DNA Preparation Coding for Ricin a and Methods of Using Same

This invention relates to DNA comprising a nucleotide sequence coding for at least a portion of a polypeptide which is a plant toxin of the ricin type, as hereinafter defined. It also relates to recombinant DNA molecules comprising a DNA sequence which codes for a polypeptide which is or is closely related to a natural plant toxin of the ricin type. Ricin, and also other plant toxins such as abrin, modeccin and viscumin, consist of two polypeptide chains (known as the A and B chains) linked by a disulphide bridge, one chain (the A chain) being primarily responsible for the cytotoxic and the other chain (the B chain) having sites enabling the molecule to bind to cell surfaces. Ricin is produced in the plant Ricinis communis (also known as the castor bean plant) via a precursor protein known as "preproricin".

Preproricin comprises a single polypeptide chain which includes a leader sequence. The leader sequence is subsequently removed in the organism to give proricin which is then cleaved to eliminate a linker region and joined by a disulphide bond to form the mature protein. The toxicity of ricin-type toxins operates in three phases: (1) binding to the cell surface via the B chain; (2) penetration of at least the A chain into the cytosol, and (3) inhibition of protein synthesis through the A chain attacking the 60S subunits of the ribosomes. Thus, separated A and B chains are essentially non-toxic, the inherently toxic A chain lacking the ability to bind to cell surfaces in the absence of the B chain.

It is also known that in ricin-type toxins the B chain binds to cell surfaces by virtue of galactose recognition sites, which react with glycoproteins or glycolipids exposed at the cell surface.

It has already been suggested that the toxicity of the ricin A chain might be exploited in anti-tumour therapy, by replacing the indiscriminately-binding B chain with a different carrier component having the ability to bind only to tumour cells. Thus, various immunotoxins have already been prepared, consisting of a conjugate of whole ricin or a separated natural ricin A chain and a tumour-specific monoclonal antibody. Although these known conjugates are of considerable potential in themselves, there is scope for improvement.

One problem with the known conjugates arises from a structural feature of the A chain from natural ricin. It is known that the natural ricin A chain becomes N-glycosylated during its synthesis, by enzymes present in Ricinus cells, and it is thought that the resulting sugar moieties are capable of non-specific interactions with cell surfaces. Thus, it appears that the known A chain conjugates are capable of a certain amount of binding with non target cells, even in the absence of the natural B chain, thus increasing the toxicity of such immunotoxins towards non target cells.

Another problem with the known ricin A chain conjugates stems from the fact that the B chain seems to have an important secondary function in that it somehow assists in the intoxication process, apart from its primary function in binding the ricin molecule to the cell surfaces. This secondary function is lost if the B chain is replaced by a different carrier component such as a monoclonal antibody.

If it were possible to prevent interactions between the cell surface via the A chain sugar moieties, whilst preserving the secondary toxicity-increasing function of the B chain, the toxicity of a whole ricin antibody conjugate towards normal cells could be reduced, and towards target cells could be increased, thus improving the therapeutic index of the immunotoxin. It is also known that the natural ricin B chain is N-glycosylated and the B chain sugar moieties may also contribute to non specific interactions, Also, the sugar moieties in both chains enable the ricin molecule to be sequestrated by reticuloendothelial cells in the liver, and so would lead to the rapid excretion from the system of a drug based on a part or the whole of the ricin molecule in which such sugar moieties were still present.

Attempts to remove all the sugar moieties from natural ricin by chemical or enzymatic methods have so far failed. Nevertheless the major obstacle confronting the use of known whole ricin-antibody conjugates is the presence of two galactose binding sites in the ricin B chain. These B chain galactose binding sites are primarily responsible for the non-specific cellular interactions of current whole ricin-antibody conjugates, particularly when used in vivo. Their presence in the natural toxin clearly eliminates or reduces the targeting specifity conferred by the antibody.

An improved immunotoxin based on ricin or another plant toxin of the ricin type, not suffering from these problems, could consist of a whole toxin molecule modified so that it is not N-glycosylated, and so that the B chain has no galactose recognition sites, but retains its secondary intoxication-promoting properties, coupled to a carrier moiety which delivers the toxin to the target cells. This could be a tumour-specific or cell/tissue specific vehicle such as a suitable monoclonal antibody.

Our research which has so far been concentrated on ricin itself, has indicated that the assembly of ricin (and the related agglutinin which consists of two ricin-like molecules with slightly modified A and B chains) does not involve the separate synthesis of the A and B chains as the products of distinct mRNA's, but rather the initial formation of a single polypeptide precursor containing both the A chain and B chain sequences. This is thought to apply in the case of other toxins of the same type.

This invention is based on the idea of preparing a genetically-engineered microorganism capable of expressing a molecule of a toxin of the ricin type, as defined above, or alternatively part of such a toxin molecule, or a precursor of such a molecule (which could be converted to the toxin molecule itself) which toxin molecule could be modified as suggested above and could be used to construct an effective toxin conjugate by combining it with a tumour-specific or cell/tissue specific monoclonal antibody or other carrier moiety, such as a hormone or lectin.

The fact that ricin is formed via a precursor polypeptide will enable a cell system to be constructed by known techniques which expresses a ricin precursor. The ricin precursor product could then be chemically or enzymatically converted to the desired modified ricin. An analagous technique could be used in the case of other ricin-type toxins as herein defined. An alternative technique would be to divide from the DNA sequence that codes for the precursor two sequences which code separately for the A and B chains, to insert these separated sequences into different cloning vehicles and to insert the resulting recombinant DNA molecules into separate host microorganisms. One such host would then express the A chain polypeptide sequence and the other the B chain polypeptide sequence. These sequences could then be combined to form the desired modified ricin molecule. This technique could obviously also be used for any ricin-type toxins in which the A and B chains are encoded by distinct mRNA gene pools. This latter approach would be preferred on safety grounds, in that separate and therefore non-toxic A and B chains would be expressed.

The drawing summarizes the preparation of a transformed host containing a DNA sequence which codes for the ricin A and B chain precursor polypeptide .

According to one aspect of the invention we provide a biologically pure and homogeneous sample of DNA comprising a nucleotide sequence coding for at least a portion of a precursor of a ricin-type toxin polypeptide, or mutants thereof.

Said portion preferably comprises the A chain or the B chain of the mature protein.

More specifically, we provide a sample of DNA including at least a substantial portion of any of the following DNA sequences, which sample is biologically-pure: nucleotides -72 to 1623, or nucleotides -72 to 801, or nucleotides 838 to 1623, as defined hereinafter at Table 1 or at least a portion of a nucleotide sequence which is equivalent thereto by virtue of degeneracy of the genetic code.

According to another aspect of this invention we provide a recombinant DNA molecule comprising a DNA sequence which codes for a polypeptide sequence present in a plant toxin of the ricin type, as defined herein.

More specifically, we provide a recombinant DNA molecule containing a DNA sequence which codes for a A and B chain precursor polypeptide of a plant toxin of the ricin type.

Alternatively, we provide a recombinant DNA molecule containing a DNA sequence coding for at least a portion of either the A chain or the B chain of a plant toxin of the ricin type.

According to a further aspect of the present invention we provide a genetically modified host microorganism which contains a recombinant DNA molecule as hereinbefore defined.

In the recombinant DNA molecules according to the present invention, the nucleotide sequence coding for the B chain may be modified to eliminate or inactivate the galactose binding sites, and the precursor polypeptide and hence in the mature protein, to eliminate or inactivate the galactose binding sites, and the sequences encoding the signals for N-glycosylation may also be modified to render them ineffective or to eliminate them. Examples of techniques which may prove useful are deletion or oligonucleotide mediated mutagenesis.

The host organism may be a plant cell or an animal cell or preferably a microorganism.

The microorganism may be a prokaryote or a eukaryote. As examples of prokaryotes may be mentioned Gram-negative bacteria, e.g. E. coli, Methylophilus methylotrophus and Alcaligenes eutrophus; and Gram-positive bacteria, e.g. Streptomyces, Bacillus subtilis and Arthrobacter. As examples of eukaryotes may be mentioned yeasts, for examples Saccharomyces cerevisiae.

The recombinant DNA molecules may comprise a cloning vector such as a plasmid or phage vector into which has been inserted the DNA sequence coding for at least a portion of a precursor polypeptide, or at least a portion of either the A chain or the B chain, of a ricin-type plant toxin.

The cloning vector is preferably a plasmid although we do not exclude the possibility that it may be a phage vector. The plasmid may be a naturally-occurring plasmid or preferably a composite derived from fragments of other plasmids. Where a composite plasmid is employed it preferably contains promoter sequences which improve expression of the ricin gene.

Examples of suitable plasmids which may be used as cloning vehicles are inter alia for Gram-negative bacteria: pBR322, pAT153, pUC8, pGSS15 and pMB9; and for S. cerevisiae: pMA91, pMA230, YRp7, and YEp6. The vector will be selected to be suitable for the particular host envisaged.

We also provide a method of obtaining a recombinant DNA molecule which comprises preparing a double-stranded DNA sequence which codes for a polypeptide sequence present in a plant toxin of the ricin type and inserting said double-stranded DNA sequence into a cloning vector.

More specifically such a method may comprise isolating the mRNA which codes for the ricin A and B chain precursor polypeptide, synthesising single-stranded cDNA from said mRNA using reverse transcriptase and a suitable primer, assembling the second DNA strand onto the template formed by said first strand by means of DNA polymerase followed by S1 nuclease, and inserting the resulting double-stranded cDNA into a cloning vector.

Alternatively, the cDNA assembled from the mRNA may be cut into separate portions which code respectively for separate portions of the ricin molecule, for example for the A and B chains, which portions are then inserted into separate cloning vectors.

As stated above the cloning vector is preferably a plasmid such as pBR322, pAT153, of pUC8, and this may be cut open by means of the restriction endonuclease Pst I, and may be tailed with oligo (dG), and annealed with the double-stranded cDNA which has been tailed with oligo (dC).

We also provide a method of producing a modified transformed host by introducing into a suitable host microorganism a recombinant DNA molecule according to this invention.

The microorganism used as the host for cloning is preferably a Gram-negative bacterium and more preferably E. coli.

After cloning, the DNA sequence coding for the ricin precursor (or the precursor of another ricin-type toxin which is formed from a precursor) may be removed from the host cloning vector. It may then be divided into two portions which code for separate regions of the toxin molecule, for example the A and B chains, these portions introduced into separate second cloning vectors and new hosts modified with each of the resulting new recombinant DNA molecules. Alternatively it may be introduced while into a second cloning vector. The second cloning vectors comprise suitable promoter sequences and the position and direction of insertion of the whole coding sequence or portion thereof into the second cloning vector are such that on introducing the new recombinant DNA molecules into suitable host microorganisms, e.g. E. coli or S. cerevisiae, expression of the desired gene sequence is obtained.

The preparation of a transformed host containing a DNA sequence which codes for the ricin A and B chain precursor polypeptide will now be described by way of example, first in general terms and then in detail. This process is summarised in the accompanying diagram.

Firstly, the mRNA encoding this precursor was enriched in known manner by sucrose density gradient centrifugation. The corresponding cDNA was assembled in single-strand form onto this mRNA in known manner using the enzyme reverse transcriptase, a growing point having first been provided on the mRNA using oligo (dT) as primer which binds onto the polyadenylated 3'-terminus of the mRNA. The immediate product of this reaction is a DNA-RNA hybrid. The RNA strand is removed by hydrolysis, leaving the single-strand DNA intact. This is converted to the double stranded form using the enzyme DNA polymerase in the presence of free nucleotides, which results in a hairpin-shaped molecule, the curved end of which is then removed by the single-strand specific nuclease S1. The resulting double strand cDNA is then tailed with oligo (dC) using terminal transferase, size fractionated to remove small molecules, or vice versa and annealed with the pBR322 or pAT153 vector which has been cut open with Pst I and tailed with oligo (dG) using terminal transferase, the cytosine tails on the DNA base pairing with the guanine tails on the vector.

The resulting chimaeric plasmids containing the DNA segment coding for the ricin precursor polypeptide were then used to transform E-coli DH1 cells, and the presence of the chimaeric plasmid was ensured by selecting cells displaying tetracycline resistance and ampicillin sensitivity. Over 1600 Tet.sup.r, Amp.sup.s clones were obtained. Colonies derived from each clone were transferred to nitrocellulose filters and clones containing the desired DNA sequence identified using a 32P-end labelled 20 mer oligonucleotide probe, having the DNA sequence ##STR1## which hybridises to DNA containing the complementary sequence ##STR2## As the ricin precursor polypeptide has been found to contain the amino sequence--Trp-Met-Phe-Lys-Asn-Asp-Gly-- the DNA sequence responsible for this is deduced using the genetic code to be the latter mentioned above.

Using appropriate hybridisation and wash conditions, e.g. as described by Singer-Sam et al in (1983) Proc. Natl. Acad. Sci. (U.S.A.), Vol. 80 pp 802-806, 80 clones were selected as positively containing the desired DNA sequence, and of these, the eight largest in the plasmid pBR322 have been initially chosen for further characterisation. Their relationship to the castor bean lectin precursor polypeptides has been confirmed using the hybrid release translation assay. Of the eight clones mentioned above, four, respectively with 1614, 1950, 1059 and 1020 base pairs, have been selected for sequencing.

In detail, the transformed host was prepared as follows:

A. cDNA Synthesis

1. mRNA Extraction and Fractionation

100-200 g of ripening Ricinus seeds were frozen and ground to a powder in liquid nitrogen, and homogenised in a Waring blender for 1-2 minutes in 50 mM tris-HCl pH 9, 150 mM NaCl, 5 mM EDTA and 5% SDS. The homogenate was extracted with an equal volume of phenol:chloroform (1:1) and the phases were separated by centrifugation. The organic phase and residue were reextracted with 0.5 volume of 20 mM tris-HCl pH 9.0, 2 mM EDTA and the resulted aqueous phase was combined with the original one. The total aqueous phase was reextracted repeatedly with equal volumes of phenol:chloroform until no material was present at the interface. RNA was precipitated by the addition of 2 volumes of cold ethanol after making the solution 200 mM in NaCl.

After overnight precipitation at -20.degree. C. The RNA was centrifuged at 10,000 rpm for 30 minutes in an MSE 18 or MSE 21 centrifuge; the pellet was then washed repeatedly in 3M NaAc pH 5.5 until no polysaccharide was detectable in the supernatant by ethanol precipitation. The final pellet was dissolved in 300 mM NaCl and precipitated as above.

mRNA molecules bearing poly(A) tails were extracted by affinity chromatography on oligo(dT)-cellulose: after hybridisation at room temperature for 30 min. in 400 mM NaCl, 20 mM tris-HCl pH 7.6, 0.2% SDS, the beads were pelleted and washed three times in the above buffer and two times in 200 mM NaCl, 20 mM tris-HCl pH 7.6, 0.1% SDS. The slurry was poured into a column and washed further with the last buffer until the A.sub.260 of the eluate reached the background level. Poly(A)-containing RNA was then eluted with 20 mM tris-HCl pH 7.6 at 50.degree. C. The eluate was monitored with an ISCO continuous flow UV cell. Poly(A)-containing RNA was precipitated overnight from 200 mM NaCl by the addition of 2 volumes of cold ethanol at -20.degree. C. and was then washed three times with 70% ethanol, and redissolved in 10 mM tris-HCl pH 7.0 to approximately 1 .mu.g/.mu.l.

The mRNA was heated for 2 min. at 65.degree. C. and quench cooled. Approximately 400 .mu.g of poly(A).sup.+ RNA was layered on top of a 10-30% ribonuclease-free sucrose (Sigma) density gradient in 100 mM tris-HCl pH 7.5, 0.5% SDS, 1 mM EDTA, and centrifuged in a Beckman L5-65B centrifuge using an SW27 rotor at 25,000 rpm at 17.degree. C. for 14 hours. 400 .mu.l fractions were collected with an ISCO density gradient fractionator using the continuous flow UV cell.

Each fraction was made 200 mM in NaCl and precipitated with 2 volumes of cold ethanol by freeze-thawing in liquid nitrogen three times, and recovered by centrifugation in an Eppendorf microcentrifuge for 30 minutes at 4.degree. C., washed once with 70% ethanol, and redissolved in 10 .mu.l of 10 mM tris-HCl pH 7.0. An aliquot from each fraction (1 .mu.l) was translated in a reticulocyte lysate cell free system and the lectin precursor immunoprecipitated to identify the fraction enriched for lectin mRNA.

2. First Stand Synthesis

Fractionated poly(A).sup.+ RNA was reverse transcribed at 0.5 .mu.g/.mu.l in the presence of 50 mM tris-HCl pH 8.3, 10 mM MgCl.sub.2, 100 mM KCl, 1 mM of dATP, dTTP and dGTP, 250 .mu.M dCTP, 0.06 .mu.g/.mu.l oligo (dT).sub.12-18, 10 mM DTT and 0.4 units/.mu.l of reverse transcriptase from avian myeloblastosis virus. (3H)dCTP or .alpha.-(.sup.32 P)dCTP were included in the reaction as appropriate.

The reaction mixture was incubated at 42.degree. C. for 45 minutes, at which point an equal volume of 5 mM tris-HCl pH 8.3, 5 mM DTT, 250 .mu.M dCTP was added along with the same amount of enzyme as previously. The reaction was incubated for a further 45 minutes at 45.degree. C. and terminated by freezing. Aliquots were analysed on 1% denaturing agarose gels along with the products of the second strand and S.sub.1 nuclease reactions.

3. Second Strand Synthesis

mRNA-cDNA hybrids were denatured by boiling the first strand reaction for 3 minutes and cooling rapidly. After pelleting insoluble material in the Eppendorf microfuge for 2 minutes the supernatant was transferred to a fresh chilled tube. For the standard reaction, reagents were added as follows, ignoring elements already present: dATP, dGTP and dTTP to 100 .mu.M, Hepes-KOH pH 6.9 to 105 mM, KCl to 92 mM, dCTP, labelled as appropriate, to 80 .mu.M, and 0.1 units/.mu.l of DNA polymerase. The reaction was allowed to proceed at 20.degree. C. for 6 hours, at which time cDNA was removed from the mixture by gel filtration on 1 ml columns of Bio-Gel P60 in 10 mM tris-HCl pH 7.6, 20 mM NaCl, 1 mM EDTA. Fractions were monitored by Cerenkov or liquid scintillation counting, and peak excluded fractions were pooled and precipitated from 0.3M NaAc pH 6 by the addition of 2 volumes of cold ethanol. Precipitates were recovered by centrifugation in the Eppendorf microcentrifuge for 30 minutes in the cold, and dissolved in water to about 2.5 .mu.g/.mu.l of RNA--equivalent material.

4. S.sub.1 Nuclease Digestion

Single-stranded regions of double-stranded cDNA were digested with S.sub.1 nuclease from Aspergillus oryzae, in the presence of 300 mM NaCl, 30 mM NaAc pH 4.5, 3 mM ZnCl.sub.2. The reaction was incubated for 15 minutes at 37.degree. C. and then for 15 minutes at 15.degree. C., and was terminated by the addition of tris-HCl pH 7.6 to 130 mM and EDTA to 10 mM; it was then extracted with an equal volume of phenol:chloroform:isoamyl alcohol (25:24:1) and precipitated from 300 mM NaAc pH 6 with 2 volumes of cold ethanol. The precipitate was dissolved in 10 mM tris-HCl pH 8, 0.1 mM EDTA to 0.25 .mu.g/.mu.l RNA equivalent.

5. Addition of Homopolymer Tails to DNA

Double-stranded DNA was tailed using terminal transferase with dCTP at 0.001-0.01 .mu.g/.mu.l in the presence of 140 mM potassium cacodylate pH 7.6, 30 mM tris base, 0.1 mM DTT, 1 mM CoCl.sub.2 and (.sup.3 H) or (.sup.32 P)-labeled dCTP in 75-150 fold excess over 3' termini. The reaction was carried out at 37.degree. C. for 6 minutes. The extent of incorporation of label was followed by assaying the amount of TCA--insoluble radioactivity as a proportion of the total radioactivity, counting in Bray's scintillant.

The reaction was stopped by chilling and adding EDTA to 10 mM, after which unincorporated material was removed by gel filtration as described. Tailed cDNA was precipitated as before, and dissolved in 1M NaAc pH 8, 10 mM tris-acetate pH 8, 1 mM EDTA ready for fractionation.

PstI-cleaved pBR322 DNA was similarly treated, except that dGTP replaced dCTP.

6. Fractionation of Tailed cDNA

cDNA was fractionated on 5-20% linear sucrose density gradients in 1M NaAc pH 8, 10 mM tris-acetate pH 8, 1 mM EDTA, and centrifuged overnight at 39,000 rpm in an SW50.1 rotor. DNA sedimentation was checked on a parallel gradient loaded with a mixture of HinfI and PstI digests of pBR322 DNA, and fractions of this gradient were run on a 1% neutral agarose gel. Fractions from the cDNA gradient were diluted with an equal volume of water and precipitated with 2 volumes of cold ethanol, and then pooled to give three final fractions, a large cDNA fraction (larger than 2,200 bp), an intermediate fraction (1,000-2,200 bp) and a fraction containing smaller cDNAs (600-1,000 bp). cDNA molecules smaller than 600 bp were discarded.

The three final fractions were dissolved to approximately 5 ng/.mu.l in 150 mM RbCl, 10 mM tris-HCl pH 7.6, 0.2 mM EDTA.

B. Annealing and Transformation

1. Annealing

dC-tailed cDNA was mixed with dG-tailed pBR322 or pAT153 in approximately equimolar quantities, at a concentration of 0.4 ng/.mu.l of vector. Buffers were as described above. The mixtures were heated to 70.degree. C. for 30 minutes and then cooled overnight to room temperature, and slowly chilled to 4.degree. C. Competent cells were added and transformed as described below.

2. Preparation of Competent Cells and Transformation

DH1 cells.sup.x were grown in 10 ml cultures of psi broth (2% tryptone, 0.5% yeast extract, 10 mM NaCl, 20 mM MgCl.sub.2, pH adjusted to 7.6 with KOH; all bacteriological reagents from Difco), and grown at 37.degree. C. in a shaking waterbath to A.sub.550 =0.3. 1 ml of this was then inoculated into 25 ml of the F..sup.-, end A1, hsd R17 (r.sub.k -, m.sub.k -), supE 44, thi-1, lambda.sup.-, recA1, gyr A96, rel A1? same medium and grown to A.sub.550 =0.48. The cells were then chilled on ice for 15 minutes and harvested at 5,000 rpm for 5 minutes in an MSE 21 centrifuge at 4.degree. C. They were then resuspended in 10 ml of 100 mM RbCl, 50 mM McCl.sub.2, 10 mM CaCl.sub.2, 35 mM NaAc pH 5.8, 15% glycerol and kept on ice for 15 minutes.

The cells were again harvested, and resuspended in 1 ml of 10 mM RbCl, 75 mM CaCl.sub.2, 10 mM MOPS-KOH pH 5.8, 15% glycerol, and kept on ice for a further 15 minutes.

100 .mu.l of the cells thus prepared were mixed with the annealed DNA samples, and incubated on ice for 30 minutes, after which they were heat-shocked at 42.degree. C. for 90-120 seconds. 1 ml of psi broth was added, and the cells were grown at 37.degree. C. for 1 hour. They were then centrifuged briefly, and resuspended in 100 .mu.l of psi broth and plate on LB plates containing 14 .mu.g/ml tetracycline (LB is 1% tryptone, 0.5% yeast extract, 170 mM NaCl, 1.5% agar).

After 18-24 hours growth at 37.degree. C., colonies were counted and spotted onto LB plates containing 33 .mu.g/ml ampicillin to identify those transformants containing recircularised or uncut plasmids. Over 1600 Tet.sup.r Amp.sup.s clones were picked and transferred in ordered arrays onto large plates of LB containing 14 .mu.g/ml tetracycline.

C. Screening

1. Labelling of Oligonucleotide

The ricin B chain specific oligomer (20 mer) was end labelled using polynucleotide kinase. 500 ng of oligonucleotide was incubated in 50 mM tris pH 8.5, 10 mM MgCl.sub.2, 5 mM DTT, 0.1 mM spermidine-HCl, 0.1 mM EDTA with 60 .mu.Ci .delta.(.sup.32 p) ATP and 1 .mu.l polynucleotide kinase (Boehringer) for 35 min at 37.degree. C. The reaction was stopped by adding an equal volume of 0.6M NH.sub.4 AC and the bulk of non incorporated .delta.ATP was removed by passage through a sephadex G25 column in 0.14M NaCl, 0.02M tris pH 7.6, 0.005M EDTA, 0.1% SDS. The probe was stored frozen at -20.degree. C.

2. Colony Hybridization Using Oligonucleotide Probe

Transformants were grown on nitrocellulose filters (Schleicher & Schuell 0.45.mu.) layered over LB plus tetracycline. The filters, in triplicate, were then transferred to LB-Tet plates containing 200 .mu.g/ml chloramphenicol for 16 h at 37.degree.. The filters were placed colony side up on two sheets of 3 mm paper wetted with 0.5M NaOH for 15 mins at room temperature. The same procedure was followed for the following two washes (1) with 1M tris pH 8.0, and (2) with 1M tris pH 8, 1.5M NaCl (30 mins). The filters were air dried and baked at 80.degree. C.

Prehybridizations and hybridizations were done in double sealed polythene bags. The filters were prehybridized in 0.9M NaCl, 0.09M tris 7.4, 0.006M EDTA, 0.5% NP40, 2.times. Denhardts, 0.2% SDS, 100 .mu.g/ml denatured single strand salmon sperm DNA and 70 .mu.g/ml tRNA. Prehybridization was done for 4 h at 55.degree. C. The prehybridization buffer was then squeezed from the bag and fresh buffer added that contained 50 ng labelled probe (to give a maximum concentration of 5 ng/ml buffer). Annealing was done overnight at 37.degree. C.

Non stringent washes were done with 6.times. SSC at room temperature. The filters were washed in 4 changes of 6.times. SSC over 3 h. The triplicate filters were then washed at three different temperatures determined from the base composition and degree of mismatch of the probe. Using 2.degree. C. for every A or T and 4.degree. C. for every C or G in the probe the wash temperatures selected were 52.degree. C., 56.degree. C. and 60.degree. C. The filters were washed at the stringent temperature in 6.times. SSC for 10 minutes and then dried thoroughly. The filters were exposed to X-ray film overnight.

D. Hybrid Selection Procedure

1. DNA Binding

Plasmid DNA was purified from the positive clone(s) and 10-15 .mu.g linearized with EcoRl. After phenol:chloroform extraction and ethanol precipitation the pellet was dissolved in 0.5 ml 0.1.times. SSC. 0.5 ml 1M NaOH was then added and the mixture allowed to stand for 15 mins at room temperature. 4 ml of a prechilled neutralizing solution (1.5M NaCl, 0.25M HCl, 0.25M tris-Cl pH 8.0) was added and the 5 ml DNA sample sucked by vacuum through swinnies containing wetted Schleicher and Schuell 0.45.mu. filter discs. 5 ml 6.times. SSC was then passed through the filter(s). These were air dried and then baked for 2 h at 80.degree. C.

2. Hybrid Selection Protocol

The filter(s) were placed in 5 ml bottles and prehybridized for 4 h at 41.degree. C. in 50% formamide, 0.4M NaCl, 10 mM pipes-NaOH pH 6.4, 4 mM EDTA, 0.5 .mu.g/ml tRNA, 10 .mu.g/ml poly(A). The buffer was removed and the filter(s) typically hybridized overnight at 41.degree. C. in 50% formamide buffer (above) containing approx. 20 .mu.g poly (A).sup.+ RNA from castor beans. The buffer was removed and the filters washed twice for 15 mins in each of the following: (1) 1.times. SSC, 0.5% SDS at room temperature, (2) 0.1.times. SSC, 0.1% SDS at room temperature, (3) 0.1.times. SSC, 0.1% SDS at 50.degree. C., (4) 0.1.times. SSC, 0.1% SDS at room temperature. The filters were drained and 200 ul hybrid release buffer (90% formamide, 10 mM pipes-NaOH pH 6.4, 1 mM EDTA, 0.5% SDS) added to each and mixed for 30 mins at 40.degree. C. The buffer was removed into a fresh eppendorf and NaCl added to 0.2M. The released mRNA was precipitated with ethanol, rinsed several times in 70% ethanol, dried and dissolved in 5 .mu.l sterile water. The sample was translated in a reticulocyte lysate cell free system and the products run out directly on an SDS-polyacrylamide gel or firstly immunoprecipitated with appropriate antisera.

The DNA sequence coding for the above-mentioned ricin precursor polypeptide of two of the above-mentioned clones, referred to herein as pBRCL 6 and pBRCL 17 (RCL=Ricinus communis lectin) has now been determined by a combination of the Sanger dideoxy method (Sanger et al, 1977--Proc. Natl. Acad. Sci. U.S.A. 74, 5463-67) and the procedure of Maxam and Gilbert (Maxam and Gilbert, 1980--Meth. Enzym. 65, 499-560). In order to determine the sequence at the ends of each insert the inserts were excised from pBR322 and Pst I and ligated into Pst I linearized, phosphatased plasmid pUC8 (Vierra and Messing, 1982--Gene 19, 259-268). E. Coli DHI cells were transformed by these recombinant plasmids. These new recombinant plasmids are referred to herein as pRCL6 and pRCL17.

It is apparent that the two inserts contain a region of common sequence and that together they represent a total ricin precursor sequence. There are no nucleotide differences between the overlapping regions of the inserts in pRCL6 and pRCL17.

A new recombinant DNA molecule was then constructed that contains the complete nucleotide sequence encoding the ricin precursor polypeptide. This was achieved by isolating a fragment, 323 base pairs in length, obtained from pRCL17 by digestion with the restriction endonuclease Sau 961, and ligating this fragment to a fragment 1561 base pairs in length isolated after a partial digestion of pRCL6 with Sau 961 Ligation was performed in 50 mM tris HCl (pH 7.4) 10 mM MgCl.sub.2, 10 mM dithiothreitol, 1 mM spermidine, 10 mM ATP, 0.1 mg/ml BSA with 5 units commercial T4 DNA ligase, and incubation proceeded overnight at 15.degree. C. After a standard phenol/chloroform extraction and ethanol precipitation the ligated DNA was pelleted, dissolved in a small volume of 10 mM tris HCl (pH 7.4), 1 mM EDTA and digested to completion with Pst I. The resulting linearised DNA was then ligated (as above) with an equal quantity of Pst I linearised, phosphatased pUC8. The new recombinant DNA molecule containing the entire DNA sequence of the ricin precursor and referred to as pRCL617, was used in conventional manner to transform E. coli DHI cells.

The nucleotide sequence of pRCL617 is shown hereinafter.

This sequence was deduced from the two overlapping cDNA inserts in clones pRCL6 and pRCL17 (the limits of the DNA inserts in each of these two clones are given below).

Nucleotide residues are numbered in the 5' to 3' direction with the first residue of the codon specifying the amino terminal residue of mature ricin A chain numbered 1 and the nucleotides on the 5' side of reside 1 indicated by negative numbers. The 5' terminal sequence does not extend to the 5' end of the mRNA whereas the 3' terminal sequence shown is followed by a poly (dA) tract 27 residues long, thus representing the complete sequence of the region. The predicted amino acid sequence is given below the nucleotide sequence and differences with the published amino acid sequence of mature ricin A and B chains (Funatsu G., Kimura, M and Funatsu, M. Agric. Biol. Chem. Vol 43, pp 2221-2224 (1979), and Yoshitake, S., Funatsu, G and Funatsu, M--Agric. Biol. Chem. Vol. 42, pp 1267-1274 (1978)) are indicated underneath. Residues absent from the published amino acid sequence are underlined with a dashed line and the position of amino acids present in the published sequence but absent from the derived sequence presented here are indicated by an asterisk. The dashed line beneath the 12 amino acid sequence linking the C-terminus of the A chain and the N-terminus of the B chain is bracketed. Amino acids are numbered from the amino terminal residue of the mature A chain and the preceding residues are indicated by negative numbers. Potential sites for asparagine linked N-glycosylation are boxed and potential poly (A) signals are underlined. The insert of pRCL6 extends from nucleotide--102 to residue 1512 and the insert of pRCL17 extends from nucleotide 733 to residue 1782.

The intervening twelve triplets code for the linker amino acid sequence which is present in the precursor polypeptide and which is enzymatically removed in the cell to separate the A and B chains, which chains are joined by a disulphide bridge during the formation of the ricin molecule itself. This linker region as well as the presumptive amino terminal leader or signal sequence (amino acids -24 to -1) are not present in the sequences already published by Funatsu et al.

Preproricin is the whole polypeptide coded for by the aforesaid DNA insert, i.e. from amino acid -24 to amino acid 541. Proricin, which is obtained from preproricin in the organism by removal of the amino acid leader sequence, extends from amino acid 1 to amino acid 541.

The following nucleotide sequence is designated the nucleotide sequence of Table 1. The following amino acid sequence is designated as the amino acid sequence of Table 1. - .sub.-100 .sub.-50 . . . . . . . . . . 5'--AAACCGGGAG GAAATACTAT TGTAATATGG ATG TAT GCA GTG GCA ACA TGG CTT TGT TTT GGA TCC ACC TCA GGG TGG TCT TTC ACA TTA GAG Met Tyr Ala Val Ala Thr Trp Leu Cys Phe Gly Ser Thr Ser Gly Trp Ser Phe Thr Leu Glu .sup.-20 .sup.-10 .sub.-1 .sub.50 . . . . . . . . . GAT AAC AAC ATA TTC CCC AAA CAA TAC CCA ATT ATA AAC TTT ACC ACA GCG GGT GCC ACT GTG CAA AGC TAC ACA AAC TTT ATC AGA GCT Asp Asn Asn Ile Phe Pro Lys Gln Tyr Pro Ile ##STR3## Thr Ala Gly Ala Thr Val Gln Ser Tyr Thr Asn Phe Ile Arg Ala .sup.-1 .sup.10 .sup.20 .sub.100 .sub.150 . . . . . . . . . GTT CGC GGT CGT TTA ACA ACT GGA GCT GAT GTG AGA CAT GAT ATA CCA GTG TTG CCA AAC AGA GTT GGT TTG CCT ATA AAC CAA CGG TTT Val Arg Gly Arg Leu Thr Thr Gly Ala Asp Val Arg His Asp Ile Pro Val Leu Pro Asn Arg Val Gly Leu Pro Ile Asn Gln Arg Phe .sup.30 .sup.40 Glu .sup.50 .sub.200 .sub.250 . . . . . . . . . ATT TTA GTT GAA CTC TCA AAT CAT GCA GAG CTT TCT GTT ACA TTA GCC CTG GAT GTC ACC AAT GCA TAT GTG GTC GCC TAC CGT GCT GGA Ile Leu Val Glu Leu Ser Asn His Ala Glu Leu Ser Val Thr Leu Ala Leu Asp Val Thr Asn Ala Tyr Val Val Gly Tyr Arg Ala Gly .sup.60 Gln .sup.70 Ser .sup.80 .sub.300 .sub.350 . . . . . . . . . AAT AGC GCA TAT TTC TTT CAT CCT GAC AAT CAG GAA GAT GCA GAA GCA ATC ACT CAT CTT TTC ACT GAT GTT CAA AAT CGA TAT ACA TTC Asn Ser Ala Tyr Phe Phe His Pro Asp Asn Gln Glu Asp Ala Glu Ala Ile Thr His Leu Phe Thr Asp Val Gln Asn Arg Tyr Thr Phe .sup.90 .sup.100 .sup.110 .sub.400 . . . . . . . . . GCC TTT GGT GGT AAT TAT GAT AGA CTT GAA CAA CTT GCT GGT AAT CTG AGA GAA AAT ATC GAG TTG GGA AAT GGT CCA CTA GAG GAG GCT Ala Phe Gly Gly Asn Tyr Asp Arg Leu Glu Gln Leu Ala Gly Asn Leu Arg Glu Asn Ile Glu Leu Gly Asn Gly Pro Leu Glu Glu Ala .sup.120 .sup.130 .sup.140 .sub.450 .sub.500 . . . . . . . . . ATC TCA GCG CTT TAT TAT TAC AGT ACT GGT GGC ACT CAG CTT CCA ACT CTG GCT CGT TCC TTT ATA ATT TGC ATC CAA ATG ATT TGA GAA Ile Ser Ala Leu Tyr Tyr Tyr Ser Thr Gly Gly Thr Gln Leu Pro Thr Leu Ala Arg Ser Phe Ile Ile Cys Ile Gln Met Ile Ser Glu .sup.150 .sup.160 .sup.170 .sub.550 .sub.600 . . . . . . . . . GCA GCA AGA TTC CAA TAT ATT GAG GGA GAA ATG CGC ACG AGA ATT AGG TAC AAC CGG AGA TCT GCA CCA GAT CCT AGC GTA ATT ACA CTT Ala Ala Arg Phe Gln Tyr Ile Glu Gly Glu Met Arg Thr Arg Ile Arg Tyr Asn Arg Arg Ser Ala Pro Asp Pro Ser Val Ile Thr Leu .sup.180 .sup.190 .sup.200 .sub.650 .sub.700 . . . . . . . . . GAG AAT AGT TGG GGG AGA CTT TCC ACT GCA ATT CAA GAG TCT AAC CAA GGA GCC TTT GCT AGT CCA ATT CAA CTG CAA AGA CGT AAT GGT Glu Asn Ser Trp Gly Arg Leu Ser Thr Ala Ile Gln Glu Ser Asn Gln Gly Ala P he Ala Ser Pro Ile Gln Leu Gln Arg ##STR4## .sup.210 .sup.220 .sup.230 - - - Asp .sub.750 .sub.800 . . . . . . . . . TCC AAA TTC AGT GTG TAC GAT GTG AGT ATA TTA ATC CCT ATC ATA GCT CTC ATG GTG TAT AGA TGC GCA CCT CCA CCA TCG TCA CAG TTT ##STR5## Phe Ser Val Tyr Asp Val Ser Ile Leu Ile Pro Ile Ile Ala Leu Met Val Tyr A rg Cys Ala Pro Pro Pro Ser Ser Gln Phe .sup.240 Leu .sup.250 - - - .sup.260 .sub.850 . . . . . . . . . TCT TTG CTT ATA AGG CCA GTG GTA CCA AAT TTT AAT GCT GAT GTT TGT ATG GAT CCT GAG CCC ATA GTG CGT ATC GTA GGT CGA AAT GGT Ser Leu Leu Ile Arg Pro Val Val Pro Asn Phe Asn Ala Asp Val Cys Met Asp Pro Glu Pro Ile Val Arg Ile Val Gly Arg Asn Gly (- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -) .sup.280 .sup.290 .sub.900 .sub.950 . . . . . . . . . CTA TGT GTT GAT GTT AGG GAT GGA AGA TTC CAC AAC GGA AAC GCA ATA CAG TTG TGG CCA TGC AAG TCT AAT ACA GAT GCA AAT CAG CTC Leu Cys Val Asp Val Arg Asp Gly Arg Phe His Asn Gly Asn Ala Ile Gln Leu Trp Pro Cys Lys Ser Asn Thr Asp Ala Asn Gln Leu .sup.300 Asn Asn His .sup.310 .sup.320 .sub.1000 .sub.1050 . . . . . . . . . TGG ACT TTG AAA AGA GAC AAT ACT ATT CGA TGT AAT GGA AAG TGT TTA ACT ACT TAC GGG TAC AGT CCG GGA GTC TAT GTG ATG ATC TAT Trp Thr Leu Lys Arg Asp Asn Thr Ile Arg Ser Asn Gly Lys Cys Leu Thr Thr Tyr Gly Tyr Ser Pro Gly Val Tyr Val Met Ile Tyr - - - .sup.330 .sup.340 Pro Ser .sub.1100 .sub.1150 . . . . . . . . . GAT TGC AAT ACT GCT GCA ACT GAT GCC ACC CGC TGG CAA ATA TGG GAT AAT GGA ACC ATC ATA AAT CCC AGA TCT AGT CTA GTT TTA GCA Asp Cys Asn Thr Ala Ala Thr Asp Ala Thr Arg Trp Gln Ile Trp ##STR6## Ile Ile Asn Pro Arg Ser Ser Leu Val Leu Ala .sup.360 Thr Asp - - - Glu Asn .sup.380 .sub.1200 .sub.1250 . . . . . . . . . GCG ACA TCA GGG AAC AGT GGT ACC ACA CTT ACG GTG CAA ACC AAC ATT TAT GCC GTT AGT CAA GGT TGG CTT CCT ACT AAT AAT ACA CAA Ala Thr Ser Gly Asn Ser Gly Thr Thr Leu Thr Val Gln Thr Asn Ile Tyr Ala V al Ser Gln Gly Trp Leu Pro ##STR7## Gln .sup.390 .sup.400 Pro Phe .sub.1300 . . . . . . . . . CCT TTT GTT ACA ACC ATT GTT GGG CTA TAT GGT CTG TGC TTG CAA GCA AAT AGT GGA CAA GTA TGG ATA GAG GAC TGT AGC AGT GAA AAG Pro Phe Val Thr Thr Ile Val Gly Leu Tyr Gly Leu Cys Leu Gln Ala Asn Ser Gly Gln Val Trp Ile Glu Asp Cys Ser Ser Glu Lys Trp .sup.420 .sup.430 Val .sup.440 Ser Cys .sub.1350 .sub.1400 . . . . . . . . . GCT GAA CAA CAG TGG GCT CTT TAT GCA GAT GGT TCA ATA CGT CCT CAC CAA AAC CGA GAT AAT TGC CTT ACA AGT GAT TCT AAT ATA CGG Ala Glu Gln Gln Trp Ala Leu Tyr Ala Asp Gly Ser Ile Arg Pro Gln Gln Asn Arg Asp Asn Cys Leu Thr Ser Asp Ser Asn Ile Arg .sup.450 Ser Asn .sup.460 Asn Arg .sup.470 .sub.1450 .sub.1500 . . . . . . . . . GAA ACA GTT GTT AGG ATC CTC TCT TGT GGC CCT GCA TCC TCT GGC CAA CGA TGG ATG TTC AAC AAT GAT GGA ACC ATT TTA AAT TTG TAT Glu Thr Val Val Lys Ile Leu Ser Cys Gly Pro Ala Ser Ser Gly Gln Arg Trp Met Phe Lys Asn Asp Gly Thr Ile Leu Asn Leu Tyr .sup.480 .sup.490 Glu .sup.500 .sub.1550 .sub.1600 . . . . . . . . . AGT GGA TTG GTG TTA GAT GTG ACG CGA TCG GAT CCG AGC CTT AAA CAA ATC ATT CTT TAC CCT CTC CAT GGT GAC CCA AAC CAA ATA TGG Ser Gly Leu Val Leu Asp Val Arg Arg Ser Asp Pro Ser Leu Lys Gln Ile Ile Leu Tyr Pro Leu His Gly Asp Pro Asn Gln Ile Trp .sup.510 Ala .sup.520 Trp * - - - - - - .sub.1650 .sub.1700 . . . . . . . . TTA CCA TTA TTT TGA TAGACAGATT ACTCTCTTGC AGTGTGTGTG TCCTGCCATG AAAATAGATG ##STR8## GTAAATTTTG TAACTGAAAG Leu Pro Leu Phe *** *Leu Pro .sub.1750 .sub.1780 . . . . . . . GACAGCAAGT TATTGCAGTC ##STR9## ACTATTGTCT TGTGCATTCT AAATTT-Poly(A)