Molecular cloning and expression of DNA encoding ovine interleukin 2

Expression of ovine IL-2 / 223

MOLECULAR CLONING AND EXPRESSION OF DNA ENCODING OVINE INTERLEUKIN 2 Raymond Bujdoso, Mairi Williamson, Doug Roy, Paul Hunt, Blacklaws, David Sargan, Ian McConnell

Barbara

We have generated DNA encoding the mature form of ovine interleukin 2 (IL-2) by polymerase chain reaction (PCR) using primers complementary to sequences at the 59 and 39 ends of human, murine and bovine IL-2 cDNA. The predicted PCR product of 400 bp was ligated into the yeast Ty-P1 galactose-inducible expression vector pOGS40 which was used to transform yeast spheroplasts. The fusion protein, with a Factor Xa proteolytic cleavage site between ovine IL-2 and the P1 fusion partner, was expressed from galactose-induced transformed yeast. P1:IL-2 fusion protein, which self-assembles into virus-like particles (VLPs) due to the interaction of the P1 protein, was purified from lysates of mechanically disrupted yeast by centrifugation on a discontinuous sucrose gradient. Fusion protein was detected in Western blot analysis with polyclonal antisera raised to recombinant bovine IL-2. Soluble recombinant ovine IL-2 was released from the P1 fusion protein by cleavage with Factor Xa enzyme. After purification recombinant ovine IL-2 was functionally active as shown by its ability to support the proliferation of Con A-activated T cells and was capable of generating maedi visna virusspecific cytotoxic T cells from primed precursor cells. The availability of recombinant ovine IL2 will greatly help the analysis of the specificity of pathogen-specific cells in the sheep.

We are interested in studying the response of the ovine immune system to important ovine pathogens including Mycobacterium paratuberculosis and the lentivirus maedi visna virus (MVV). An essential element of this work is the in vitro analysis of the repertoire and functional capabilities of ovine T cells specific for bacterial or viral proteins derived from these pathogens. This type of analysis is possible through the understanding that antigen-specific clonal expansion of T cells occurs via a signal transduction cascade, initiated by the interaction of the T cell receptor for antigen with a complex of MHC molecule plus antigenic peptide, leading to the expression of IL-21 and its receptor.2 Combination of IL-2 with highaffinity IL-2 receptors on mature peripheral T cells leads to expression of transferrin receptors3 allowing the uptake of iron, an essential co-factor for the enzyme ribonucleotide-reductase involved in purine and pyrimidine biosynthesis which are required for DNA replication in the S phase of the cell cycle. The pivotal role of IL-2 in T cell biology4 is reflected by the fact

From Department of Veterinary Pathology, Summerhall, University of Edinburgh, Edinburgh EH9 1QH Correspondence to: Dr R. Bujdoso, Department of Clinical Veterinary Medicine, University of Cambridge, Madingley Road, Cambridge CB3 0ES Received 30 June 1994; accepted for publication 16 September 1994 © 1995 Academic Press Limited 1043-4666/95/03022319 $08.00/0 KEY WORDS: IL-3/DNA/sequence/expression/protein CYTOKINE, Vol. 7, No. 3 (April), 1995: pp 223–231

that antigen-specific T cells may be maintained indefinitely in vitro by alternate cycles of antigen stimulation and expansion with IL-2 alone. To provide unrestricted supplies of ovine IL-2 to grow antigen-specific ovine T cells in vitro, we sought to clone PCR-generated DNA encoding ovine IL-2 and to express this DNA as recombinant protein. A variety of eukaryotic expression vectors are available for the expression of cloned DNA. The yeast Ty retrotransposon vector pOGS40 allows rapid and efficient purification of recombinant proteins because of the physical properties of the expressed product. The native Ty element comprises two overlapping genes TYA and TYB. Expression of TYA results in 51 kD protein termed P1.5 TYB is expressed by a complex mechanism in which translation starts at the beginning of TYA, proceeds to the TYA:TYB overlap region where an infrequent riobosomal frame shift occurs, and finishes at the 39 end of TYB. This results in a 190 Kd TYA:TYB hybrid protein termed P3.5, 6 P1 selfassembles into virus-like particles (VLPs).5, 7 We have used the vector pOGS40 which contains that part of the TYA gene which encodes the first 381 amino acids of P1.8 By ligating foreign DNA at the 39 end of the truncated P1 gene within pOGS40, and separating the two genes by the coding region of a Factor Xa enzyme cleavage site, VLPs which comprise recombinant protein can be readily purified, and cleaved to release the protein of interest. Here we report the successful production of functional ovine recombinant IL-2 by this approach. The sheep is an important experimental sys223

224 / Bujdoso et al.

tem for the study of lentiviral and mycobacterial disease. The availability of ovine recombinant IL-2 will greatly assist in the generation of antigen-specific T cells to determine the role of the ovine immune system in these infections.

RESULTS Ovine IL-2 cDNA sequence We have generated DNA encoding the mature form of ovine IL-2 by PCR using primers which contain common sequences found in the coding regions of bovine, human and murine IL-2. The predicted PCR product of 408 bp was cloned into the phagemid pTZ18R and used to transform E. coli strain JM101. Insert DNA from two independent clones were fully sequenced and found to be identical, except for nucleotide position 74 (T↔C) and nucleotide position 213 (C↔T). Both clones predicted the same amino acid sequence. The nucleotide sequence of one of these clones is shown in Figure 1. During the course of this work a nucleotide sequence of the entire coding region of ovine IL-2 was reported which was identical to that shown here.9 This confirmed that the N-terminal amino acid sequence of ovine IL-2 is Asp-Pro-Thr-Ser-SerSer-Thr as was encoded in the 59 PRC primer used here.

CYTOKINE, Vol. 7, No. 3 (April 1995: 223–231)

The ovine IL-2 nucleotide sequence has 66%, 80% and 96% identity with the nucleotide sequences encoding mouse,10 human11 and bovine12 IL-2, respectively. As expected, the highest degree of identity was found with bovine IL-2 which differed by only 16 nucleotides from the sequence of mature ovine IL-2. The predicted sequence of mature ovine IL-2 which differed by only 16 nucleotides from the sequence of mature ovine IL-2. The predicted sequence of mature ovine IL-2 has 135 amino acids and a MW of 15.8 kD. A comparison of the predicted amino acid sequence of ovine IL-2 with IL-2 from several other species is shown in Figure 2. Conservation of amino acids occurs at the N-terminus (Asp-Pro-Thr-Ser-Ser-Ser-Thr) and towards the C-terminus (Phe-Cys-Gln-Ser-Ile-X-SerMet). Four Cys residues are highly conserved. Ovine IL-2 contains a potential glycosylation site at amino acids 59-61 which is present in bovine IL-2 but is not present in IL-2 from other species.12 Like IL-2 molecules from other species, the ovine IL-2 lacks a stretch of 12 Gln residues found so far only in murine IL-2.10 As is the case with human, bovine and mouse IL-2, the mature form of ovine IL-2 is predicted to have a neutral isoelectric point having an equal number of acidic (Asp 1 Glu) and basic (Arg 1 Lys 1 His) amino acid residues.

Expression of ovine IL-2 cDNA in yeast The coding sequence of the IL-2 PRC primers were flanked with BamHI sites to facilitate cloning into the yeast Ty-VLP expression system. The 59 primer contained the coding region of the Factor Xa cleavage site IEKR. BamHI-cut IL-2 insert from a pTZ18R clone was ligated into the unique BamHI site of the shuttle vector pOGS40 (Fig. 3). Recombinant clones in the E. coli strain JM83 were sequenced through the insertion site to select a clone with the correct orien-

Figure 1. Nucleotide sequence of mature ovine IL-2 DNA. The nucleotides are numbered from the first nucleotide of the sequence and an asterisk marks the stop codon. The nucleotide sequences of the primers used in the PCR are underlined. The EMBL accession number of this DNA sequence is X60148.

Figure 2. The predicted amino acid sequences (in single letter symbols) of ovine, bovine, murine and human IL-2. # signifies the positions of identical amino acids and ∼ the conserved amino acids within the different forms of IL-2.

Expression of ovine IL-2 / 225 TYA P1 Factor Xa site Interleukin 2 CCG GGA TCC ATA GAA GGT AGA GCA CCT .... ATG ACT

Bam H1

AP TY

1 PG K1 A m

pi

lin

r e s i s t a n ce

PAL P

ci l

pOGS40

LEU2

Figure 3. pOGS40 expression vector for the production of recombinant ovine IL-2.

tation and with a maintained insert reading frame. One plasmid, pTy-IL-2.1, was chosen for co-transformation of yeast spheroplasts with the vector pUG41S. Plasmid pUG41S overexpresses GAL4 protein under the control of a galactose-inducible promoter.13 Co-transformation of yeast with this vector and the VLP plasmid, pOGS40 containing a PAL promoter,8 results in increased expression levels of P1 fusion protein since a major restraint on the level of protein expression in galactose-inducible systems is the low level of GAL4 gene expression.14 Crude extracts of yeast transformants grown in liquid culture with galactose-induction were analysed by SDS-PAGE followed by Coomassie staining. Transformants showed the presence of a band of approximately 67 kD which was not present in the

lysates of untransformed yeast (Fig. 4, tracks 1 and 2). This band corresponds to the predicted size of the P1:IL-2 fusion protein (MW of monomer native P1 is approximately 51 kD). Expression of IL-2 was confirmed by Western blot analysis using a rabbit polyclonal anti-bovine IL-2 antisera which recognized a single prominent band at 67 kD in extracts from transformed yeast (Fig. 4, track 5). One yeast transformant yTy-IL2.1 was chosen for further work. The yTy-IL-2.1 transformant grew at a similar rate to non-transformed cells following induction in galactose medium. Various times of induction with galactose were investigated for their effect on the yield of fusion protein. No significant increase in the yield of the fusion protein in batch cultures was seen after 24 h of culture and this time was routinely used for induction of yTy-IL-2.1 yeast. Ty:IL2 fusion protein was purified by centrifugation of clarified yeast lysates on a sucrose gradient. Fractions containing fusion protein, detected by Coomassie staining and Western blot analysis with anti-IL-2.1 antisera, were pooled (Fig. 4, tracks 3 and 6). The yield of fusion protein was approximately 10 mg/litre of yeast culture. The DNA encoding the P1:IL-2 fusion protein was engineered to contain a Factor Xa protease cleavage site to allow the separation of IL-2 from Ty-encoded P1 protein.7 Efficient cleavage of fusion protein occurred with the neutral detergent sodium deoxycholate acid at an optimum final concentration of 0.09%. A time course for the cleavage of the P1:IL-2 fusion protein is shown in Figure 5A. Protein bands corresponding to native P1 (∼51 kd) and recombinant ovine IL-2 (∼16 kd) were evident after 2 h of the start of Factor Xa digestion. Further incubation with Factor Xa enzyme resulted in an increase in the intensity of these products but also in the appearance of an additional band of 11 kD. This lower molecular mass protein band may represent a cleavage product of IL-2, as prolonged incubation of fusion protein with Factor Xa enzyme

4

Figure 4. Expression and purification of Ty-IL-2 fusion protein. SDS-PAGE using a 5–20% acrylamide gradient gel was performed on lysates and purified Ty-IL-2 fusion protein derived from galactose-induced yTy-IL-2 transformed yeast. The gels were visualized by Coomassie stain (tracks 1–3) or by Western blot (tracks 4–6) using rabbit anti-bovine IL-2 sera. Tracks: 1 and 4, untransformed yeast; 2 and 5, yeast transformed with yTyIL-2; 3 and 6, purified Ty-IL-2 fusion protein.

5

6

226 / Bujdoso et al.

CYTOKINE, Vol. 7, No. 3 (April 1995: 223–231)

Figure 5. (A) Cleavage of Ty-IL-2 fusion protein with Factor Xa enzyme. 100 µl aliquots of Ty-IL-2 fusion protein at 1 mg/ml were cleaved with 0.01 mg/ml Factor Xa enzyme in cleavage buffer containing 0.09% deoxycholate acid at room temperature for the indicated times. (B) Preparation of soluble recombinant ovine IL-2. Sucrose gradient-purified Ty-IL-2 was digested as described in Materials and Methods, and soluble IL-2 purified by centrifugation. In both cases proteins were separated through a 15% acrylamide SDS-PAGE gel and visualized by silver stain.

resulted in a decrease in the intensity of the 16 kD IL-2 band and a corresponding increase in the intensity of 11 kD band (data not shown). Routinely, digestion times of 1–2 h were used to ensure production of full length IL-2. Recombinant IL-2 was removed from P1 protein by centrifugation of the Factor Xa cleavage reaction in the presence of 2-mercaptoethanol. Under these conditions VLPs comprising P1 protein were pelleted and IL-2 left in the supernatant. SDS-PAGE analysis followed by silver staining of liberated recombinant ovine IL-2 is shown in Figure 5B. The final yield of recombinant ovine IL-2 was ∼100 µg/l of yeast culture.

lymphocytes from MVV-infected sheep with live virus antigen. All three concentrations of ovine IL-2 allowed the generation of MVV-specific CTL (Fig. 7). CTL generated in the presence of ovine recombinant IL-2 had similar levels of activity as CTL generated using human recombinant IL-2 (60% lysis at E:T of 5:1). High concentrations of ovine recombinant IL-2 allowed induction of lymphokine-activated killer (LAK) activity and therefore, non-MHC restricted recognition of virus infected heterologous cells. In this CTL assay, the lymphocytes mediating lysis have been shown to be CD1.16 CTL can be generated by culture without IL-2, but cell recoveries are poor and the lytic activity of the lymphocytes low.

Functional activity of recombinant ovine IL-2

100 000

H-thymidine cpm

80 000 60 000 40 000

3

The function of IL-2 is to bind to IL-2 receptors and allow activated cells to progress into the S-phase of the cell cycle and onto proliferation.1 Figure 6 shows that ovine recombinant IL-2 stimulates the proliferation of Con A blasts in a dose-dependent manner. The response titrated in the picomolar range and plateaued at approximately 100 pM of added recombinant ovine IL-2. High affinity IL-2 receptors in the human system are saturated by concentrations of IL-2 greater than 100 pM.15 This suggests that the ovine recombinant IL2 generated here is capable of binding to and stimulating high affinity IL-2 receptors on sheep cells. The negative control for this proliferation assay was supernatant from Factor Xa-exposed native P1 protein that had been treated in same manner as P1:IL-2 fusion protein and prepared to similar concentrations. This control supernatant did not stimulate the proliferation of Con A blasts. The functional capability of ovine recombinant IL-2 was also assessed by its ability to aid the generation of MVV-specific cytotoxic T lymphocytes (CTL). Three dilutions of ovine IL-2 (10, 33 and 100 pM) were tested in 14 day cultures of peripheral blood

20 000

0

0.75 3 6 25 100 400 Ovine recombinant IL-2 pM (log2)

Figure 6. Stimulation of Con A-blast proliferation by recombinant ovine IL-2. Ovine Con A-blast cells were cultured at 5 3 10 4 cells/well for 3 days and proliferation was measured by 3H-thymidine-uptake over the last 5 h. Open circles: recombinant ovine IL-2. Closed circles: proliferation obtained with equivalent dilutions of supernatant from Factor Xa-treated native P1 protein.

Expression of ovine IL-2 / 227

A: Hu IL-2

B: 10 pM Ov IL-2

C: 30 pM Ov IL-2

D: 100 pM Ov IL-2

80 60

Percent specific 51Cr release

40 20 0 80 60 40 20 0 0

5

10

15

20

0

5

10

15

20

Effector : Target ratio

Figure 7. Specificity of visna-specific cytotoxic T cells generated by use of recombinant ovine IL-2. Peripheral blood lymphocytes from a persistently MVV infected sheep were cultured for 14 days with live virus and either 0, 10, 33 or 100 pM ovine recombinant IL-2, or 5 units/ml human recombinant IL-2. Live lymphocytes were purified and diluted to the appropriate effector:target ratios for use in a 6 h 51 Cr-release assay. Target cells were autologous (circles) or heterologous (triangles) skin cells mock- (open symbols) or MVV-infected for 72 h (closed symbols).

DISCUSSION We have used PCR to generate DNA encoding the entire open reading frame of mature ovine IL-2. The generated DNA has significant similarity with DNA sequences encoding mature IL-2 in other species, in particular bovine IL-2 with which it has 98% nucleotide identity. At the amino acid level, mature ovine IL-2 has approximately 60% identity with human IL-2. The conservation of structurally important Cys residues and the observation that human IL-2 will bind and stimulate physiologically relevant high-affinity IL-2 receptors across a wide species range, including sheep,17 suggests that the different species forms of IL-2 will share a similar overall structure. Human IL-2 has been the subject of several structure-function studies. Mutational analyses have identified a disulfide bridge, between Cys58-Cys105,18 certain amino acids towards the Nterminus19 and amino acids at the C-terminus19, 20 as being critical to the activity of the protein. On the basis of extensive immunological and more limited physical, biochemical and mutational data,21 a model for the structure of human IL-2 has been proposed in which the polypeptide folds to form four antiparallel-helices.

The helices are designated A, B, C and D and are all required for full biological activity of the IL-2 molecule. X-ray diffraction studies of human IL-2 have confirmed the presence of at least three helices, although the resolution was insufficient to specify the relative orientations of each helix.22 Indirect evidence to support the proposed pivotal role of these helical structures is that the first 26 amino acids of mature human IL-2 can be deleted from the molecule without loss of activity.23 Also, the greatest difference between human IL-2 and mouse IL-2 is the additional 12 glutamines at amino acids 15–26 of mouse IL-2. Both of these regions are proximal to the proposed most Nterminal α-helix A. The PCR product generated here was successfully expressed via the yeast retrotransposon Ty. The capability of P1 fusion proteins, like native P1, to selfassemble into Ty-VLPs7 provided a convenient system for the purification of IL-2 fusion protein. Purified preparations of fusion protein, at an average yield of 100 mg/litre of yeast culture, were obtained by the centrifugation of Ty-IL-2-VLPs over a discontinuous sucrose gradient. Insertion of a Factor Xa enzyme site

228 / Bujdoso et al.

CYTOKINE, Vol. 7, No. 3 (April 1995: 223–231)

between the ovine IL-2 and P1 protein sequences allowed the subsequent release of the 16 kD ovine IL-2 from the purified fusion protein allowing cleavage with Factor Xa enzyme. The requirement for the detergent sodium deoxycholate acid suggested during Factor Xa cleavage that some degree of unfolding of fusion protein is required to expose the proteolytic cleavage site to Factor Xa enzyme. Extended Factor Xa reaction times led to the generation of a second cleavage reaction product of ∼11 kD. It is likely that this 11 kD protein represents a breakdown-product of ovine recombinant IL-2 as judged by a decrease in intensity of the 16 kD band with a concomitant increase in intensity of the 11 kD band during extended cleavage. Inspection of the mature ovine IL-2 protein sequence shows the presence of sequence NIKR at amino acid residues 90–93. Cleavage by Factor Xa of the mature 16 kD ovine IL-2 at this site would result in two peptides: one of 11 kD, a candidate for the protein seen during extended reaction times, and one of 5 kD. Treatment of other fusion proteins with Factor Xa have shown cleavage at sites other than its proposed single specific recognition sequence of IEKR.24 Examples include the prothrombin sequence IDGR25 (Magnusson et al., 1984), the HIV p24 sequence ISPR26 (Gilmour et al., 1989), the IL-1α sequence IKPR and IL-1β sequence IEEK 27 (fiskerstrand et al., 1992), and the Mycobacterium paratuberculosis hsp60 sequence PKGR (A. Colston et al. unpublished). This promiscuous cleavage activity of Factor Xa enzyme may limit the range of fusion proteins which may be successfully cleaved. Despite the potential for inadvertent cleavage of mature ovine recombinant IL-2, full length lymphokine protein was successfully produced in sufficient quantities to be used for T cell expansion. The ovine recombinant IL-2 liberated from cleaved fusion protein was active in bioassays which measured either proliferation alone, in the case of the Con A blasts response, or proliferation and differentiation, in the case of the generation of cytotoxic T cells. Ovine recombinant IL-2 supported the proliferation of activated Con A blasts in the picomolar range. We have previously reported that it is in this concentration range that human and bovine recombinant IL-2 stimulate

proliferation of activated ovine T cells, and that IL-2 binds to physiologically relevant high-affinity IL-2 receptors which are responsible for signal transduction during T cell proliferation. Ovine recombinant IL-2 was also functionally active in the in vitro generation of MVV-specific cytotoxic T cells from precursor cells in ovine peripheral blood. IL-2 is known to be necessary for the induction of CTL to tumour and viral antigens.28,29 The culture system used here showed that addition of exogenous ovine IL-2 increased the recovery of viable cells and their activity at some effector to target concentrations. These results clearly demonstrate that ovine viral-specific CTL can be generated in vitro using ovine recombinant IL-2 and will allow analysis of important MVV antigens through the use of MVV-specific T cells.16 Collectively, our report describes the usefulness of combining PCR technique and use of the yeast TyVLP expression system for the rapid cloning and generation of recombinant protein. Other ovine lymphokines have been generated using this approach including TNF-α,30 IL-1α and IL-1β.27 We have found that TNF-α is expressed in greater amounts as a Ty fusion protein than is IL-2 or IL-1α and IL-1β. So far, we have been unable to establish criteria which allow predictions to be made of the level of expression of different fusion proteins in this system. However, for those proteins which are produced in reasonable amounts as Ty-VLPs, this system allows the purification of functionally active, soluble, recombinant protein material, or fusion protein in a particulate form which may be highly immunogenic for the production of monoclonal antibodies.

MATERIALS AND METHODS Cloning of IL-2 DNA Polyadenylated RNA was prepared from ovine peripheral blood mononuclear cells, previously stimulated with 5µg/ ml Con A for 10–16 h in vitro, by oligo-dT cellulose column chromatography. It was used to generate cDNA for use as substrate in a polymerase chain reaction (PCR) using primers complementary to sequences at the 59 and 39 ends of bovine, human and mouse IL-2. The full primer sequences were as follows:

59 primer

BamHI Factor Xa site first 18 bp of IL-2 TCCCGGGATCCATAGAAGGTAGAGCACCTACTTCAAGCTCT

39 primer

BamHI Stop Last 27 bp of IL-2 AGACCCGGGATCCTTCTATCAAGTCATTGTTGAGTAGATGCTTTG

Expression of ovine IL-2 / 229

PCR was carried out as described by Saiki31 using 0.1 nM of each primer and 2 mM MgC12 . The 400 bp reaction product was blunt-end ligated into the phagemids pTZ19R and pTZ18R and rescued by transformation of JM101 bacteria. The 408 bp insert of three independent transformants were fully sequenced by the di-deoxy method.32 A BamHI insert from one transformant containing plasmid designated pOvIL-2.1 was ligated into the unique BamHI site of the yeast/E. coli shuttle vector pOGS408 (a gift from British Biotechnology Ltd, Oxford) and rescued by transformation of JM83 bacteria. Appropriate transformants grown on LB-ampicillin agar plates were selected by restriction digest of DNA and double stranded sequencing through 59 vector-insert junction of pOGS40 to confirm those which contained the 408 bp ovine IL-2 DNA in the correct orientation and with a maintained reading frame for fusion protein expression. One plasmid designated pTy-IL-2.1 which satisfied these criteria was purified by ultracentrifugation on a caesium/ethidium bromide gradient, and used for expression in yeast.

Yeast transformation Media, yeast culture conditions and induction for protein expression were as described by Kingsman et al.8 Briefly, spheroplasts of the protease deficient Saccharomyces cerevisiae strain BJ 2168 (a, leu2, trp1, ura3-52, prb1-1122, pep4-3, pcr1-407, gal2)33 were simultaneously transformed to leucine and uracil independence with plasmids pTy-IL-2.1 and pUG41S then subsequently plated in regeneration agar {0.67% yeast nitrogen base (YNB), 1 M sorbitol, 1% glucose and 30% Difco agar} and incubated at 30°C for 3–7 days. Yeast colonies were restreaked onto 0.67% YNB plates and subsequently grown in small scale liquid culture for 48 h at 30°C in YNB containing 1% glucose with 0.002% tryptophan followed by induction for 24–48 h in media containing 1% galactose plus 0.3% glucose with 0.002% tryptophan.34

Purification of recombinant ovine IL-2 Ty-IL-2 fusion protein was purified from 5–10 litres of galactose-induced yeast culture grown in conical flasks on an orbital shaker at 30°C. Pelleted cells were mechanically disrupted by vortexing with glass beads in TEN buffer (100 mM TrisHCl pH 7.4, 2 mM EDTA pH 8, 140 mM NaCl) containing the protease inhibitors (all obtained from Sigma) apoprotin, antipain, chymostatin, leupeptin and chymotrypsin all at 0.625 µg/ml, and PMSF at 5 mM. The yeast lysate was clarified by centrifugation at 10 000 3 g for 30 min and the supernatant containing VLPs applied to a 15, 25, 35, 45% discontinuous sucrose gradient, underlayered with 60% sucrose, then centrifuged at 120 000 3 g for 3 h at 4°C. Fractions of 1.5 ml volume were collected from the bottom of the gradient by tube puncture and analysed by SDS-PAGE and Western blot analysis with rabbit anti-Ty or rabbit antibovine IL-2 sera. Samples containing hybrid Ty-pIL-2 VLPs were pooled, dialysed into cleavage buffer (50 mM Tris HCl pH 7.4, 100 mM NaCl, 1 mM CaCl2) and digested with Factor Xa enzyme (Boehringer Mannheim, cat. no. 1179-896) at a ratio of 100:1 (w/w of VLPs: enzyme) in the presence of 0.09% sodium deoxycholic acid for 1–2 h at room temperature. 2-mercaptoethanol was added to a final concentration

of 5% and the cleavage reaction centrifuged at 350 000 g for 1 h 4°C to pellet Ty-VLPs. The supernatant which contained solubilised IL-2 was dialysed extensively against PBS and stored at 270°C.

Protein analysis Protein preparations were analysed by SDS-PAGE run under reducing conditions. Total proteins were detected and quantified by staining gels with 0.25% Coomassie Brilliant Blue G-250 in 20% methanol, 5% acetic acid along with standard protein samples. For Western blot analysis, proteins were electrophoresed as above then transferred to nitrocellulose membranes (Hybond-C, Amersham) using a semi-dry electroblotter. After blocking in PBS plus 5% dried milk powder (Marvel), blots were probed by incubation overnight in specific antibody and washed five times in PBS plus 1% dried milk powder plus 0.1% Tween 80 (PBS-MP) over 25 min. Blots were incubated for 1 h at room temperature with affinity purified anti-mouse IgG-biotin conjugate (Sigma, cat. no. B7264), washed with PBS-MP as before and incubated for 1 h at room temperature with a complex of streptavidin (Boehringer Mannheim, cat. no. 973-190) preincubated with a biotin-alkaline phosphatase conjugate (Boehringer Mannheim, cat. no. 1-119-834). Blots were washed again before development with 330 µg/ml nitro-blue tetrazolium (Sigma, cat. no. N6876) and 165 µg/ml 5-bromo4-chloro-3-indolyl phosphate (Sigma, cat. no. B0274) in 0.1M Tris HCl, pH 9.5, 100 mM NaCl and 5 mM MgCl2 .

Proliferation of Con A blast cells Medium: RPMI 1640 (Gibco Biocult, Uxbridge, cat. no. 074-1800) was supplemented with 2 mM L-glutamine, 100 U/ ml benzyl penicillin and 100 U/ml streptomycin, 5 3 1025 M 2-mercaptoethanol and 2 g/litre sodium bicarbonate. Wash medium consisted of RPMI 1640 and supplements with 1% FCS (flow Laboratories, Irvine, Ayrshire). Culture medium contained 10% serum. Preparation of Con A-blast cells: PBMC were isolated by centrifugation of heparinized venous blood at 800 3 g for 20 min to obtain buffy coat cells which were subsequently centrifuged at 600 3 g for 20 min at room temperature over Lymphoprep (Nycomed, Denmark). Cells collected at the Lymphoprep density medium interface were washed three times. Cells were suspended at 2 3 106 /ml in culture medium with 5 µg/ml Con A (Sigma, cat. no. C-2010) for 4 days at 37°C. Viable blast cells were harvested over Lymphoprep and resuspended at 1 3 106/ml and dispensed in 100 µl aliquots into 96-well flat-bottom plates. Triplicate dilutions of IL-2 were added in 100 µl volumes and the culture incubated at 37°C in a humidified atmosphere of 5% CO2/95% air for 3 days. Proliferation was measured by 3H-thymidine-uptake over the last 5 h of culture.

Cytotoxic assay Generation of cytotoxic T cells: CTL were prepared as described elsewhere.16 Briefly, PBMC prepared as above were washed twice with wash medium plus 2% FCS and resuspended at 2 3 10 6 cells/ml in culture medium supplemented with 20 mM Hepes pH 7.2 and either 10, 33 or 100

230 / Bujdoso et al.

pM ovine recombinant IL-2, or 5 units/ml human recombinant IL-2. 2 3 106 PBMC were added to 105 autologous skin cells which had previously been infected with 0.5 tissue culture infectious dose50 /cell MVV EV-135 in DMEM / 2%FCS for 90–120 min at 37°C in 24-well plates. PBMC and infected skin cells were cultured at 37°C in 5% CO2/95% air for 3 days and re-fed with 0.5 ml IL-2 containing medium before culture for a further 4 days. Viable cells were harvested over Lymphoprep and resuspended at 2 3 106 cells/ml in culture medium with IL-2 then replated on infected autologous skin cells and cultured for 7 days with feeding of IL-2 as above. Viable cells were finally harvested by centrifugation over Lymphoprep and washed twice before resuspension in culture medium. Cytotoxic T lymphocyte assay: Targets were 10 4 autologous or heterologous skin cells either mock infected or infected with 0.5 TCID50/cell MVV EV-1 for 56 h and labelled with 2.5 µCi51Cr (sodium chromate, 350–600 mCi/ ml51Cr, Amersham International PLC) per 104 cells over the last 24 h at 5% CO2/95% air before washing four times with RPMI / 2% FCS. Lymphocyte effectors and targets were mixed at effector : target ratios of 2.5:1 to 20:1 in culture medium in a total volume 200 µl before incubation for 16 h at 37°C in 5% CO2/95% air after which time 100 µl of culture supernatant was counted in a LKB Wallac 1274 RIAGAMMA machine. The results are expressed as percentage specific 51Cr release.16 All values were obtained from triplicate samples and spontaneous 51 Cr release was always less than 30% of the maximum release.

Acknowledgements This work was supported by AFRC Link Grant LRG31. BAB and DJR are supported by Wellcome Trust Programme Grant 035157.

REFERENCES 1. Smith K (1984) Interleukin 2. Annu Rev Immunol 2: 319–333. 2. Minami Y, Kono T, Miyazaki T, Taniguchi T (1993) The IL-2 receptor complex—its structure, function and target genes. Annu Rev Immunol 11:245–268. 3. Neckers LM, Cossman J (1983) Transferrin receptor induction in mitogen-stimulated human T lymphocytes is required for DNA synthesis and cell division and is regulated by interleukin 2. Proc Natl Acad Sci USA 80:3494–3498. 4. Smith K (1992) Interleukin 2. Curr Opin Immunol 4: 271–276. 5. Mellor J, Fulton SM, Dobson MJ, Wilson W, Kingsman SM, Kingsman AJ (1985) A retrovirus-like strategy for expression of a fusion protein encoded by yeast transposon Ty-1. Nature 313:243– 246. 6. Wilson W, Malim MH, Mellor J, Kingsman AJ, Kingsman SM (1986) Expression strategies of the yeast retrotransposon Ty—a short sequence directs ribosomal frameshifting. Nucleic Acids Res 14:7001–7016. 7. Adams SE, Mellor J, Gull K, Simms RB, Tuite MF, Kingsman SM, Kingsman AJ (1987) The functions and relationships of Ty-VLP proteins in yeast reflect those in mammalian retroviral proteins. Cell 49:111–119. 8. Kingsman SM, Cousens D, Stanway C, Chambers A, Wilson W, Kingsman AJ (1990) High efficiency yeast expression vectors

CYTOKINE, Vol. 7, No. 3 (April 1995: 223–231) based on the promoter of the phosphoglycerate kinase gene. Methods Enzymol 185:329–341. 9. Goodall JC, Emery DC, Perry ACF, English LS, Hall L (1990) cDNA cloning of ovine interleukin 2 by PCR. Nucleic Acids Res 18:5883. 10. Kashima N, Nishitakaoka C, Fujita T, Taki S, Yamada G, Hamuro J, Taniguchi T (1985) Unique structure of murine interleukin 2 as deduced from cloned cDNAs. Nature 313:402–404. 11. Taniguchi T, Matsui H, Fujita T, Takaoka C, Kashima N, Yoshimoto R, Hamuro J (1983) Structure and expression of a cloned cDNA for human interleukin 2. Nature 302:305–310. 12. Reeves R, Spies AG, Nissen MS, Buck CD, Weinberg AD, Barr PJ, Magnuson NS, Magnuson JA (1986) Molecular-cloning of a functional bovine interleukin 2 cDNA. Proc Natl Acad Sci USA 83:3228–3232. 13. Lue NF, Chasman DI, Buchman, AR, Kornberg RD (1987) Interaction of GAL4 and GAL80 gene regulatory proteins in vitro. Mol Cell Biol 7:3446–3451. 14. Johnston SA, Hopper JE (1982) Isolation of the yeast regulatory gene GAL4 and analysis of its dosage effects on the galactose/melobiose regulon. Proc Natl Acad Sci USA 79:6971–6975. 15. Robb RJ, Greene WC, Rusk CM (1984) Low and high affinity cellular receptors for interleukin 2. Implications for the level of Tac antigen. J Exp Med 160:1126–1146. 16. Blacklaws BA, Bird P, Allen D, McConnell I (1994) Circulating cytotoxic T lymphocyte precursors in maedi visna virus-infected sheep. J Gen Virol 75:1589–1596. 17. Bujdoso R, Young P, Hopkins J, McConnell I (1990) IL-2like activity in lymph fluid following in vivo antigen challenge. Immunol 69:45–51. 18. Wang A, Lu SD, Mark DF (1984) Site-specific mutagenesis of the human interleukin 2 gene-structure-function analysis of the cysteine residues. Science 224:1431–1433. 19. Ju G, Collins L, Kaffka KL, Tsien WH, Chizzonite R, Crowl R, Bhatt R, Kilian PL (1987) Structure-function analysis of human interleukin 2—identification of amino-acid residues required for biological activity. J Biol Chem 262:5723–5731. 20. Liang SM, Thatcher DR, Liang CM, Allet B (1986) Studies of structure-activity relationships of human interleukin 2. J Biol Chem 261:334–337. 21. Cohen FE, Kosen PA, Kuntz ID, Epstein LB, Ciardelli TL, Smith KA (1986) Structure-activity studies of interleukin 2. Science 234:349–352. 22. Brandhuber BJ, Boone T, Kenney WC, Mackay DB (1987) Crystals and a low resolution structure of interleukin 2. J Biol Chem 262:12306–12308. 23. Zurawski SM, Mosmann TR, Benedik M, Zurawski G (1986) Alterations in the amino-terminal 3rd of mouse interleukin 2—effects on biological-activity and immunoreactivity. J Immunol 137:3354–3360. 24. Nagai K, Thogersen HC (1984) Generation of β-globin by sequence-specific proteolysis of a hybrid protein produced in Escherichia coli.. Nature 309:810–812. 25. Magnusson S, Petersen TE, Sottrup-Jensen L, Cleays H (1975) In Reich E, Rifkin D, Shaw E (eds) Proteases and Biological Control. Cold Spring Harbour Laboratory, New York, pp123–149. 26. Gilmour JEM, Senior JM, Burns NR, Esnouf MP, Gull K, Kingsman SM, Kingsman AJ, Adams SE (1989) A novel method for the purification of HIV-1 p24 protein from hybrid Ty virus-like particles (Ty-VLPs). AIDS 3:717–723. 27. fiskerstrand CE, Roy DJ, Green IR, Sargan DR (1992) Cloning, expression and characterisation of ovine interleukins-1α and β. Cytokine 4:418–428. 28. Simon MM, Landolfo S, Diamanstein T, Hochgeschwender V (1986) Antigen-sensitised and lectin-sensitised murine cytolytic lymphocyte-T-precursors require IL-2 and endogenously produced IFN-γ for their growth and differentiation. Curr Top Microbiol Immunol 126:173–185.

Expression of ovine IL-2 / 231 29. Kos F, Mullbacher A (1992) Specific epitope-induced conversion of CD81 memory cells into effector cytotoxic T lymphocytes in vitro. Eur J Immunol 22:1595–1601. 30. Green IR, fiskerstrand CE, Bertoni G, Roy DJ, Peterhans E, Sargan DR (1993) Expression and characterisation of bioactive recombinant ovine TNF-α: some species specificity in cytokine response to TNF. Cytokine 5:213–223. 31. Saiki RK (1989) The design and optimisation of the PCR. In Erlich, HA (ed) Polymerase Chain Reaction. Cold Spring Harbour Laboratory, New York, CR Technology, Stockton Press, pp25– 30.

32. Sanger F, Nicklen S, Coulson AR (1977) DNA sequencing with chain terminating inhibitors. Proc Natl Acad Sci USA 74:5463– 5477. 33. Jones EW (1991) Tackling the protease problem in Saccharomyces cerevisiae. Methods Enzymol 194:428–453. 34. Hinnen A, Hicks JB, fink GR (1978) Transformation of yeast. Proc Natl Acad Sci USA 75:1929–1933. 35. Sargan DR, Bennet ID, Cousens C, Roy DJ, Blacklaws BA, Dalziel RG, Watt NJ, McConnell I (1991) Nucleotide sequence of EV-1, a British isolate of maedi-visna virus. J Gen Virol 72:1893–1903.