sequence (TRS) compAsing at least a promoter and pl~ bly an ~nh~n-~çr, each of which may either be capable of non-specific eA~ s~ion 25 intleppndent of the type of cell in which e-.plcs~ion is occurring or may exhibit a selectivity of eAl,lt;ssion ~iepPn~ent upon the cellular environment. Preferred TRSs are non-specific, potent promoter/enh~ncer combinations such as cytomegalovirus promoter/enh~ncPr, SV40 promoter/enh~n~Pr and retroviral long terminal repeat promoter/enh~ncer. Other prefelled TRSs include those of ~B-actin, glyceraldehyde-S-phosphate and tubulin.

SUBSTITUTE SHEET (RULE 26) Also ~rc~cllcd are TRSs exhibiting cell-type dependPnt cAlJlc~ion in which case the selection of the TRS, in particular the promoter and çnh~ncer sequence, will depend on the targetted cell type. Fy~mples include the albumin (ALB) and alpha-fe~lutein (AFP) TRS for normal h~ tc~y~s and transformed hep~ yLcs lc~pecLi~rely~ the TRS for carrinoemhryonic antigen 5 (CEA) for use in transformed cells of the ga~Llointe,~ l tract, lung, breast and other tissues:
the TRS for tyrosine hydroxylase, choline acetyl transferase or neuron specific enolase for use in neuroblastomas: the TRS for glial fibro acidic protein for use in glioblastomas and the TRS for insulin for use in tumours of the pancreas. Further examples include the TRS
specific for gamma-glutamyltranspeptidase for use in certain liver tumours and dopa 10 decarboxylase for use in certain tumours of the lung.

In ~ ition the TRS for certain oncogenes may be used as these are c~ cssed predominantly in certain tumour types. These include the HER-2/neu oncogene TRS which is expressed in breast tumours and the TRS specific for the N-myc oncogene for neuroblastomas.
The ALB and AFP genes exhibit extensive homology with regard to nucleic acid sequence, gene structure, amino acid sequence and protein secontl~ry folding (for review see Ingram et al Proc. Natl. Acad. Sci. (USA) 78, 4694-4698 (1981)). These genes are in~ependlontly but lecil.iocally eAylcssed in ontogeny. In normal development ALB tr~n~rrir)tion is initi~t~d 20 shortly before birth and continues throughout adulthood. Tr~n~rrirtional c~ ssion of ALB
in the adult is confinrd to the liver. AFP is normally cA~lc~scd in foetal liver, the visceral en~odPrm of the yolk sac and the foetal gastrointestin~l tract, but clerlin~-s to lm(letert~hle levels shortly after birth and is not signifir~ntly e,~lessed in nonp~thogenic or non-regenrr~ting adult liver or in other normal adult tissues. However, AFP transcription in 25 adult liver often increases riram~tir~lly in hept~r~llnl~r carcinoma (HCC). In ~d~itio~
tr~n~rriI)tion may also be elevated in non-seminomatous and fixed carcinoma of the testis:
in ~n~o~l~rmal sinus tumours in certain ~;l Lul~ cinomas and in certain gastrointestin~l tumours. Liver-specific c,L~rcssion of AFP and ALB is the result of interactions of the regulatory sequences of their genes with trans-activating transcriptional factors found in SUBSTITUTE SHEET (RULE 26) WO 96/16179 ~ ~D5SIO2716 nuclear extracts from liver. The AFP and ALB TRSs are preferred for gen~tin~ he~o~
specific or general liver-specific ~pression f~s~e~;Lively of molesul~rly combined genes since - the AFP and ALB genes are regulated at the tr~nc~rirtional level and their mRNAs are among the most abundant polymerase II Sr~n~crirts in the liver.
c 5 Several m~mm~ n ALB and AFP promoter and enhancer sequences have been identified(for review see Genes and Develop 1, 268-276 (1987); Science, 235, 53-58 (1987); J. Biol.
Chem. 262, 4812-4818 (1987)). These sequences enable the selective and specific eA~ sion of genes in liver hep~tocytes (normal and transformed) and hepalo"las respectively.
Similar to the regulatory structure of the ALB gene the regulatory elem~ntc of the AFP genes promote tissue-specific e~lession in certain liver pathologies, such as HCC (Mol. Cel. Biol.
6, 477-487 (1986); Science, 235, 53-58 (1987)). The regulatory elements of a m~mm~ n AFP gene consist of a specific 5' promoter proximal region (located in some m~mm~ n 15 species be~ eell 85 and 52 bp 5' to the gene). This sequence is escPnti~l for transcription in hepatomas. In ~ itic)n there are u~lr~alll (5') regulatory el~mentC well defined for the murine AFP gene which behave as classical Pnh~ncPrs (Mol. Cell Biol. 6, 477-487 (1986);
Sri~pn~e~ 235, 53-58 (1987)). These upstream regulatory elemPntc are desiEn~tP~ elemPntc I, II and III and are located between 1,000 to 7,600 bp 5' to the fr~nccrirtion initiation site 20 for the AFP murine gene. These three enh~ncer domains are not f~lnction~lly equivalent at genP~tin~ tissue-specific e:~p~ssion of AFP. FlPmentc I and II have the greatest capacity to direct liver-specific e,.~ression of AFP. It is illlpol~nt to note that the regulatory s~uences of the alpha-r~:loplolein gene advantageously contain the sequences not only for tissue-specific tr~nccrirtional activation but also for repression of e~res~ion in tissues which 25 should not express AFP. In a similar fashion the regulatory regions of the human alpha-felo~rulein gene have been char~tPriced (J. Biol. Chem. 262, 4812-4818 (1987)). A
structural gene placed in the correct oriPnt~tic~n 3' to the AFP regulatory sequences will enable that structural gene to be selectively e~ressed in fetal liver h~ato~as, non-S~IIIn~ OLIS carcinomas of the testis, certain teratocarcinomas, certain gastrointestin~l SUBSTITUTE SHEET (RULE Z6) CA 0220~93 1997-0~-16 tumours and other normal and pathological tissues which spe~ifi~lly express AFP.
The promoter and çnh~nrer sequences preferably are select~-d from the TRS for one of albumin (ALB), alphaÇelopl~cill (AFP), carcinoembryonic antigen (CEA) a. DNA
5 Sequencing and Mapping, Vol 4, 185-196), tyrosine hydroxylase, choline acetyl transferase, neuron-specific enrl~cp~ glial fibro acid protein, insulin or gamaglutamytranspeptidase, dopadecarboxylase, HER-2/neu or N-myc oncogene or other suitable genes. Most l"ert;ldbly the TRS for ALB or AFP are used to direct liver specific or hepatollla specific e,~les~ion respectively.
In ~lc~t;lled tlllbo~ ,Pnt~ of the present invention the molecular çhim~Pr~ is selectively eA~l~ssed in a target cell population. This may be taken to mean that the chim~Pra is e~ressed at a higher level in the target than in the non-target cell population and is preferably ~ ressed predomin~ntly or exclusively in that population. Selective eA~ s~ion 15 may be achieved by inclusion of a target-cell specific TRS (promoter with or without enhancer) as described above or may be a product of the method of delivery of the çhim~
to the target cell. Methods capable of providing target cell specific delivery of the chim7~Pr~
with subsequent stable integr~ti()n and c~re~:,ion, include the techniques of c~lcium phosphate tr~n~fection~ ele~;Ll~ol~tion, microinjection, liposomal transfer, ballistic barrage 20 or retroviral infection or lnfection using adenovirus or adeno-~cco~i~tPd virus. For a review of this subject see Biotechniques Vol. 6 No. 7 (1988).

Such selectivity may be obtained by a variety of such techniques. Physiologically localised delivery of the Ch;lll~ d for the target cells will reduce the possibility of non-target cells 25 c~r~ssing the ~.:hil~ This may be achieved when for eY~mrle using retroviral or liposome mPAi~tP~ delivery and would involve direct injection to a blood vessel known to supply the target cells. Selectivity may also be obtained using retroviral me~ tPd chim~Pra delivery in the therapy of hyperproliferative disorders. Retroviruses only infect dividing cells and would therefore only introduce chim~Pras to dividing cells. Liposome technology SUBSTITUTE SHEET (RULE 26) CA 0220~93 1997-0~-16 permits the delivery of the chim~Pra cont~in~P-d therein to be targetted to a particular cell type based on ~,o~,iaLe modifications made to the liposome coat structure. In a p,efe,led - embodiment according to the present invention, a number of such methods for obL~ining selectivity will be comlJinp~ to improve the fidelity of selective ~ ssion. For eY~mrle in S the h~t~ t of hPp~tocP~ r carcinoma, the chim~Pra may comprise TRSs derived from liver-specific gene promoters such as ALB or AFP, and will be delivered in a retrovirus directly to the hepatic artery. Hence a three-fold level of speçifi~ity will be obtained, firstly by the localised delivery which ensures that effectively all the retroviruses infect liver cells, secondly by use of a retrovirus which only infects dividing carcinoma cells and thirdly by 10 liver-specific ~,es~ion of the TRSs.

One particular method according to the present invention for obtaining selective e,.l"ession of a mole~ul~r çhim~Pra of the present invention delivered using a retrovirus is accompliched by promoting selective infection of liver cells. This technique involves the retroviral env 15 gene present in the p~k~ging cell line which defines the spe~-ificity for host infection. The env gene used in constructing the p~k~gjng cell line is mo~ifi~d to generate artificial infective virions that selectively infect hepatocytes. As an example a retroviral env gene introduced into the p~rk~ging cell may be modified in such a way that the artificial infective virion's envelope gl~coprotein selectively infect hepatocytes via the specific receptor 20 mPAi~tP~ binding utilised by the h~p~titic B virus (HBV). HBV primarily infects hep~tocytes via specific ,~ce~Lor mP~i~tP~ binding. The HBV proteins encoded by the pre-S 1 and pre-S2 sequences play a major role in the ~tt~chment of HBV to hepatocytes (see ~Ppa~ln~ Viruses edited Robinson et al 189-203, 205-221 (1987)). The env gene of the pack~ging cell is modified to include the hPp~focyte binding site of the large S HBV envelope protein. Such 25 morlifi~tion~ of the env gene introduced into the p~ ging cell may be performed by standard molP~ r biology techniques well known in the art and will f~cilit~tP viral uptake in the target tissue.

ln the methotls of obt~ining selective activation which comprise target cell specific delivery SUBSTITUTE SHEET (RULE 26) systems, the TRS need not be target cell specific and TRSs derived from genes such as ~-actin, glyc~pra~ phyde-3-phosrhatp and cytomegalovirus (e.g. imme~i~tP early gene) (see Huber, e~ al Cancer Research, 53, 4619-4626 (1993)) and references therein) may be used.

5 The mol~ular chimaP~ of the present invention may be made utili~ing standard recombinant DNA techniques. Thus the coding sequence and polyadenylation signal of for example the cytosine d~ atP gene is placed in the proper 3' orientation to the ALB or AFP TRS.
These molecular cl-;.l.~ enable the selective ~AlJlG~ion of cytosine f~ minatP in cells which normally express from ALB or AFP TRSs respectively.
Accordingly there is also provided a method of constructing a molecular chima~P-ra comprising operatively linking a DNA sequence compricing a TRS capable of being activated in a mamm~lian cell to a DNA sequence encoding an enzyme capable of passing through the cell membrane and capable of catalysing the conversion of the prodrug into a ~;ylot~l~ic or 15 Cylo~Lic agent.

The technique of retroviral infection of cells to integrate artificial genes employs retroviral shuttle vectors which are known in the art (see for example Mol. and Cell Biol. 6, 2895-2902 (1986)). F~entially retroviral shuttle vectors are generatPd using the DNA form of the 20 retrovirus containP~ in a plasmid. These pla~mi~ also contain sequences ne~e~ry for s~ *c-~ and growth in b~cte~i~ Retroviral shuttle vectors are constructed using standard m-)lecular biology techniques well known in the art. Retroviral shuttle vectors have the parental endogenous retroviral genes (e.g. gag pol and env) removed and the DNA sequence of interest inserted, such as the molecular chimaeras which have been described. They 25 however contain appl~liate retroviral regulatory sequences for viral encapsidation, proviral insertion into the target genome, message splicing, termination and polyadenylation.
Retroviral shuttle vectors have been derived from the Moloney murine leuk~emia virus (Mo-MLV) but it will be appreciated that other retroviruses can be used such as the closely related Mol<~ney murine sarcoma virus. Certain DNA viruses may also prove to be useful as a SUBSTITUTE SHEET (RULE 26) ~ ~ =
CA 0220.7.793 1 997 - 0.7 - 1 6 WO 96/16179 PCT/~b5~,J'~2716 delivery system. The bovine papilloma virus (BPV) replicates eYtr~chromosomally so that delivery system based on BPV have the advantage that the delivered gene is m~int~ined in - a nonint~gratPd manner. Adenoviruses and adeno-~cco~i~ted viruses may also be used.

5 Thus according to a further aspect of the present invention there is provided a retroviral shuttle vector con~i~;ning a molecular chim~Pra as hereinbefore definecl.

The advantages of a retroviral-mPAi~ttoA gene transfer system are the high PfficiPn~y of the gene delivery to the targeted tissue sequence specific integration rega~di,lg the viral genome 10 (at the 5' and 3' long terminal repeat (LTR) sequences) and little rearrangements of delivered DNA colllpaleA to other DNA delivery systems.

Accordingly in a pl~felled embodiment of the present invention there is provided a retroviral shuttle vector compricing a DNA sequence comprising a 5' viral LTR sequence, a cis acting 15 psi encapsidation sequence, a molecular chimaera as hereinbefore defineA and a 31 viral LTR
sequence.

In a plere-l~A embodiment and to help elimin~te non-target-specific ~ s~ion of the molecular chim~Pra, the molecular chim~era is placed in opposite tr~ncçriptional orient~tion 20 to the 5' retroviral LTR. In ~ itic)n a dominant sele~t~hle marker gene may also be inclllded which is tr~ncc7iptionally driven from the 5' LTR sequence. Such a dominant s~le~ct~hle marker gene may be the b~ctPri~l neomycin-resict~nce gene NEO (aminoglycoside-3-phosphotr~ncfer~ce type II) which confers on eukaryotic cells recict~nce to the neomycin analogue G418 sulphate (Gen~ticin - trade mark). The NEO gene aids in the selection of 25 p~ck~ging cells which contain these sequences.

The retroviral vector used may be based on the Moloney murine leuk~Pmi~ virus but it will be a~pleciated that other vectors may be used. Such vectors cont~ining a NEO gene as a c~ole~t~hle marker have been described, for example, the N2 vector (Science, 230, 1395-1398 SUBSTITUTE SHEET (RULE 26) CA 0220~93 1997-0~-16 (1985))-A theoretical problem ~Ccoci~tçd with retroviral shuttle vectors is the pot~,llial of retroviral long tPrmin~l repeat (LTR) regulatory sequences transcriptionally activating a cellular S oncogene at the site of int~r~tion in the host genome. This problem may be iiminichçd by creating SIN vectors. SIN vectors are self-inactivating vectors which contain a deletion comprising the promoter and enh~ncer regions in the retroviral LTR. The LTR sequences of SIN vectors do not tranccrirtionally activate 5 or 3 genomic sequences. The transcriptional inactivation of the viral LTR sequences diminichçc insertional activation of 10 ~dj~ce~nt target cell DNA sequences and also aids in the s~lected e~p-ession of the delivered molec~ r chim~era SIN vectors are created by removal of approximately 299 bp in the 3 viral LTR sequence (Biotechniques, 4, 504-512 (1986)).

Thus preferably the retroviral shuttle vector of the present invention are SIN vectors.
Since the parental retroviral gag, pol and env genes have been removed from these shuttle vectors a helper virus system may be utilised to provide the gag pol and env retroviral gene products trans to package or encapsidate the retroviral vector into an infective virion. This is accomplished by utilicing speci~licPd "p~cl~Eing" cell lines which are capable of 20 gener~ting infectious synthetic virus yet are clefil~iPnt in the ability to produce any det~-ct~ble wild-type virus. In this way the artificial synthetic virus contains a chim~e~ of the present invention packaged into synthetic artificial infectious virions free of wild-type helper virus.
This is based on the fact that the helper virus that is stably inlegldted into the p~c~ ing cell c--nt~inc the viral structural genes but is lacking the psi site and cis acting regulatory 25 sequence which must be cont~inPd in the viral genomic RNA molecule for it to be ~nr~pcid~t~ into an infectious viral particle.

Accordingly the present invention provides an infective virion comrricinp a retroviral shuttle vector as hereinbefore described said vector being enc~rsi~l~tçd within viral proteins to create SUBSTITUTE SHEET tRULE Z6~

CA 0220~93 1997-0~-16 WO 96/16179 PCrlGB95102716 an artificial infective repli~tioll-defective retrovirus.

In ~rl~lition to removal of the psi site ~ itiol~l alterations can be made to the helper virus LTR regulatory sequences to ensure that the helper virus is not packaged in virions and is 5 blocl~Pd at the level of reverse transcription and viral integration.

~ltPrn~tively helper virus structural genes (i.e. gag, pol and env) may be individually and in-lepPn~Pntly tr~ncfPrred into the p~ck~in~ cell line. Since these viral structural genes are s~p~ ed within the p~rlr~ginE cell's genome, there is little chance of covert recombin~tionc 10 generating wild-type virus.

In a further aspect of the present invention there is provided a method for procluçing infective virions of the present invention by delivering the artificial retroviral shuttle vector comprising a molecul~r -him~Pr~ of the invention as hereinbefore desçrihe~ into a p~ck~ing cell line.
The p~rk~gin~ cell line may have stably intPgr~tP~ within it a helper virus lacking a psi site and other regulatory sequence as hereinbefore described or alternatively the p~ck~in~ cell line may be el-gi.-PP-~d so as to contain helper virus structural genes within its genome.

20 The present invention further provides an infective virion as hereinbefore dPs~ribe~ for use in therapy particularly for use in the tre~tmp-nt of cancer and more particularly for use in the llc~t~ of hep~toc~Pll~ r carcinoma, non-semin~m~tous carcinoma of the testis, certain Ir~toc~ o...~c and certain gastrointPstin~l tumours.

25 The infective virion according to the invention may be formulated by techniques well known in the art and may be presenLed as a formulation with a ph~rm~ceutic~lly acceptable carrier therefor. Pharm~eutic~l acceptable carriers in this inct~nce may comprise a liquid mPtlinm sl~it~hle for use as vehicles to introduce the infective virion into the patient. -An PY~mple of such a carrier is saline. The infective virion may be a solution or suspPncinn in such a SUBSTITUTE SH EET (RULE 26) WO 96/16179 P~ ~b5slo27l6 vehicle. Stabilisers and ~ntio~ ntc and or other eYcirientc may also be present in such ~h~""~e"~ form~ tionc which may be ~lminiStered to a m~mm~l by any conv~Pntion~lmethod e.g. oral or parenteral routes. In particular the infective virion may be a~minictpred by intra-venous or intra-arterial infusion. In the case of treating HCC intra-hepatic arterial 5 infusion may be advantageous.

Accordingly the invention also provides pharm~- eutic~l forml~l~tionc comprising a molec~ r chim~Pr~ of the present invention con~ ~ within one of, an infective virion or a liposome or a p~rl~ging cell mix, in a~ c with a pharm~feutie~lly acceptable carrier, and10 pharm~-euti-~l formnl~tionc comprising a molecular çhim~Pra virion, vector, liposome or p~ ging cell mix of the present invention in ad~ ulc with a pharmaceuti~lly acceptable carner.

Ad~ition~lly the present invention provides methods of making pharm~ceutical formulations 15 as herein ies~ribe~ comrricing mixing an artificial infective virion co-.~ ing a molecular c~im~Pr~ with a pharm~ceuti~-~lly acceptable carrier.

The invention also includes the use of any molecular chim~er~ vector, virion, liposome or pharm~eutic~l formulation of the present invention in human therapy and in the manufacture 20 of a meAit~mPnt for use in the tre~tment of pathological states.

The invention also int~ll-des mçtho(ls of medical therapy compricing the use of any molecular çhim~Pr~, vector, virion, liposome or pharm~eutic~l formulation of the present invention.

25 Also in~ ed within the scope of the present invention is a protein encoded by a molecul~r cllim~ of the present invention and any combination of such a protein and a prodrug which can be catalysed by the enzyme component of that protein.

The precise dosage to be ~rlminict~red to a patient will l~ltim~t~ly be depçndent upon the SUBSTITUTE SHEET (RULE 26) WO 96/16179 PCI~/GB95/02716 discretion and profeccion~l judgement of the ~ttPn~l~nt physician and will be a product of the particular ~E~c~ g mPrh~nicm chosen. Rerclc. ces col-t~intod herein to the effi~iency of ~Ctlillg of retroviruses, liposo---e etc. may be used to determine a~ru~liate dosage levels.

S The invention is illllCt~t~d by the following eY~mples in which reference is made to the acco...p~,yillg drawings. In the drawings:

FIG 1 is a map of pl~cmid 268G-HCPAI which includes the human 286G-HCPAI in a suitable form for cAl~lcssion in HT1080 cells;
FIG 2 shows inhibition of HT1080 cell growth in cells cA~l-,ssing secreted 268G-HCPAI
with various prodrug tre~tmentc;

FIG 3 shows production of methotrexate from prodrug B (N-(4-(((2,4-~ minQ-6-15 pteridinyl)methyl)methylamino)benzoyl-L-glutan-l-yl-3-cyclobutyl-L-phenyl~l~minp) by culture mP~ m from cells eA~lcssing secreted 268G-HCPAI;

FIG 4 shows cellular location of ,~ ct~m~ce activity in m~mm~ n cells transfected with a se~;lctoly ,~ t~m~cç construct;
FIG 5 shows growth curves of E. coli DH5~x and DH5~ transformed with pCMV-pgal or pCMV-pga2.

FIG 6 shows a schPm~tic rcpresent;~tion of pCMV-pga clones and E. coli pga genomic 25 sequcnce.

SUBSTITUTE SHEET (RULE 26) W O 96/16179 PC~rlGB95/02716 F.Y~ nrl~ 1 (i) F~ g of HT1080 Cell;s To Express 268~HCPA1.
A Hind III-Xba I fr~pmrnt which cont~ined the entire coding sequence of human 268G-5 HCPAlwas inserted into the eucaryotic e,.~r~s,ion vector, pcDNAINeo anvitrogen, Inc.)also restrict~d with Hind III and Xba I. The rec--lting plasmid construct design~t~A 268G-HCPAl (see Figure 1) was transfected into HT1080 cells using the cationic lipid nli,.lul~, Lipofectinsu, (Gibco BRL, Inc.) and recombinant cells were s--lP~ted in G418 (500 g/ml).
Individual cell clones were obtained by the use of cloning rings and were char~ctP-ri7ed by 10 t,he HCPAl enzymatic activity produced in the cell culture mPAium.

(ii) Inhibition of HT1080 Cell Growth in Cells Expressing Secreted 268G-HCPAl with Prodrug Tre~tm~nt HT1080 cells, either control (pcDNA) or .ongineered to express 268G-HCPAl (268G) were 15 plated into 96-well plates in normal medium with 10 % human serum which was p~ edted at pH 3.2 for 2 h at room lW~ dlUlt; followed by neut~ali7~ti~n After 24 h the inriir~ted drug was added at the specified concentration Plates were incubated at 37 C for three days and cell viability was determined using the Ml~ assay. The results are shown in Figure 2.

20 (iii) Production of Methotrexate from Prodrug B by culture Medium from Cells Ex~ g Secreted 268~HCPA1.
HT1080 cells as above were plated into dishes in the in~ t~d mPAil-m and were incub~ted at 37 C for three days. Prodrug B (N-(4-(((2,4-diamino-6-pteridinyl)methyl)methylamino) benzoyl)-L-glutam-l-yl-3-cyclobutyl-L-phenyl~l~nine; 20~M) was added to the cells and 25 aliquots were removed at lh, 3h and 20h for analysis of methotrexate production. The following cells were tested: pcDNA-control cells; 268G-cells e~ essillg 268G-HCPAl;
5%HS-cells plated in 5 % human serum; 5%HS-Acid-cells plated in 5% human serum SUBSTITUTE SHEET (RULE 26) WO 96tl6179 P~ g5~ 716 p~t~eatcd in acid as above.

F.Y~n~rlP 2 Se~--r~ of ~nmon CarL~ ce Al cDNA
S A Hind III - Sal I cDNA fr~gmPtlt çnco-ling the rat pr~n~c~l,oxy~cptidase Al (CPAl) (Gardell et aZ., Nature, 3l 7, 551-555 (1985)) was isolated from pMP36 provided by Dr. M.
Phillips (IJCSF). The 1.2 kb Hind III - Sal I rat CPAl cDNA was radiolabelled with l32P]dCTP and Prime-it kit (Stratagene) and then used to screen a lambda gtll human pancreas cDNA library (Clontech) according to the method of Grunstein and Hogness (Proc.
10 Natl. Acad. Sci. (USA), 72, 5016-5020 (1975)). Purified icol~tPd plaques that hybridised to the rat CPAl cDNA probe were obtained after three rounds of sclæning.

Lambda DNA from purified plaques was prepared from overnight growth liquid cultures using Qiagen columns and according to the manufacturer's protocol (Qiagen). Human CPAl 15 cDAN inserts were liberated from lambda DNA by digestion with EcoRI and purification of the cDNA inserts by low-melting agarose gel electrophoresis. The EcoRI cDNA inserts were cloned into the EcoRI site of pGEM 7z(+) (Promega) and denoted as pHCPA pl~cmitl5 Following overnight growth in liquid culture, plasmid pHCPA DNA was prepared using Qiagen ct l~lmnc DNA seq~len~ing of icol~tPd lambda plaques and of pl~cmi~ls cont~ining HCPAl cDNA was pclru,.l.ed with [35S]dATP and the fmol sequencing system (Promega). Oligonucleotide ~JlilllClS that flanked the cloning site of lambda gtl 1 or the mllltirle cloning sites in pGEM
7(+) and the TA vector (Invitrogen) were used initially to determine cDNA sequence.
25 Oligon~ P~ti~le pAmers that c~llc~olld to regions of the determined sequence were synshPci~P~ which permitted the entire cDNA sequence of both strands of HCPAl to be . determined. The sequence is given in SEQ. ID. NOS. 1 and 2.

SUBSTITUTE SHEET (RULE 26) F.Y~mrl~ 3 (i) C'!(!nin~ of E. coli ~-T.~ ce for ~lm~n Cell Exprescion We have constructed unique DNA constructs cont~ining the b~ tP~i~l b-l~cPm~ce gene which, 5 when delivered to human cells, result in ~A~res~ion of functioll~l b-l~t~m~ce. The advantages of b-l~rt~m~ce as a prodrug activating enzyme are 1) the enzyme is kinPti~ y very effici~Pnt and 2) because of a unique activation mPch~nicm, a prodrug of virtually any drug can be made as an çffic i~nt substrate for the enzyme. The imrli~tinnc of this to cancer therapy is that it permits the use of combination prodrug therapy to counter resict~nce 10 phenomena as well as allows one to choose drugs a~"upliate to the tumor target. For eY~mple, to target lung cancer, prodrugs of methotrexate (5798W93) and 5-fluorouracil (1614W94) have been synth~pci7ed (ii) Con truction of Secretory ,B-I~ct~m~ce Constructs 15 To create a DNA construct which would express secrelo.y ~ t~m~ce in human cells, the coding region of TEM ,B-l~t~t~m~ce (Sykes and Matthews, J. Antimicrob. Chemo., 2, 115-157 (1976); Ambler and Scott Proc. Natl. Acad. Sci. USA, 75, 3732-3736 (1978)) was used.
Since it exists in the pPrirl~cm of b~ctP~i~ the unmo-lifiPrl coding region of TEM ,~-l~Cl~n~ce cont~inc a signal peptide (Sutcliffe, Proc. Natl. Acad. Sci. USA, 75, 3737-3741 20 (1978)). Sequences useful for the cloning and t;~ ssion of this gene in a eukaryote were added to fl~nking sequence during PCR by inrlu~ing the sequences in the PCR primers. The sequence of the fo,~vdrd primer (JM1) was:
5'-TTGCATAAGC:TrGCCACCATGAGTATTCAACATTTCCGTGTC (42-mer - SEQ ID
NO 3). The sequence of the reverse primer aM2) was:
25 5'-GATCTGTCTAGATTACCAATGCTTAATCAGTGAGGC (36-mer - SEQ ID NO 4).
The fol~.l primer cont~inC a Hind III rPctri~tiQn site (AAGCTT) for subsequent cloning of the PCR product, and a sequence (GCCACC) which confers optimal t~nCl~tion effciency SUBSTITUTE SHEET (RULE 26) in vertebrates (Kozak, J. Cell Biol. 115, 887-903 (1991)) immPAi~t~ly 5-prime to the initi~lor mpthioninp~ codon (ATG) of the ~ rt~m~ce coding region. The reverse primer cont~inc an Xba I rest~ctio~ site (TCTAGA) adjac~Pnt to the stop codon (TAA) of the ~B-ce coding region.
S
The PCR reaction was carried out for 25 cycles using standard con-litionc and using Vent DNA Polymerase (New Fngl~nd Biolabs, Inc., Beverly, MA, USA) in 4 mM MgSO4 and 200 ~M of each dNTP and 1 pmoV~l forward and reverse primers. PCR thermal cycling cQnditions were 95C, 1 min; 60C, 1 min; 75C, 1 min, 25 cycles then 75C, S min. The 10 approxim~tP-ly 800 base pair PCR product was gel-purified using the Glass-Max kit (Life Technologies, Inc., Gaithersburg, MD, USA). The purified PCR product was restriction digested with Hind III and Xba I, re-purified by gel electrophoresis, and ligated into the mllltirle cloning site of the pRc/CMV vector (InVitrogen, Inc., San Diego, CA). The oriPnt~tion of the ,B-l~ct~m~ce insert in this vector places the ~ t~m~ce gene under the 15 t~nc~rirtional regulation of the interme~i~tP/early CMV promoter as well as followed a bovine growth hormone poly(A) addition signal. The sequence of the construct (decign~ted pCMV-BL) is shown in SEQ IDS NOS 5 and 6 along with the amino acid sequence of inserted seclcluly~ t~m~ce 20 (iii) Determin~fion of C~ r T oc~tiorlc of Targeted ,B T ~rt~m~ce Protein Confirm~tion of the predicted location of the ,B-l~t~m~ce constructs was carried out using transient DNA transfection in a m~mm~ n cell line. Transfection was carried out by li~osollle-mPAi~ted DNA delivery using lipofect~mine (Life Technologies, Inc., Gaithersburg, MD, USA). FxrerimPntc were pelroll.led according to manufacturer's instructions, varying 25 the number of cells, amount of tr~ncfection reagent, and amount of DNA to determine optinlulll conditions. Typically, 60 x 15mm tissue culture plates cont~ining aRroximately 3 x 105 to 1 x 106 cells were employed. After transfection using pCMV-BL, transfected cells SUBSTITUTE SHEET (RULE 26) were resl-spçn~ed in 50 mM Tris-Cl (pH 7.4), 0.1 mM EDTA cont~ining PMSF and leupeptin, swollen on ice for 10 min, then lysed using a Dounce homogenizer. After centrifugation at 800 x g for 6 min, supernatant (cytosolic fraction) was recentrifuged at 30 psi for 20 .,.inules in a Re~l~m~n AirFuge. Pellets from both centrifug~tinn~ (which include 5 membranes and nuclei) were combined. Each fraction was assayed for activity using the chromogenic substrate PADAC, added to a final concentration of 20 mM (Calbiochem, Corp.). Absorbence at 570 nm was measured using the auto-rate assay of a Kontron Model 9310 spectrophotometer. To assess secreted ,~ rt~m~e levels, PADAC assays were carried out on the cell-free media after transfections. ~ t~m~se enzyme activity was measured 10 using PADAC (Calbiochem, Corp.) which serves as a chromogenic substrate of ~B-l~t~m~ce activity (Sçhin(llPr and Huber, Enzyme Inhibitors, Brodbede, Ed., pp 169-176, Verlag Chemie, Weinheim (1980). A 500 ~M PADAC stock was made in water, filtered through a 0.22 ~m filter, and added to media to give a final concentration of 20 ,uM. Decreases in absorbance at 570 nm were measured using the auto-rate assay of a Kontron UV/Vis15 s~ ;Llo~hotometer.

The data in Figure 4 show that at 48 hours after transfection with lipofect~mine, large amounts of ,B-l~rt~m~e are secreted from cells transfected with pCMV-BL. The cellular activity seen with this construct is presumably the enzyme co~t~ined in secl~to~y granules 20 in the process of being exported. Based on the m~gnitl-~e of this activity, we e~tim~ that the enzyme from the se~iletoly ~B-l~ct~m~e construct re~l~sents 5-10% of total cellular protein made per 24 hours per cell.

(iv) ,B-~ ~ct~m~ce Delivery to Cells Confers Se-LsiLi~ to Cephalosporin Prodrugs A. ,B-I~ ce effi~ ntly activates 5798W93 and 1614W94 Prodrugs of metho~Aa~ (5798W93) and 5-fluorouracil (1614W94) .~i~sent the parent SUBSTITUTE SHEET (RULE 26) drugs linked to cephalothin. The kinetic p~r~mPters of prodrug activation were measured by inrub~ting various cQnrent~tion~ of prodrug with purified ~ rt~m~ç followed by HPLC
analysis to determine the rate of prodrug conversion. ~B-T ~rt~m~P efficiently activates both 5798W93 and 1614W94 with a kC~/KM (specificity constant) of 272 and 67 seclmM-', S res~ecli-rely.

B. C'n~ in~ti~-n of the ,a T._r~ ce Gene with 5798W93 and 1614W94 Confers Toxicity We have evaluated the in-vitro toxicity of the ,B-l~ct~m~ce prodrugs in the presence and 10 ~hs~.nr,e of the ~ rt~m~e gene. Cytotoxicity was qn~ntit~ted by dete~ ining IC50s in treated A549 human lung adenocarcinoma cells using an SRB-based growth inhibition assay (Nair et al., J. Med. Chem. 32, 1277-1279 (1989)).

In the ~bsenre of the B-l~rt~m~e gene, methotrexate was 10-fold more toxic than the 15 methoL.~;,Late prodrug 5798W93, and fluorouracil was 20-fold more toxic than the fluorouracil prodrug 1614W94 (Table 1). When A549 cells which cont~inPd stable intPgr~tPA
copie(s) of the secrel~"y ,s l~rt~m~e gene (A549-BL) were tested, methotrexate and its prodrug 5798W93 were equally toxic (Table 1). This ~ lP-nt implies that the delivery of the ,B-l~rt~m~e gene to tumor cells will make them sensitive to cephalosporin prodrugs.
The relatively small di~rere,ltial between the toxicity of methotrexate and S-fluorouracil and their l~e~ e prodrugs in the absence of the ,B-l~ct~m~ce gene was unexpected. This is bec~--se, for both parent drugs, the mPçh~ni~m of action is well understood and the chPmi~l morlific~tion made by ~tt~rhing cephalothin to these compounds should clearly detoxify the 25 drugs. For eY~mple, transport of methotrexate into cells depends on availability of the terminal glut~m~tP moiety which is blocked in 5798W93. Toxicity of 5-fluorouracil depPn~
on the availability of the Nl group since this group is nP~e~ry for glycosidic bond SUBSTITUTE SHEET (R~JLE 26) WO 96/16179 PCI/GB9!;/02716 forrn~ti~ n and conco,.li~lt nuclPosi~le formation. The Nl group is blocked in 1614W94. It is clear that the observed toxicity of these prodrugs in-vitro reflects some degree of ch~mir~l instability of the ~rudlugs which could result in cignifi~nt breakdown of the prodrugs during the 72-hour incubation utilized in the IC5" de~ tion.
s Support for this notion comes from two lines of evidence. The first is that no toxicity is observed when either prodrug is given to mice at a dose equivalent to an LDIoo for the parental drug. The lack of toxicity in these cases is eypl~ined by the relatively short half life of the drug in-vivo (ttn # 20 minutes) in contrast to the e~osu, c of cells to the prodrug for 10 72 hours in-vitro.

The second line of evidence is shown by direct measurement of in-vitro toxicity by short-term assays (3 hour e~o~le of cells to prodrug). Using a sensitive assay for cell toxicity, a 6-r'Hl-deoxyuridine based assay which measures inhibition of thymidylate synthase and 15 DNA synthesis, we could measure toxicity over time as short as a three hour interval.
During this shorter interval, the differential between prodrug and parent drug increased cignifi~ntly (Table 2). These data are consistent with the idea that the prodrug toxiciti~s reported in Table 1 result from chl-mic~l instability of the prodrugs over the long time-course (72 hours) of those e7crtorim~ontC
(v) A~ r Ev~ t;Qn of Secretory ,B-T ~. I;....~ce in vivo Using Liposome-t~l DNA Delivery Se~;lcloly ~B-l~rt~m~ce and cytosine de~min~ce DNA constructs were cor .pared for antitumour effects in mice bearing subcutaneous (s.c.) A549 human lung adenocarcinoma tumours.
25 Results are shown in Table 3. Plasmid DNA e~cssion vectors encoding either cytosine de~."i.l~cP (CD) or s~ielo~ r~m~ce (BL) under the transcriptional control of the non-specific CMV promoter were encapsidated in cationic liposomes (25~g DNA; 25 nmol SUBSTITUTE SHEET (RULE 26) WO 96116179 1~ ;h55/02716 liposo".es). Mice bearing A549 s.c. tumours were treated with five i~t~ ~tl~..oral injections of liposo...al DNA. Prodrug therapy (1614W94 (50 mg/kg; i.p., qd x 5) or S-FC (500 mg/kg; i.p., qd x 5) was initi~tPd two days after DNA tre~tm~nt Inhibition of tumour growth was determined on day 47. Both CD and BL constructs resulted in similar 5 ~ntitumour activity in vivo. 1614W94 ~mini~tration resulted in about 60% inhibition of tumour growth (Table 3). 5-FC ~r~minictration resulted in about 70% inhibition of tumour growth, whereas DNA liposollles alone and 5-FU alone (25mg/kg, i.p., qd x 5) resulted in only about 20% inhihition of tumour growth (Table 3). Thus, liposol-lal DNA/5-FU prodrug combin~tion~ resulted in s.c. tumour regressions.
Secretory ,~ t~m~ce and cytosine cl~min~e DNA constructs were also evaluated by int~thoracic (i.t.) injection of liposomal DNA into the pleural space of mice bearing tumors.
Results are shown in Table 4. Mice bearing human large cell lung H460 i.t. tumours received DNA encoding either CD or BL under the transcriptional control of the CMV5 promoter. DNA was dosed by i.t. injection on days 6, 7, 12 and 13. Prodrugs for the e enzyme were dosed on days 7-16 (5-FC, 500 mg/kg; 1614W94, 70 mg/kg; i.p., qd x 10). Animal survival was evaluated 30 days after tumour implantation. All nontreated mice and mice treated with 5-FU (30 mg/kg i.p., qd x 5) died from tumour by 30 days.
CMV-BL/1614W94 trP~tmPnt increased survival to 60%, and CMV-CD/5-FC tre~tmPnt also0 increased the survival to 40% CTable 4).

SUBSTITUTE SHEET (RULE 26) WO 96116179 P(,~ i51~716 Table 1 Cytot~Yir;ty (SRB ~ssay) ICso 5 MethoL.c~ate 10 nM N.D.
1 ~M, 3h 5798W93 100 nM N.D.
1 ~M, 3h S-Fluorouracil 1.9 ~M 1.4 ~lM
10 1 M, Sh 1614W94 40 ~M 1.7 ~M
1 ~LM, Sh Table 2 CytotoYirity (~H]-Deoxyuridine Assay) lS % Inhibition ~L2 AS49-BL
Methotrexate 82 i 4 88 + 6 1 ~M, 3h 5798W93 3 + 2 38 + 3 20 1 ~M, 3h S-Fluorouracil 95 + 8 91 + 10 1 ~LM, 5h 1614W94 -2 + 3 33 + 6 1 ~M, Sh SUBSTITUTE SHEET (RULE 26) CA 0220~93 1997-0~-16 WO 96/16179 pcrlGB9sM27l6 Table 3 A.. ~;1.. :.. - Effects of Se.. ~to~ T a~ ce and Cytosine De~ Genes in Mice I~e&l~g Subc~Pn~o~c A549 ~nm~n Lung A~lenoc~rcinoma Therapy Group Tunour Volume (mm3) re.c~nlage Inhibition (Relative to Control) phosph~tto Burr~;d 1268 + 212 0 Saline (No DNA) CMV-BLand 1614W94 511 ~ 86 60 CMV-CD and 380 + 237 70 S-Fluorocytosine S-Fluorollldcil alone 1021 + 37 19 SUBSTITUTE SHEET (RULE 26~

WO 96/16179 1 ~ 5lo27l6 Table 4 A~ nmlr Effects of Se~.~tol,~ ,B~ t~m~ce and Cytosine Des~ e Genes in Mice Intrathoracic H460 ~nm~n Large Cell Lung Tumours s Therapy Group Mean Days of p Value I~ ased Life SuF~i~al Span (%) P h o s p h a t e 20 - -Buffered Saline (No DNA) CMV-CD ~ PBS 23 0.245 15 CMV-CD + 5- 27 0.009 40 Fluolucylosine CMV-BL + 32 0.012 60 F~mrl~ 4 (i) General Al ~. o~.h Acylation of an essenti~l amino group of a wide variety of Cy~OtO~LiC drugs leads to a ~ignific~nt loss of toxicity. Penirillin G amidase and the related penicillin V ~miti~e activate 20 these CyloLo~ic drugs by removing a phenylacetyl group from these nontoxic drugs (Senter, The FASEB Journal, 4, 18~-193 (1990); Kerr et al., Cancer Tmmnnol. Tmmllnother., 31, SU8STITUTE SHEET (RULE 26) CA 0220~93 l997-0~-l6 wo 96/16179 PCr/Gss5/027l6 202-206 (1990); Bignami, et al., Cancer Research, 52, 5759-5764 (1992)). The goal is to pursue a gene directed enzyme prodrug therapy (GDEPT) approacll in order to take advantage of the yottlltial for long-term high levels of protein ~cumul~ti- n of enzyme at the tumor site using this approach. In order for the GDEPT approach to be succeccful using the 5 peni~-illin G ~mirl~e gene, the gene must be t;A~l~ssed in human cells. In E. coli, ~A~le~ion of the penit~illin G ~mid~ce gene is complex. The enzyme consists of two subunits with molecular weights of 24,000 daltons (a) and 65,000 daltons (b), produced by proteolytic cleavage of a 94,000 dalton precursor (Bock, et al, FEMS Microbil. Lett., 20, 141-144 (1983)) to remove an amino-terminal signal peptide (Oliver, e~ al. Gene 40, 9-14 (1985)).
10 and a 54 amino acid endopeptide linking the two subunits (Schumacher et al, Nucl. Acids Res. 14, 5713-5727 (1986)). In E. coli, p~ocec~ g of the enzyme occurs during translocation of the enzyme to the peripl~cm. We show that the unmodified coding region l~ep~sç~-ti--g this enzyme can be e~ rc:ssed, albeit at low levels, in human cells.

15 (ii) Materials and Methods Preparation of genomic DNA from E. coli ATCC 11105 -- R~.t~ri~l genomic DNA was prepared using a mo-lific~tion of Current Protocols in Molecular Biology (Section 2.4.1).
50 ml of a confluent culture of ATCC 11105 (American Type Culture Collection, Rockville, MD, USA) grown in LB broth was pelleted and the supern~t~nt removed. The pellet was 20 resuspended in 2.2 ml of 10 mM Tris-Cl, pH 7.5, EDTA 1 mM, 0.5% SDS, and 0.1 mg/ml proteinase K and incub~ted for 1 hour at 37 C with occasional mixing. An equal volume of buffer-saturated phenol was added (Life Technologies. Inc., Gaithersburg, MD, USA), gently mixed by inversion, and subsequently centrifuged. The aqueous phase was removed using a large bore pipette and treated with phenol two more times as above. Ammonium acetate 25 was added to a final collcPntration of 2M followed by the addition of 2 volumes of 100%
ethanol at ambient telnpel~tu-e. Using a sealed Pasteur pipette, DNA was swirled around pipette and then transferred to new tube. The sample was centrifuged to remove excess SUBSTITUTE SHEET (RULE 26) WO 96tl6179 PCTIGB95102716 liquid. The pellet was washed twice with 70% ethanol. After the second wash, DNA was res~cpen~ed in 300 ~l of 10 mM Tris-Cl, pH 7.5, 1 mM EDTA. The DNA was sheared by passage through a 27-gauge needle followed by digestion of the DNA with DNAse-free RNAse from bovine pancreas (Ro~ er M~nnhrim) at S miclu~ l~,l,s/ml for 1 hour at 37C.
5 The DNA was then purified using the Promega DNA Clean Up Kit (Promega Corp.). The DNA was ethanol precipitated and resllcpen~çd in 10 mM Tris-Cl, pH 7.5, 1 mM EDTA
and stored at 4C.

Isolation of the E. coli penicillin G arnidase Coding Region -- The sequence of the E. coli 10 penicillin G ~mirl~cç gene from ATCC 11105 has been reported (Oh, et al, Gene, 56, 87-97 (1987), Sch~lm~rhPr et al, Nucl. Acids Res. 14, 5713-5727 (1986)). Oligonucleotide primers s~nting fl~nking regions of E. coli pt~nicillin G ~mi~ce were synthPci7ed (OligoTherapeutics Inc., Wilsonville, OR, USA) to produce a PCR product l~ies~nting the E. coli p~nicillin G ~mitl~ce sequence from nucleotides 199 (numbering from published sequence) 15 to 2799 as well as ~tt~rhed fl~nking Hind III and Xba I restrictinn sites. An appruxi"~trly

2.7 Kb PCR product was purified from an agarose gel using a Glass-Max kit (Life Technologies, Gaithersburg, MD, USA) and ~ligçsted with Hind III and Xba I. The ~1igl~cted product was ligated to Hind III and Xba I digested pRc-CMV vector (InVitrogen, Inc.) and used to transform electroco-l-pe~ellt DH5cY cells (Molecular Cloning, Sambrook et al, Cold 20 Spring Harbor Press (1989)). Sequencing of the pCMV-pgal insert was l ~lrol--,ed using the fmol DNA Sequencing System (Promega Corp., Madison, WI, USA) ( Mead et al, Promega Notes, 16, (1988)). Two earlier reports on the sequence of E. coli penirillin G ~mi~l~ce differed at six positions (Oh et al (1987) and Schl-m~rht~-r et al (1986) supra). At all six positions, our sequence agreed completely with Sch~lm~her et al., (1986) supra. Also, our 25 pPnirillin G ~mit1Acç sequence did not contain the 4 ~itio~ nucleotides reported by Valle et al. Gene, 50, 115-122 (1986)) in the ribosome binding site.

SUBSTITUTE SHEET (RULE 26) -WO 96116179 PCI~ 5_J~,2716 Rec~llce co~ctitutive e:A~le~:,ion of penicillin G ~mi~ce produced by pCMV-pgal was toxic during growth of the bactrri~l host carrying the modified penicillin G ~mi~ce gene, pCMV-pgal was further modified to give rise to pCMV-pga2 by removing all ba~tçri~l promoter sequel~ces . One ~ ition~l motlifir~tion in this clone was the addition of a Kozak concpn~s 5 (ACCGCC) sequence im...~Ai~tely S' to the ATG start codon for possible enh~nced tr~ncl~tion~l effi~ienry in eukaryotic cells.

Penicillin G Amidase Assays -- The catalytic activity of prnicillin G ~mill~ce was routinely de~,ll,ined from the increase in A405 during hydrolysis of 1 ml 0.4 mM 6-nitro-10 phenyl~ret~mi-lo-benzoic acid in 50 mM phosphate pH 7.5 (K-lt7b~ch et al, Hoppe-Seyler's Z. Physiol. Chem., 354, 45-53 (1974)).

In vitro Expression -- Human cell lines A549 (lung ~çnoc~rcinoma) and NCI-NCI-H441 (lung Clara-like) were obtained from ATCC (Rockville, MD, USA). Cells were tr~nci~ntly 15 transfected via eleclr~,pol~lion (manufacturer's instructions, Bio-Rad, Hercules, CA).
Transfections using 6 mg of pl~cmid DNA were carried out in RPMI 16-40 media +
glllt~mine (MediaTech, Washington, DC, USA) and 10 mM glucose and 0.1 mM Dl-r at0.2 kilovolts and 960 microfarads in ele~;l.opol~tion cuvettes co.,tilhling a 0.4 cm gap (BioRad, Hercules, CA, USA). 24 hours after transfections, 4199W93 in 100% DMSO was 20 added to a final cQnrpntr~tion of 10 f.M (DMSO concentration was 2% final). Viable cells were ~ ntil~ted after 48 hours by counting the cells in the presence of trypan blue dye.

RNA Analysis -- A549 human lung ~ noc~rcinoma cells were cotransfected as descrihed above with pCMV-pgal and pCMV-BL, a construct identir~l to pCMV-pgal with the 25 exception of re~l~ring the prnicillin G ~mi~l~ce coding region with the coding region of TEM
~B-l~rt~m~ce RNA was prepared as described in Current Protocols in Molec~ r Biology (Section 4.2.4). 10 ~g of purified RNA was slot blotted onto Hybond-N filters (~m~rch~m, SUBSTITUTE SHEET (RULE 26) ~,lin~Lo,~ ~eightc~ IL), cross-linked by UV irr~ tion~ and probed with r32P]-labeled DNA
lc~,esæn~ either plasmid pCMV-pgal or pCMV-BL. Plasmid DNAs were labeled by a random random ~ llel~ DNA labeling system (Life Technologies, Gaithersburg, MD) following m~mlf~rtl-rer's instructions. The relative amounts of RNA determined by S hybri~i7~tion of the t32P]-labeled probe was qll~ntit~t~d using a phoshQrim~ger (Molecular Dynamics) using ImageQuant sorl~vare.
(Abbreviations: E. coli, Escherichia coli; HSV, Herpes Simplex Virus; DTT, dithiothreitol;
PCR, polymerase chain reaction.) 10 (iii) Ren-ltc Prodrugs; Cytotoxicity and Activation - Phenylacetylation of an ç~rnti~l amino group of a wide variety of ;yLotu~cic drugs leads to a ~ignifir~nt loss of cytotoxicity (Senter et al (1990) supra). We synthP~i7pd N-(2-phenylacetyl) daunomycin (4199W93) and N-(2-phenylacetyl) methotrexate and determined the cytotoxicity of these prodrugs in WiDr colon 15 carcinoma cells relative to their ~ ecLive parent drugs. The results are shown in Table 5.
As eYpe~tP~ phenylacetylation of daunomycin reduced the ~;yloto~icity of daunomycin and resulted in an increase of several orders of m~gnitllde in its IC50 value. In contrast, phenylacetylation of methotrexate increased its IC50 value by only S-fold (Table 5). In order to investigate the reasons for these differences we followed the g~n~r~tic~ll of the parent drugs 20 during these experimPnt~ by reverse phase liquid chromatography (RPLC). We observed that N-(2-phenylacetyl) daunomycin was very stable throughout the duration of the experiment as no daunomycin was detect~d in these cultures (data not shown). In contrast, a ~ignific~nt arnount of methoL,~a~e was genP~tt~d in tissue cultures inc-lb~trd with N-(2-phenylacetyl) methoL~ ate (data not shown). The half life for N-(2-phenylacetyl) methotrexate in these 25 e~p~ ;..,Pnt~ was about 50 hr.

PPnirillin G ami~e (PGA) can remove a phenylacetyl group from many different SUBSTITUTE SHEET (RULE 26) col.lp.,unds (Senter et al (1990) supra). A purified preparation of PGA efficiently converted N-(2-phenylacetyl) daunomycin to daunomycin as followed by RPLC. However, we were unable to determine the det~ilPA kin~tics of activation due to the insolubility of the prodrug.
We determined the kinPti~s of activation of N-(2-phenylacetyl) methotrexate by purified PGA
5 using RPLC. The activation reaction was linear as a function of time and enzyme concentration when the substrate conc~ntration was at saturation, and the enzyme followed normal ~i~h~t~lic-Menton kinptics (data not shown). The values of kCI~ and Km for PGA with N-(2-phenylacetyl) methotrexate as a substrate were determined and were 1.0 sec~l and 0.12 mM, respectively. This co~lespollded to a substrate specificity constant (kC,,,/K~,,) of 8.3 10 sec-lmM-E

Construction of Eucaryotic Expression Vectors - Two eukaryotic ~A~l~s~ion vectors, con~ il-g the E. coli pPnicillin G ~mit~ce coding region, dçcign~t~d pCMV-pgal and pCMV-pga2 were constructed. Purific~tioll of pCMV-pgal was problematic because the 15 pl~cmirl caused lysis of its b~cteri~l host. When E. coli DNS transformed with pCMV-pgal was grown at 37~C, the recombinant strain displayed the same growth kinetics as the parental DH5a strain until reaching an OD600 of 0.4, after which there followed rapid and a~palently complete lysis (Figure 5). R~ctPri~l lysis, as judged by the abnormal colony morphology, was also obvious when DHSa cont~ining pCMV-pgal was grown on the surface 20 of agar plates. The b~ tPri~l lysis induced by pCMV-pgal was due to production of active pPnic~illin G ~mid~cP (data not shown).

To f~rilit~tP plasmid production of pCMV-pga clones, the interfering ~A~l~;ssion of the penicillin G ~mi~ce gene in E. coli was blocked. To do this, pCMV-pgal was modified to 25 give rise to pCMV-pga2 by removing all b~rtPri~l promoter sequences (Figure 6). One additional mo~ifi~tion in this clone was the addition of a Kozak consensus sequence imm~Ai~t~]y S~ to the ATG start codon for possible enh~nced translational efficiency in SUBSTITUTE SHEET (RULE 26) CA 0220~93 1997-0~-16 eukaryotic cells. Colonies harboring this plasmid had normal growth ch~r~rtrri~tirs and produced large qll~ntiti~s of plasmid in plasmid pnrific~tio~lc. When tr~n~i~ntly transfected into A549 lung ~Pnoc~-cinoma and NCI-H441 lung Clara-like cell lines, pCMV-pga2 conferred the same degree of sPn~iti7~tion to these cell lines to N-acetyl phenyl daunomycin S as pCMV-pga2 (data not shown).

E~cpression of PGA in Hllman Lung Tumor Cell Lines - We employed a cell-based assay to cle~ e whether ~ n~;F-~t transfection of cultured cells with pCMV-pgal results in increased sensitivity of the cells to a relatively nontoxic and stable prodrug N-acetyl phenyl 10 daunomycin (Table 6). Two human lung cancer cell lines, NCI-H441 and A549 were transiently transfected with pCMV-pgal. Twenty-four hours after the transfection, the prodrug was added at a final concentration of lO ~M. The results are shown in Table 6.
Addition of prodrug alone had little effect on the cell morphology or on total number of viable A549 or H441 cells. In addition, transfection with control constructs in combination 15 with the prodrug did not result in any signifir~nt reductionc in viable cell number.
Tr~n~fe~tiQn of both NCI-H441 and A549 with pCMV-pgal plus prodrug resulted in arlr~m~tiC reduction in the number of viable cells after 48 hours and altered morphology and caused .cignifir,~nt accnm~ tiQn of cell debris. In contrast, transfection with control constructs in combination with the prodrug did not result in any ~ignific~nt reductions in 20 viable cell number.

By-Stan~er Tumor Cell ~ytotoxiciry an~ Low Levels of the Enyme - When id~ntir~l t~n~fection con~litions were ~ rd in A549 using a ,~-galactosidase gene reporter, we found that only approximately 0. l % of the total cells were transfected. Thus, the observed 25 decrease in cell number in cells treated with the prodrug and pCMV-pgal must be due to transfer of a killing effect of the transfected cells to the non-transfected cells, presumably through diffusion of daunomycin catalytically produced by extracellular enzyme.

SUBSTITUTE SH EET (RULE 26) CA 0220~93 1997-0~-16 WO 96/16179 PCTIGB95/n2716 However, the following observations suggest that the amount of extr~-e~ r enzyme activity - produced is relatively low. The con~Pntration of the prodrug used in above sensitivity expPrimPntc was about 200-fold more than a concentration of daunomycin which reduces the S viability of WiDr by 100% . Thus, an equivalent dose of parent drug was not generated in the experiments described above, in(iic~ting that either activity of the produced enzyme or the levels of the enzyme are relatively low.

The relatively low levels of enzyme produced could be due to several factors inclll~ing, 1) 10 instability of mRNA in human cells, 2) instability of the protein, 3) inPfficiency in post-translational processing and secretion of the precursor. To investigate the factors involved in the low levels of penicllin G ~mi-l~ce produced in transfections, we PY~minPA the first step in ~A~lession and qu~ntit~tPd the levels of penicillin G ~mi~ce mRNA steady-state levels relative to another gene produced from the same ~:Ayles~ion vector utili7in~ the same 15 ~A~l~e;,~;on regulatory signals, such as the CMV imme~i~tP/early promoter. In these PYpPrimPntc~ A549 cells were tr~nciently cotransfected with pCMV-pgal and pCMV-BL, an piession vector which ~,~plesses the TEM ,~ ct~m~ce gene. Total RNA was preparedfrom the cells, equal amounts were blotted onto a solid support, and each probed with either a penicillin G amidase or a TEM ,~-lactamase [32P]-labeled probed, respectively. We found 20 that the steady-state levels of penicillin G ~mid~ce mRNA were cignific~ntly lower than the rt~m~ce levels (data not shown). Thus, the low levels of ~Apies~ion pPnicillin G ~mi~l~ce may be in part accounted for by low steady-state levels of mRNA.

(iv) Diccl-scion 25 We have shown that the pCMV-pga constructs give rise to active penicillin G ~mi~l~ce when delivered to tumor cells. Since the unprocessed pl~cu~or protein of penicillin G ~mi-l~ce has been shown to be inactive (Burtscher et al, Brit. J. Biochem. 205, 77-83 (1992)), SUBSTITUTE SHEET (RULE 26) WO 96tl6179 P~ s95/Ot716 tAl.lcs~ion of functional enzyme from the gene lc~l~se~ting the ~le. ul~or in human cells implies that human cells are capable of removing part or all of the endopeptide sequence from the Fenirillin G ~mi~ce l~l~ul~or protein. In E. coli, the penicillin G ~mi~ e lJlC~cul~Ol is ~roces~ed as it is being tr~n~lor~t~d across the p~orirl~mic membrane. It is 5 possible that human cells have a homologous enzyme, or that the endopeptide isproteolytically sensitive and is removed by general protease activity present in the cell.

We have also provided evidence that the enzyme produced by the pCMV-pga vectors is secreted from human cells. This is consistent with the fact that the enzyme cont~inc a 10 b~rteri~l signal peptide, and that b~rteri~l signal peptides have been shown to be functional in eukaryotic cells (Muller et al, J. Biol. Chem. 25~, 11860-11863 (1982)). Extracp~ r prodrug activation by a secreted enzyme has the potential advantage of ma~cimi7ing bystander effects, i.e. transfer of cytotoxic effects from transfected cells to neighboring cells. This r~h~r~rt~ri~tic is critical given the limit~tioll~ of current gene transfer technology. Of course, 15 this effect would be influrnred by other factors such as diffusion of the enzyme away from the tumor site and stability of the enzyme in the extr~rplll~l~r space.

The fact that p-~nirillin G ~miti~e is capable of activating multiple prodrugs allows the evaluation of multiple drugs for efficacy. Since each drug and prodrug will display its own 20 ch~r~t~ri~tic pharmacokinetif~s and pharmacodynamics, each drug and drug combination must be explored under many conditions. Increased bystander effects could also conceivably be achieved through judicious choice of a prodrug(s).

SUBSTITUTE SHEET (RULE 26) PCT/GB9!5/02716 Table 5 ICso, ~M

S Daunomycin 0-07 N-(2-Phenylacetyl) Daunomycin > 100 MethuL~t;Aate 0.02 N-(2-Phenylacetyl) MethoLIt;Aale 0.11 10 Table 6 % Cell Viability Cell Line Addition A549 H441 pCMV-pgal 94 96 4199W93 + pCMV-pgal 30 35 SUBSTITUTE SH EET (RULE 26) ~Qu~w~ LISTING

~`Nr~'RAT~ lN~-~RMATION:

(i) APPLICANT:
(A) NAME: The Wellcome Foundation Limited (B) STREET: Unicorn Hou~e (C) CITY: 160, Euston Road (D) STATE: London (E) CO~h~: Great Britain (F) POSTAL CODE (ZIP): NWl 2BP

(ii) TITLE OF lN~wLlON: Enzyme Prodrug Therapy (iii) NUMBER OF ~QD~N~S: 6 (iv) COMPUTER R~An~Rr.~ FORM:
(A) MEDIUM TYPE: Floppy disk (B) COMPUTER: IBM PC compatible (C) OPERATING SYSTEM: PC-DOS/MS-DOS
(D) SOFTWARE: PatentIn Release ~l.O, Version ~1.30 (EPO) (2) INFOR~ATION FOR SEQ ID NO: l:

(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 1260 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: double (D) TOPOLOGY l~nk-( ii ) M~T~r~CrJr~r~ TYPE: cDNA

(ix) FEATURE:
(A) NAME/KEY: CDS
(B) LOCATION:l..1257 SlJBSTITUTE SHEET (RULE 26) CA 0220~93 l997-0~-l6 W O96/16179 PCT/GB95/n2716 (Xi) S~'QD~:N~: DESCRIPTION: SEQ ID NO: 1:

Met arg Gly Leu Leu Val Leu Ser Val Leu Leu Gly Ala Val Phe Gly LYB Glu Asp Phe Val Gly Hi~ Gln Val Leu Arg Ile Ser Val Ala ABP

Glu Ala Gln Val Gln LYB Val Ly~ Glu Leu Glu A~p Leu Glu His Leu Gln Leu Asp Phe Trp Arg Gly Pro Ala His Pro Gly Ser Pro Ile A~p Val Arg Val Pro Phe Pro Ser Ile Gln Ala Val Lys Ile Phe Leu Glu 65 70 ~5 80 Ser His Gly Ile Ser Tyr Glu Thr Met Ile Glu Asp Val Gln Ser Leu Leu Asp Glu Glu Gln Glu Gln Met Phe Ala Phe Arg Ser Arg Ala Arg Ser Thr A~p Thr Phe Asn Tyr Ala Thr Tyr His Thr Leu Glu Glu Ile Tyr ABP Phe Leu Asp Leu Leu Val Ala Glu Asn Pro His Leu Val Ser SUBSTITUTE SHEET (RULE 26) CA 0220~93 1997-0~-16 WO96/16179 PCTtGB95/n2716 Lys Ile Gln Ile Gly Asn Thr Tyr Glu Gly Arg Pro Ile Tyr Val Leu LYD Phe Ser Thr Gly Gly Ser Lys Arg Pro Ala Ile Trp Ile Asp Thr Gly Ile His Ser Arg Glu Trp Val Thr Gln Ala Ser Gly Val Trp Phe Ala Lys Lys Ile Thr Gln Asp Tyr Gly Gln Asp Ala Ala Phe Thr Ala Ile Leu Asp Thr Leu Asp Ile Phe Leu Glu Ile Val Thr Asn Pro A~p Gly Phe Ala Phe Thr His Ser Thr Asn Arg Met Trp Arg Lys Thr Arg Ser His Thr Ala Gly Ser Leu Cys Ile Gly Val Asp Pro Asn Arg Asn Trp A~p Ala Gly Phe Gly Leu Ser Gly Ala Ser Ser A~n Pro Cys Ser Glu Thr Tyr His Gly Lys Phe Ala Asn Ser Glu Val Glu Val Lys Ser Ile Val Asp Phe Val Lys Asp His Gly Asn Ile Lys Ala Phe Ile Ser SUBSTITUTE SHEET (RULE 26) CA 0220~93 1997-0~-16 WO96/16179 P~ 5slo27l6 Ile His Ser Tyr Ser Gln Leu Leu Met Tyr Pro Tyr Gly Tyr Lys Thr Glu Pro Val Pro Asp Gln Asp Glu Leu Asp Gln Leu Ser Lys Ala Ala Val Thr Ala Leu Ala Ser Leu Tyr Gly Thr Lys Phe A~n Tyr Gly Ser Ile Ile Lys Ala Ile Tyr Gln Ala Ser Gly Ser Thr Ile Asp Trp Thr Tyr Ser Gln Gly Ile Lys Tyr Ser Phe Thr Phe Glu Leu Arg Asp Thr Gly Arg Tyr Gly Phe Leu Leu Pro Ala Ser Gln Ile Ile Pro Thr Ala Lys Glu Thr Trp Leu Ala Leu Leu Thr Ile Met Glu His Thr Leu Asn His Pro Tyr (2) INFORMATION FOR SEQ ID NO: 2:

QU ~'N~ CHARACTERISTICS:

SUBSTITUTE SHEET (RULE 26) CA 0220~93 l997-0~-l6 (A) LENGTB: 419 amino acids ~B) TYPE: amino acid (D) TOPOLOGY: linear (ii) MnnT~'GUT~ TYPE: protein (xi) S~yu~.._~ D~TPTION: SEQ ID NO: 2:

Met Arg Gly Leu Leu Val Leu Ser Val Leu Leu Gly Ala Val Phe Gly ys Glu Asp Phe Val Gly Hi~ Gln Val Leu Arg Ile Ser Val Ala Asp Glu Ala Gln Val Gln Lys Val Ly~ Glu Leu Glu Asp Leu Glu His Leu Gln Leu Asp Phe Trp Arg Gly Pro Ala His Pro Gly Ser Pro Ile Asp Val Arg Val Pro Phe Pro Ser Ile Gln Ala Val Lys Ile Phe Leu Glu er His Gly Ile Ser Tyr Glu Thr Met Ile Glu Asp Val Gln Ser Leu eu Asp Glu Glu Gln Glu Gln Met Phe Ala Phe Arg Ser Arg Ala Arg Ser Thr Asp Thr Phe Asn Tyr Ala Thr Tyr His Thr Leu Glu Glu Ile Tyr Asp Phe Leu Asp Leu Leu Val Ala Glu Asn Pro His Leu Val Ser Ly~ Ile Gln Ile Gly Asn Thr Tyr Glu Gly Arg Pro Ile Tyr Val Leu SUBSTITUTE SH EET (RULE 26) CA 0220~93 1997-0~-16 W O96/16179 PCTIGB9Sl02716 Lys Phe Ser Thr Gly Gly Ser Lys Arg Pro Ala Ile Trp Ile Asp Thr Gly Ile His Ser Arg Glu Trp Val Thr Gln Ala Ser Gly Val Trp Phe Ala Lys Lys Ile Thr Gln Asp Tyr Gly Gln Asp Ala Ala Phe Thr Ala Ile Leu Asp Thr Leu Asp Ile Phe Leu Glu Ile Val Thr Asn Pro Asp Gly Phe Ala Phe Thr His Ser Thr Asn Arg Met Trp Arg Lys Thr Arg Ser His Thr Ala Gly Ser Leu Cys Ile Gly Val Asp Pro Asn Arg Asn Trp Asp Ala Gly Phe Gly Leu Ser Gly Ala Ser Ser Asn Pro Cys Ser Glu Thr Tyr His Gly Lys Phe Ala Asn Ser Glu Val Glu Val Lys Ser Ile Val Asp Phe Val Lys Asp His Gly Asn Ile Lys Ala Phe Ile Ser Ile His Ser Tyr Ser Gln Leu Leu Met Tyr Pro Tyr Gly Tyr Lys Thr Glu Pro Val Pro Asp Gln Asp Glu Leu Asp Gln Leu Ser Lys Ala Ala Val Thr Ala Leu Ala Ser Leu Tyr Gly Thr Lys Phe Asn Tyr Gly Ser Ile Ile Lys Ala Ile Tyr Gln Ala Ser Gly Ser Thr Ile Asp Trp Thr SUBSTITUTE SHEET (RULE 26) Tyr Ser Gln Gly Ile Lys Tyr Ser Phe Thr Phe Glu Leu Arg A~p Thr Gly Arg Tyr Gly Phe Leu Leu Pro Ala Ser Gln Ile Ile Pro Thr Ala Lys Glu Thr Trp Leu Ala Leu Leu Thr Ile Met Glu His Thr Leu Asn His Pro Tyr (2) lN~OkMATION FOR SEQ ID NO: 3:

(L) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 42 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: I~nkn. . "

( ii ) MnT T`CUT T'` TYPE: cDNA

(Xi) ~QU~N~ DESCRIPTION: SEQ ID NO: 3:

TTGCATAAGC TTGcr~cr~T GAGTATTCAA CA... CCG.G TC 42 (2) INFORMATION FOR SEQ ID NO: 4:

(i) ~yu~N~ CHU~RACTERISTICS:
(A) LENGTH: 36 ba~e pairs (B) TYPE: nucleic acid (C) STR~NnT~'nNESS: 3 ingle (D) TOPOLOGY: t~nknC~l, ( ii ) MnT T'`CUT T~ TYPE: cDNA

SUBSTITUTE SHEET (RULE 26) CA 0220~93 1997-0~-16 (xi) S~YU~N~ DESCRIPTION: SEQ ID NO: 4:

GA.~ A GATTAC~T GCTTAATCAG TGAGGC 36 (2) INFORMATION FOR SEQ ID NO: 5:

(i) ~yu~N~ CHARACTERISTICS:
(A) LENGTH: 6229 ba~e pair~
(B) TYPE: nucleic acid (C) sT~ANn~nNEss: double (D) TOPOLOGY: I~nknrS/,, (ii) MOLECULE TYPE: cDNA

(ix) FEATURE:
(A) NAMEtXEY: CDS
(B) LOCATION:693. 1550 (xi) ~y~;N~ DESCRIPTION: SEQ ID NO: 5:

CGATGTACGG GccAr-ATATA CGCGTTGACA TTGATTATTG ACTAGTTATT AATAGTAATC 60 AAATGGCCCG CCTGGCTGAC CGCC~AACGA CCCCCGCCCA TTGACGTCAA TAATGACGTA 180 ~ CC~ATA GTAACGCCAA T~GG~-ACTTT CCATTGACGT CAATGGGTGG ACTATTTACG 240 CGTCAATGAC GGTAAATGGC CCGC~,GGCA TTATGCCCAG TACATGACCT TATGGGACTT 360 TCCTACTTGG CAGTACATCT ACGTATTAGT CATCGCTATT ACCATGGTGA TGCGG ~,G 420 ; 49 SUBSTITUTE SH EET (RULE 26) CA 0220~93 1997-0~-16 GCAGTACATC AATGGGCGTG GATAGCGGTT TGACTCACGG GGATTTCCAA G,~.C~ACCC 480 CATTGACGTC AATGGGAGTT .~..,.GGCA Cr~AA~TCAA CGGGACTTTC CAAAATGTCG 540 ~AArAAcTcc GCCCCATTGA CGCAAATGGG CGGTAGGCGT GTACGGTGGG AGGTCTATAT 600 AArr~GAr-cT C,~-~G~.AA cT~r-Ar-AAcc CACTGCTTAA CTGGCTTATC GAAATTAATA 660 CGACTCACTA TAGGr-Ar-ArC AAGCTTGCCA CC ATG AGT ATT CAA CAT TTC CGT 713 Met Ser Ile Gln His Phe Arg Val Ala Leu Ile Pro Phe Phe Ala Ala Phe Cys Leu Pro Val Phe Ala , 430 435 440 His Pro Glu Thr Leu Val Lys Val Lyq Asp Ala Glu Asp Gln Leu Gly Ala Arg Val Gly Tyr Ile Glu Leu Asp Leu Aqn Ser Gly Lys Ile Leu Glu Ser Phe Arg Pro Glu Glu Arg Phe Pro Met Met Ser Thr Phe Lyq Val Leu Leu Cys Gly Ala Val Leu Ser Arg Ile Asp Ala Gly Gln Glu Gln Leu Gly Arg Arg Ile His Tyr Ser Gln Asn Asp Leu Val Glu Tyr SUBSTITUTE SHEET tRULE 26) CA 0220~93 l997-0~-l6 W O96/16179 PCTI~SJ'~2716 Ser Pro Val Thr Glu Lya His Leu Thr Asp Gly Met Thr Val Arg Glu Leu Cy~ Ser Ala Ala Ile Thr Met Ser Asp Asn Thr Ala Ala Asn Leu Leu Leu Thr Thr Ile Gly Gly Pro Lys Glu Leu Thr Ala Phe Leu His Asn Met Gly ABP His Val Thr Arg Leu Asp Arg Trp Glu Pro Glu Leu Asn Glu Ala Ile Pro Asn Asp Glu Arg Asp Thr Thr Met Pro Val Ala Met Ala Thr Thr Leu Arg Lys Leu Leu Thr Gly Glu Leu Leu Thr Leu Ala Ser Arg Gln Gln Leu Ile Asp Trp Met Glu Ala Asp Lys Val Ala Gly Pro Leu Leu Arg Ser Ala Leu Pro Ala Gly Trp Phe Ile Ala Asp A~A TCT GGA GCC GGT GAG CGT GGG TCT CGC GGT ATC ATT GCA GCA CTG 1433 Lys Ser Gly Ala Gly Glu Arg Gly Ser Arg Gly Ile Ile Ala Ala Leu .

SUBSTITUTE SH EET (RULE 26) ~ = = - ~
CA 0220~93 1997-0~-16 GGG CCA GAT GGT A~G CCC TCC CGT ATC GTA GTT ATC TAC ACG ACG GGG 1481 Gly Pro Asp Gly Lys Pro Ser Arg Ile Val Val Ile Tyr Thr Thr Gly Ser Gln Ala Thr ~et A~p Glu Arg Asn Arg Gln Ile Ala Glu Ile Gly GCC TCA CTG ATT AAG CAT TGG TAATCTAGAG GGCC~LATTC TATAGTGTCA 1580 Ala Ser Leu Ile Lys His Trp CCTA~TGCT AGAGCTCGCT GATCAGCCTC GACTGTGCCT TCTAGTTGCC AGCCATCTGT 1640 TGTTTGCCCC ~CCCCC~,GC ~..C~..GAC CCTGGAAGGT GCCACTCCCA ~.~.C~..~C 1700 CTA~T~ T GAGGA~ATTG CATCGCATTG TCTGAGTAGG TGTCATTCTA TTCTGGGGGG 1760 ~GGG~lGGGG CAGGACAGCA AGGGGGAGGA TTGGGAAGAC AATAGCAGGC A~G~-GGGGA 1820 .GC~.GGGC TCTATGGAAC CAGCTGGGGC TCGAGGGGGG ATCCCC~rGC GCC~-~-AGC 1880 GGCGCATTAA GCGCGGCGGG i~G~-G~-- ACGCGCAGCG TGACCGCTAC ACTTGCCAGC 1940 GCCCTAGCGC CCG~C~,~ CG~...-~C CC-,C~,,.C TCGCCACGTT CGCCGGCTTT 2000 CCCC~.~AAG CTCTAAATCG GGGCATCCCT TTAGGG~CC GATTTAGTGC TTTACGGCAC 2060 CTCr-~rCCr~ AA~AACTTGA TTAGGGTGAT GGTTCACGTA GTGGGCCATC GCCCTGATAG 2120 ACG~,,,.C GCCTTTACTG AGCACTCTTT AATAGTGGAC ,~..~,, '~A AACTGGAACA 2180 ACACTCAACC CTA~ GG, CTA,. GATTTATAAG ATTTCCATCG CCATGTAAAA 2240 GTGTTACAAT TAGCATTAAA TTA~,,~,~, ATATGCTACT A'-~-L--GG ~1-CC. ~AC 2300 GGGG~GG~A CCGAGCTCGA A.,~,~,GGA A~ ~A GTTAGGGTGT GGA~AGTCCC 2360 SUBSTITUTE SHEET (RULE 26~

-CA 0220~93 1997-0~-16 W 096116179 P~ 5l~27l6 TGTGGAAAGT CCCr~GGCTC CCCAGCAGGC AGAAGTATGC AAAGCATGCA TCTCAATTAG 2480 GCCCATTCTC CGCCC~ATGG CTGACTAATT ~..... ATTT ATGCAGAGGC CGAGGCCGCC 2600 TCGGCCTCTG AGCTATTCCA GAAGTAGTGA GGAGGCTTTT TTGGAGGCCT AGG~..~.GC 2660 AAAAAGCTCC CGGGAGCTTG GATATCCATT TTCGGATCTG ATrAAG~r~C AGGATGAGGA 2720 ~`~...CGCA TGATTGAACA AGATGGATTG CACGCAGGTT CTCCGGCCGC TTGGGTGGAG 2780 AGGCTATTCG GCTATGACTG GGrAr~Ar~G ACAATCGGCT GCTCTGATGC CGCC~-.C 2840 CGGCTGTCAG CGCAGGGGCG CCCG~.~.. TTTGTCAAGA CCGACCTGTC CGGTGCCCTG 2900 AATGAACTGC AGr,~Cr,PGGC AGCGCGGCTA TCGTGGCTGG CCACGACGGG C~l-C~--GC 2960 CCGGGGCAGG A.`-.C~.~.C ATCTCACCTT GCTCCTGCCG AGAAAGTATC CATCATGGCT 3080 GATGCAATGC GGCGGCTGCA TACG~.~GAT CCGGCTACCT GCCCATTCGA Cr~CrAAr-CG 3140 AAACATCGCA TCGAGCGAGC ACGTACTCGG ATGGAAGCCG C~l..~.CGA TCAGGATGAT 3200 CTGr-~rr-~G AGCATCAGGG GCTCGCGCCA GCCGAACTGT TCGCCAGGCT CAAGGCGCGC 3260 ATGCCCr-~rG GCGAGGATCT C~.CG-GACC CATGGCGATG CCTGCTTGCC GAATATCATG 3320 GTGr-~APTG GCCGCTTTTC TGGATTCATC GACTGTGGCC GG~.GGG.~. GGCGGACCGC 3380 TATr-AGr-Ar~ TAGCGTTGGC TACCCGTGAT ATTGCTGAAG AGCTTGGCGG CGAATGGGCT 3440 GACCG~..CC TCGTGCTTTA CGGTATCGCC GCTCCCGATT CGCAGCGCAT CGC~.~.AT 3500 SUBSTITUTE SHEET (RULE 26) CA 0220~93 1997-0~-16 CGC~.~ G ACGAGTTCTT CTGAGCGGGA CTCTGGGGTT CGA~ATGACC GACCAAGCGA 3560 CGCCrAArCT GCCATCACGA GATTTCGATT CCACCGCCGC CTTCTATGAA AG~GGG~I 3620 TCGGAATCGT ~CC~GGAC GCCGGCTGGA TGA~C~.C~A GCGCGGGGAT CTCATGCTGG 3680 A~.~ CGC crArCCrAAr ~ ATTG CAGCTTATAA TGGTTACAAA TAAAGCAATA 3740 GCATCACAAA TTTcArAAAT AAAGCATTTT TTTCACTGCA TTCTAGTTGT G~ C~A 3800 AACTCATCAA TGTATCTTAT CA~ aGA .CCC~CGAC CTCGAGAGCT TGGCGTAATC 3860 ATGGTCATAG C~ C~.G TGTGAAATTG TTATCCGCTC ACAATTCCAC ArAAr~TACG 3920 AGCCGGAAGC ATA~AGTGTA AAGCCTGGGG TGCCTAATGA GTGAGCTAAC TCACATTAAT 3980 ~GC~.~GCGC TCACTGCCCG C...CCAGTC GGGAAACCTG TCGTGCCAGC TGCATTAATG 4040 AATCGGCCAA CGCGCGGGGA GAGGCGGTTT GCGTATTGGG CG~I~.~CCG ~C~l`G~ 4100 CACTGACTCG CTGCGCTCGG ~C~--CGGCT GCGGCGAGCG GTATCAGCTC ACTCAAAGGC 4160 GGTAATArGG TTATCCACAG AATCAGGGGA TAACGCAGGA AAr~AArA~TGT GAGCAAAAGG 4220 CCAGCAAAAG GCCAGGAACC GTAAAAAGGC CGCGTTGCTG GC~ .CC ATAGGCTCCG 4280 CCCCC~.GAC GAGCATCACA AAAATCGACG CTCAAGTCAG AGGTGGCGAA ArCCrArAGG 4340 ACTATAAAG~ TACCAGGCGT ~CCCC~lGG AAG~-CC~C GTGCGCTCTC ~.~..CCGAC 4400 CCTGCCGCTT ACCGGATACC TGTCCGCCTT T~`C~l~CG GGAAGCGTGG CG~---~--A 4460 ATGCTCACGC TGTAGGTATC TCA~-~CG~- GTAGG~ CGCTCCAAGC TGGGCTGTGT 4520 GC~Cr-AACCC CCC~AGC CCGACCGCTG CGCCTTATCC GGTAACTATC GTCTTGAGTC 4580 CAArccG~A Ar~ArAcrAcT TATCGCCACT GGCAGCAGCC ACTGGTAACA GGATTAGCAG 4640 SUBSTITUTE SH EET (RULE 26) CA 0220~93 1997-0~-16 Ar7cr~AGGTAT GTAGGCG~LG CTACA~AGTT CTTGAAGTGG TGGCCTAACT ACGGCTACAC 4700 TArAAr-r-~rA GTA,,,~,A ~GCGC-~- GCTGAAGCCA GTTACCTTCG GAAAAAr-~r,T 4760 TGGTAGCTCT TGATCCGGCA AArAAArrAr CGCTGGTAGC GG~G6~ GCAA 4820 GCAGCAGATT ACGCGCAGAA AAAAAr-r-ATC TrAAr-AAr-AT CCTTTGATCT TTTCTACGGG 4880 GTCTGACGCT CAGTGGAACG AAAACTCACG TTAAGGGATT ..GG~ATGA GATTATCAAA 4940 ATATGAGTAA AC~ ~6~LG ACAGTTACCA ATGCTTAATC AGTGAGGCAC CTATCTCAGC 5060 GA~ A ~ CAT CCATAGTTGC CTGACTCCCC G~ AGA TAACTACGAT 5120 GGCTCCAGAT TTATCAGCAA TAAAC~Ar,CC AGCCGGAAGG GCC~-~GCGCA GAAGTGGTCC 5240 TTCGCCAGTT AATAGTTTGC GCAACGTTGT TGCCATTGCT ACAGGCATCG ~GGL~.~ACG 5360 ~,C~ , GGTATGGCTT CATTCAGCTC CG~CC~AA CGATCAAGGC GAGTTACATG 5420 A-CCCC~TG TTGTGCAAAA AAGCGGTTAG ~- ~LLCG61 CCTCCGATCG TTGTCAGAAG 5480 CATGCCATCC GTAAGATGCT ~ GAC TGGTGAGTAC TCAACCAAGT CATTCTGAGA 5600 ATAGTGTATG CGGCGACCr-~A ~G~iG CCCGGCGTCA ATACGGGATA ATACCGCGCC 5660 Ar~TAGrArA ACTTTAAAAG TGCTCATCAT TGr-AAAAcGT TCTTCGGGGC GAAAACTCTC 5720 SUBSTITUTE SHEET (RULE 26) CA 0220~93 1997-0~-16 TTCAGCATCT TTTACTTTCA CCAGCGTTTC TGGGTGAGCA AAAAr7~r-&AA GGCAA~ATGC 5840 CG~AAAAAAr, GGAATAAr-GG cr-ArArGGAA ATGTTGAATA CTCATACTCT ~C~L~A 5900 ATATTATTGA AGCATTTATC AGGGTTATTG TCTCATGAGC GGATA~ATAT TTGAATGTAT 5960 TTAr-AAAAAT AAAr~AATAr, GG~--CCGCG CACATTTCCC Cr-AAAAGTGC CACCTGACGT 6020 Cr-ArGr-ATCG GGAGATCTCC CGA.CCC~-A .~G.CGACTC TCAGTACAAT CTGCTCTGAT 6080 GccGrATAr-T TAAGCCAGTA TCTGCTCCCT GC~.~.~.~ TGGAGGTCGC TGAGTAGTGC 6140 GCGAGCAAAA TTTAAGCTAC AACAAGGCAA GGCTTGACCG ACAATTGCAT ~-AAr-AATcTG 6200 CTTAGGGTTA GGC~.. ~.GC GCTGCTTCG 6229 (2) INFORMATION FOR SEQ ID NO: 6:

u~N~ CHARACTERISTICS:
(A) LENGTH: 286 amino acids (B) TYPE: amino acid (D) TOPOLOGY: linear ( ii ) ~T~T~`cuT~T! TYPE: protein (xi) SEQUENCE DESCRIPTION: SEQ ID NO: 6:

Met Ser Ile Gln Hin Phe Arg Val Ala Leu Ile Pro Phe Phe Ala Ala Phe Cys Leu Pro Val Phe Ala His Pro Glu Thr Leu Val Lys Val Lys Asp Ala Glu Asp Gln Leu Gly Ala Arg Val Gly Tyr Ile Glu Leu Asp Leu Asn Ser Gly Lys Ile Leu Glu Ser Phe Arg Pro Glu Glu Arg Phe SUBSTITUTE SHEET (RULE 26) -- ~ =
CA 0220~93 l997-0~-l6 W 096/16179 P~~ SJ~27l6 Pro Met Met Ser Thr Phe LYB Val Leu Leu Cys Gly Ala Val Leu Ser Arg Ile Asp Ala Gly Gln Glu Gln Leu Gly Arg Arg Ile His Tyr Ser Gln Asn A~p Leu Val Glu Tyr Ser Pro Val Thr Glu Lys His Leu Thr A~p Gly Met Thr Val Arg Glu Leu Cys Ser Ala Ala Ile Thr Met Ser Asp Asn Thr Ala Ala Asn Leu Leu Leu Thr Thr Ile Gly Gly Pro LYG

Glu Leu Thr Ala Phe Leu His Asn Met Gly Asp His Val Thr Arg Leu A~p Arg Trp Glu Pro Glu Leu Asn Glu Ala Ile Pro Asn Asp Glu Arg Asp Thr Thr Met Pro Val Ala Met Ala Thr Thr Leu Arg Lys Leu Leu Thr Gly Glu Leu Leu Thr Leu Ala Ser Arg Gln Gln Leu Ile Asp Trp Met Glu Ala Asp Lys Val Ala Gly Pro Leu Leu Arg Ser Ala Leu Pro Ala Gly Trp Phe Ile Ala Asp Lys Ser Gly Ala Gly Glu Arg Gly Ser Arg Gly Ile Ile Ala Ala Leu Gly Pro Asp Gly Lys Pro Ser Arg Ile Val Val Ile Tyr Thr Thr Gly Ser Gln Ala Thr Met Asp Glu Arg Asn SUBSTITUTE SHEET (RULE 26) W 096116179 P~ g5J'~716 Arg Gln Ile Ala Glu Ile Gly Ala Ser Leu Ile Ly Hi~ Trp SUBSTITUTE SHEET (RULE 26)

Claims (22)

CLAIMS:

1. A molecular chimaera for use with a prodrug, the molecular chimaera comprising a transcriptional regulatory DNA sequence capable of being activated in a targetted mammalian cell and a DNA coding sequence operatively linked to the transcriptional regulatory DNA
sequence and encoding a secretion signal peptide and a heterologous enzyme, such that on expression of said coding sequence in the targetted cell, the heterologous enzyme is capable of passing through the cell plasma membrane and is capable of catalysing extracellular conversion of the prodrug into a cytotoxic or cytostatic agent.

2. A molecular chimaera according to Claim 1 wherein the heterologous enzyme is obtained from an enzyme precursor by cleavage at a proteolytic cleavage site by a pathology associated protease extracellular to the targetted mammalian cell.

3. A chimaera according to Claim 2 wherein the pathology associated protease is a tumour associated protease.

4. A chimaera according to Claim 2 or 3 wherein the proteolytic cleavage site has a sequence susceptible to cleavage by a tumour-associated matrix metalloprotease.

5. A chimaera according to any of Claims 1 to 4 wherein the secretion signal peptide is encoded by a DNA sequence engineered to that encoding the heterologous enzyme.

6. A chimaera according to any of Claims 1 to 5 wherein the heterologous enzyme is a mutant mammalian enzyme.

7. A chimaera according to any of Claims 1 to 5 wherein the heterologous enzyme is a non-mammalian enzyme.

8. A chimaera according to any of Claims 1 to 7 wherein the transcriptional regulatory DNA sequence is selectively activated in the targetted mammalian cell.

9. A chimaera according to Claim 8 wherein the transcriptional regulatory DNA
sequence is selected from the transcriptional regulatory DNA sequences of the genes for albumin, alphafetoprotein, carcinoembryonic antigen, tyrosine hydroxylase, choline acetyl transferase, neuron specific enolase, glial fibro acid protein, insulin, gammaglutamyltranspeptidase, dopa decarboxylase, HER2/neu, and N-myc.

10. A vector containing a chimaera as claimed in any of Claims 1 to 9.

11. A packaging cell line containing a vector as claimed in Claim 10.

12. An infective virion generated from a packaging cell line a claimed in Claim 11.

13. An infective virion according to Claim 12 which is a retrovirus, an adenovirus or an adeno-associated virus.

14. An infective virion for use with a prodrug, capable of selective infection of a targetted mammalian cell, said virion encapsidating a molecular chimaera comprising a transcriptional regulatory DNA sequence capable of being activated in a targetted mammalian cell and a DNA coding sequence operatively linked to the transcriptional regulatory DNA sequence and encoding a signal peptide and a heterologous enzyme, such that on expression of said coding sequence in the targetted cell, the heterologous enzyme is capable of passing through the cell plasma membrane and is capable of catalysing extracellular conversion of the prodrug into a cytotoxic or cytostatic agent.

15. An infective virion according to Claim 14 wherein the heterologous enzyme is obtained from an enzyme precursor by cleavage at a proteolytic cleavage site by a pathology associated protease extracellular to the targetted mammalian cell.

16. An infective virion according to Claim 14 or 15 wherein the signal peptide is encoded by a DNA sequence engineered to that encoding the heterologous enzyme.

17. An infective virion according to any of Claims 14 to 16 wherein the transcriptional regulatory DNA sequence is selected from the transcriptional regulatory DNA sequences of the genes for cytomegalovirus, glyceraldehyde-3-phosphate, .beta.-actin or tubulin, albumin, alphafetoprotein, carcinoembryonic antigen, tyrosine hydroxylase, choline acetyl transferase, neuron specific enolase, glial fibro acid protein, insulin, gammaglutamyltranspeptidase, dopa decarboxylase, HER2/neu, and N-myc.

18. An infective virion according to any of Claims 14 to 17 being a retrovirus, an adenovirus or an adeno-associated virus.

19. A packaging cell line capable of producing an infective virion as claimed in any of Claims 14 to 18.

20. Use of an infective virion according to any of Claims 14 to 18 for the manufacture of a medicament for use in therapy wherein the therapy comprises the selective infection of a targetted mammalian cell using retroviral infection, adenoviral infection or adeno-associated viral infection, physically localised delivery or engineered viral coat proteins.

21. A pharmaceutical formulation comprising a molecular chimaera as claimed in any of Claims 1 to 9, a vector as claimed in Claim 10, a packaging cell line as claim in either of claims 11 and 19 or an infective virion as claimed in any of Claims 12 to 18.

22. Use of a molecular chimaera according to any of Claims 1 to 9 for the manufacture of a medicament for use in therapy wherein the therapy comprises administering said molecular chimaera by a method selected from calcium phosphate transfection, electroporation, microinjection, liposomal transfer, ballistic barrage, retroviral infection and adeno or adeno-associated virus infection.