Hajor histocompatibility complex (HHC) class II gene products are cell surface heterodimeric glycoproteins composed of 34 kilodalton α- and 29 kilodalton β-chains. These proteins play a crucial role in the immunological processes of self-tolerance, HHC restriction and antigen presentation. Each of these processes involves the interaction of HHC molecules with antigen, T cell α-β receptors (TCRs) and accessory molecules such as CD4. The relationship between these interactions and the biology of tolerance induction and HHC-restricted antigen presentation is, as yet, not fully understood.

Both HHC and TCR gene products are characterised by extensive structural complexity (Natural History of the Major Histocompatibility Complex, 1986, J. Klein - Viley & Sons). An important functional consequence of this complexity is diversity in antigen recognition. In the evolution of independent mammalian species, MHC and TCR complexes have diverged significantly in gene structure, number and regulation.

awrance et al (Cell, 1989, 5£J, 583-594) report that despite considerable divergence between the amino acid sequences of the human OR-α and murine IEα-chains of HHC class II gene complex it is possible to construct transgenic DRα-Eβ mice. Such mice express the human DRα gene with no apparent modulation of immune function.

Nishimura et al (J. Immunology, 1990, JL45, 353-360) describe a line of transgenic HLA-DQw6 mice. A mixture of DQw6A and DQw6B DNA was injected into 40 fertilised eggs from C57BL/6 mice. Four neonate pups were obtained out of which one mouse carried both the DQw6A and DQw6B genes. However despite the desirability of further MHC class II double transgenes there have been no further reports of these.

In a first aspect of the present invention we now provide a transgenic non-human mammal which is α,β-double transgenic for an MHC class II HLA-DR gene complex.

Examples of convenient MHC class II H A-DR gene complexes include those disclosed by Kappes and Strominger in Ann. Rev. Biochem., 1988, 57, 991-1028; Harsh and Bodmer in "Gene Registry, HLA class II nucleotide sequences, 1991"; Lanchbury in Clinical and Experimental Perspectives in Cyclosporin Therapy, 1992, 1_ (1), 15-17; Wordsworth et al in Am. J. Hum. Genet., 1992, 51 _^ , 585-591 and P.N.A.S., 1989, 8_6, 10049-10053. Such complexes include DR1, DRwlO and DR4 subtypes including DR4Dw4, DR4Dwl3, DR4Dwl4.1, DR4Dw 14.2 and DR4Dwl5. A particular complex is DR4Dwl4. A further particular complex is the human HLA-DR4Dw4-α,β complex.

The mammal is any convenient non-human mammal, for example as used in laboratory test procedures, such as rodents for example mice or rats, more conveniently mice.

The transgenic non-human mammal of the invention may be either heterozygous or homozygous for the desired α,β-tr-insgene, but is conveniently homozygous.

A preferred non-human mammal of the invention is a mouse transgenic for the human H A-DR4Dw4-α,β complex, find preferably homozygous for the α and β transgenes.

The transgenic non-human mammal of the invention is conveniently obtained by mating appropriate individual non-human mammals who are α- transgenic and β-transgenic respectively for the desired MHC class II DR gene complex. Conventional mating and breeding procedures may be employed. The desired double transgenic progeny are preferably back-crossed into a background strain which does not have any significant deletion in either the α- or β- gene. This is to minimise any unwanted human/non-human gene pairings and to promote reliable expression of the desired human α,β-transgene.

The level of α,β-transgene expression is conveniently at least 1/lOth that of mouse MHC class II and more conveniently the same.

In a further aspect of the present invention we now provide

a human β-DR4Dw4 transgene.

The transgenic non-human mammals of the invention have a number of uses including the study of any aspect of the immune system relating to MHC class II as well as drug and vaccine assays for example to determine the interaction of compounds with these cell surface proteins. By way of example tissue can be removed and (i) cultured (eg. "Culture of Animal Cells - A Manual of Basic Technique" - R. Ian Freshney, A.R. Liss Inc. New York), (ii) used in assays, (iii) immortalised in cell lines using procedures such as those outlined by Caroline MacDonald in Critical Reviews in Biotechnology, 1990, _10 (2), 155-178 or (iv) used to prepare stem cells such as those outlined in "Teratocarcinomas and Embryonic Stem Cells - a practical approach" Ed. E.J. Robertson, IRL Press.

In a further aspect of the invention we provide a mammalian cell line which is heterozygous, homozygous or chimeric for expression of an MHC class II α,β-double trans ene. Examples of convenient MHC class II α,β-double transgenes include α,β-double transgenes of the DR class of sequences, especially of DR1; the Dw class of sequences, especially of Dw4, Dwl4.1 and Dw 14.2. A particular α,β-double transgene is of DR4Dwl4. A more particular transgene is the α,β-double transgene of human HLA-DR4Dw4.

Any convenient cell line may be used. These include those disclosed as above such as immortalised cell lines. The cell line may be pluripotent and/or totipotent.

According to a further aspect of the present invention we provide a polypeptide expressed by the α,β-double transgene of the invention according to any preceeding aspect of the invention.

It is known that DR MHC class II gene products expressed by the transgenic non-human mammals of the present invention may interact with other factors to induce proliferation in human T cells.

Therefore in a further aspect of the present invention we provide a method for the identification of compounds which modulate

the production or release of an immunologically active mediator which method comprises contacting a test compound with a polypeptide expressed by the α,β-double transgene of the invention and a factor which binds to the polypeptide to form a complex, which complex interacts with T cells and leads to the production or release of the immunologically active mediator, and detecting any modulation of production or release.

Whilst we do not wish to be bound by theoretical considerations the test compound may modulate the production or release of the immunologically active mediator by inhibiting binding of the polypeptide to the factor or any subsequent event.

Examples of convenient immunological events which may be used to detect any modulation of production or release of an immunologically active mediator include T cell proliferation, antibody production, and an inflammatory response.

Any convenient test compound may be used. For control purposes or otherwise, antibodies such as monoclonal antibodies may be used to block the polypeptide expressed by the α,β-double transgene.

The factor is conveniently a peptide, conveniently a synthetic peptide. A particular peptide for use in the above method in conjunction with the HLA-DR4Dw4 transgene is FHA- ₀₇ _ ₃₂₀ (Pro-Lys-Tyr-Val-Lys-Gln-Asn-Thr-Leu-Lys-Leu-Ala-Thr-Gly).

The invention will now be illustrated but not limited by reference to the following Figures and Examples wherein:

Figure 1 shows the construction of the DRα clone T9C. The boxes represent exons of the DRα gene and the arrow indicates the direction of transcription. 5'SS is the signal sequence, αl and α2 are the extracellular domains. TC are the transmembrane and cytoplasmic domains. 3'UT is the 3' untranslated region, βl is an exon of the DRβ gene. Restriction maps for the EcoRI, BamHI and Sail

restriction enzymes are also shown. The 22kb BamHI-Sall fragment, including the 5 DRα exons, was microinjected into mouse oocytes to generate a DRα transgenic.

Figure 2 shows the generation of the DRβ construct. The boxes represent exons of the mouse IEα construct and the arrow indicates the direction of transcription. L is the leader peptide. 1 and 2 are the extracellular domains. TC are the transmembrane and cytoplasmic domains. 3'UT is the 3' untranslated region. Restriction sites for the BamHI and Xbal restriction enzymes are also shown. The human DRβ cDNA was cloned into the first exon of the mouse IEα gene, between the transcription initiation site and the translation start codon.

Figure 3(a) shows an autoradiograph of a Southern Blot including T9C DNA, liver DNA from non-transgenic mice, BSH DNA and DNA from the tails of transgenic and non-transgenic progeny hybridised to DRα cDNA.

Figure 3(b) shows an autoradiograph of a Southern Blot including WE32β DNA, liver DNA from non-transgenic mice and DNA from transgenic and non-transgenic progeny hybridised to DRβ cDNA.

Figure 4(a) shows an autoradiograph of a Northern Blot including RNA from liver, spleen, lung and kidney of DRα transgenic 1530 and RNA from BSH cell line hybridised to DR-α cDNA.

Figure 4(b) shows an autoradiograph of a Northern Blot including liver, lung, spleen and brain RNA from DRβ transgenic 1499; liver, lung and heart RNA from DRβ transgenic 1502, plus RNA from BSM cell line hybridised to DRβ oligonucleotide.

Figure 5 shows the FACS profiles of blood cells from a mouse containing HLA-DR4Dw4 and transgenes stained with antibodies directed towards HLA-DR (MAb L243) and mouse MHC Class II protein (MAb 10-3.6.2 ,

Figure 6 shows two FACS histograms of spleen cells stained with MAb L234 plus phycoerythrin-conjugated goat anti-maouse IgG (Fc)FITC followed by FITC-conjugated rat-anti-IA (Serotec clone) or FITC-conjugated rat-anti-Thy-1 (Serotec clone) MAbs. The expressing I-A molecules also express HLA-DR4DW4 molecules (population B of Figure 6a) , while the HL -DR4Dw4-negative cells also do not express I-A molecules. Furthermore, mouse T cells (CD3-positive spleen cells) which do not express mouse MHC Class II Molecules, also do not express HLA-DR4Dw4 molecules (Population B of Figure 6b) .

Figure 7 shows the responses of LN T cells obtained from 10 individual mice (5 HLA-DR4Dw4-transgenic mice; 5 non-transgenic littermates) immunised with FHA307-320. It can be seen that addition of FHA307-320 to primed LN T cells from HLA-DR4Dw4-transgenic mice produces a dose-dependent proliferative response, whereas primed LN T cells from non-transgenic littermates do not proliferate when FHA307-320 is added.

Figure 8 shows the effect of anti-HLA-DR monoclonal antibodies on recognition of FHA307-320 by primed T cells from HLA-DR4Dw4-transgenic mice. Spleen cells and primed LN T cells were prepared as described for Example 4, and a stimulatory concentration of FHA307-320 (10 μM) was added. Monoclonal antibody L243 which binds HLA-DR4Dw4 molecules, but not mouse MHC Class II molecules were also added, and the cultures incubated, harvested and counted as described. It can be seen that addition of MAb L243 completely inhibits prolifereation of primed LN T cells from transgenic mice. In contrast, a control monoclonal antibody (MKD6) which would not react with HLA-DR4Dw4 molecules has no inhibitory activity, showing that the effect is specific to HLA-DR4Dw4.

Figure 9 shows that HLA-DR4Dw4 transgenic mice produce recall delayed type hypersensitity responses to the HLA-DR4Dw4-binding peptide FHA307-320. It can be seen that administration of an HLA-DR4Dw4-binding antagonist peptide to HLA-DR4Dw4 transgenic mice results in measurable inhibition of the delayed type hypersensitivity response. In contrast, an antagonist of mouse MHC Class II molecules

(SNase61-80) demonstrated no inhibitory activity. These results confirm that the rsponse is HLA-DR4Dw4 restricted, and that this method can be used to detect HLA-DR4 antagonist peptides.

GENERATION OF TRANSGENIC MICE EXPRESSING HUMAN HLA DR4Dw4

Generation of a transgenic mouse which expreses human HLA DR4Dw4 requires the addition of two human genes, a nonpolymorphic a gene and a polymorphic β gene.

Source of the human DRα gene

A cosraid clone, T9C, containing the vhole of the human DR4 α gene vas provided by T. Spies (Spies et al., Proc. Natl. Acad. Sci. USA 82:5165 1985). A 22Kb BamHI to Sail fragment containing the 7Kb gene and 5' and 3' flanking sequence vas generated from T9C (Figure 1) and isolated from agarose gels for icroinjection (cf. Hogan et al., Manipulating the mouse embryo - A laboratory manual, Cold Spring Harbour Laboratory Press, 1986) at a concentration of 2ng/μl.

Construction of a human DR4Dv4 β cDNA construct

Human DR4Dv4 β chain cDNA vas amplified from a human β-lymphoblastoid cell line (BSH - a transformed Epstein-Barr virus obtained from R. Bolhuis, ETNO, Rijsvijk, NL). Total RNA vas prepared from BSH cells and Poly A+ RNA selected by passing 50μg of total RNA down an oligonucleotide dT cellulose column. 5μg of polyA÷ RNA vas used as a template to generate double stranded cDNA. The cDNA vas then used in a PCR reaction with oligonucleotides 238 (GTCGAATTCTCCAGCATGGTGTGTCTGAAG) and

239(CATGTCGACTTCAGCTCAGGAATCCTGTTG) to generate an 820 bp product. This vas subsequently cloned betveen the EcoRI and Sail sites of the bacteriophage cloning vectors H13mpl8 and H13mpl9 (Boehringer Mannheim) and sequenced. Double standed (RF) DNA vas prepared from the mpl8 clone, and the insert subcloned into eukaryotic expression vector pMSGCMVα [pHSG - Pharmacia, modified by (i) removal of the HHTV-LTR and replaced by a cytomegalovirus (CHV) immediate early promoter/enhancer (as disclosed by U. Schaffner, Cell, 41_, 521-530 and isolated on a 619 bp Thai fragment from pcm5029 - Gene, 45, 101-105) and (ii) replacement of gpt by the pSV2 neo marker provided by P. Berg (J. of Mol. and App. Genetics, 1982, 1_, 327-341)]. The folloving linker vas then ligated at the EcoRI site 5' to the cDNA AATTAGATCT

TCTAGATTAA thus generating a Bglll site. The construct comprising the DRβ cDNA, SV40 splice site and poly A recognition sequence vas purified on a Bglll to BamHI fragment and susequently subcloned into the BamHI site of plasmid pϋE32 provided by D. Hathis, Stanford University (Dorn et al, Proc. Natl. Acad. Sci.USA, 1987, 84, 6249-6253). pWE32 contains the mouse IEα gene modified, for clonine

purposes, by removal of a BamHI site from the first intron and introduction of a BamHI site in the untranslated region of the first exon. The IEα regulatory sequences and gene now including the DRβ cDNA in the first exon (Figure 2) were isolated by Bglll digestion. The 9.4 Kb fragment was then gel purified and resuspended at 2ng/μl in TE (0.5 mM Tris - pH 7.5, 0.1 mM EDTA) for microinjection.

Generation of Transgenic mice

All manipulations were performed as described by Hogan et al. , Manipulating the mouse embryo - a laboratory manual, Cold Spring Harbour Laboratory Press, 1986.

CBA-Ca x C57B1/6 (Fl) female mice (ZENECA - Alderley Park) were superovulated and mated with Fl males. Fertilzed oocytes were microinjected with either the 22Kb BamHI - Sail fragment containing the DRα gene, or the 9.4 Kb Bgll fragment containing the DRβ gene and incubated overnight. Viable embryos were transferred to the oviducts of pseudopregnant Fl females (mated with vasectomised males) and allowed to develop to term.

Analysis of Transgenic mice

The resulting progeny were analysed to determine whether the transgene was present. Approximately 0.5cm of tail was obtained from each animal, sliced with a scalpel blade, and digested overnight at 55°C in 0.5 ml lysis buffer (100 m Tris.HCl pH 8.5, 5mM EDTA, 0.2% SDS, 200 mM NaCl, 100 μg Prσteinase K/ml. The solution was extracted with phenol/chloroform and DNA was precipatated. DNA was precipitated with an equal volume of isopropanol, washed in 70% ethanol and resuspended in 500 μl TE (0.1M Tris pH8, 0.01M EDTA) at 55°C. PCR analysis using DRα or DRβ specific primers was used to identify transgenic animals. Approximately 100 ng of tail DNA were used in PCR reactions with Taq polymerase (Cetus - Amplitaq) using conditions described by the manufacturer (Perkin-Elmer Cetus) on a Techne thermal cycling machine. The amplimers used were:

DRα 5'AAGAACATGTGATCATCCAGGCC (EXON), 5' CCCCAGTGCTTGAGAAGAGGCTCA (EXON 2) at 92°C for 2 min, 65°C for 2 min and 72°C for 2 min for 35 cycles giving a 1018 bp product.

DRβ 5'GAGATGCATCCAGCAATAAGGAG ( -500 bp in 5' flanking sequence of mouse IEa) , 5' TGCACTGTGAAGCTCTCACCAAC (374 bp from translation start site of human DR4Dw4 β cDNA) at 92°C for 2 min, 64°C for 2 mins, 72 ^e C for 2 min for 35 cycles giving an 886 bp product.

Transgenic animals were confirmed by Southern analysis of

EcoRI digested tail DNA. DNA (10 μG) was transferred to Hybond membrane (Amersham) using 20 x SSC (3M NaCl, 0.3 M tri-sodium citrate - Na C H 0 .2H O) , UV fixed, prehybridised and hybridised 3 6 5 7 2 according to the manufacturers recommendations with a P labelled probe of either the DRα or DRβ cDNA using standard protocols. The results are shown in Figures 3 and 4.

Founder animals

Six 'founder' DRα animals were identified, two transmitted the transgene with high frequency (DRα08 and DRαlO) a third animal (DRαl2) produced transgenic offspring although the litter size was small. Of the other three 'founders' two did not transmit (DRαl9 and DRα93) and one gave rise to progeny which appeared to contain rearranged sequences (DRα21) . DR 08 was chosen to establish a transgenic line and further reference to DRα relates to DRα08. Microinjection of the DRβ construct resulted in 3 'founders', β325, β380 and β356. β325 and β380 did not transmit the transgene to their progeny. A line was therefore established from mouse/human hydrid β356 and further reference to DRβ relates to this line.

Northern Analysis

Total RNA was extracted from mouse tissues using the method of Chirgwin et. al., Biochemistry 1979, 18 No. 24: 5294. 10 μg of RNA were loaded onto formaldehyde gels and electrophoresis carried out at

100 volts for 4 hours. RNA was blotted onto Hybond membrane

(Amersham) using 20 x SSC and UV fixed. Prehydridisation and hybridisation were carried out in either 0.5M sodium phosphate pH 7.2,

7%SDS, lmM EDTA (for DRα cDNA probe) or 0.144 M Na HPO , 0.056M

NaH PO , lmM EDTA, 1% BSA, 7.2% SDS, 15% deionised formamide (for DRβ 2 4

oligonucleotide probe probe 5ΑCTGTGAAGCTCTCACCAACCCCG) . Highest levels of DRα and DRβ expression were found in spleen and lung.

Fluorescent in situ hybridisation (FISH)

Integration of the transgenes in the mouse genome was observed by fluorescent in situ hybridisation (FISH) . Either cosmid T9C DNA of pWE32 was biotinolated (BRL- Bionick Kit, according to the manufacturers instructions) . Metaphse chromosome spreads from fibroblasts of transgenic animals were treated with RNase (100 μg/ml in 2xSSC) for 60 minites at 37°C, washed and dehydrated through an alcohol gradient, denatured in 70% formamide, 2xSSC at 75°C for 3-5 minutes and dehydrated (denaturation may also be performed at 90°C for 5 minutes) . 100 ng of labelled DNA was dissolved in 25 μl of hydridisation mix (10% dextran sulphate, 50% formamide, 2xSSC, 0.1 mM EDTA, 0.05mM Tris pH7.5, 100 μg/ml denatured salmon sperm DNA) and denatured at 80°C for 10 minutes. Denatured probe was hybridised to the slides under a coverslip at 37°C for 48 hours. Slides were washed 3x5 minutes in 50% formamide/2xSSC pH7.0, at 42°C and 3x5 minutes in 2xSSC at 42°C and blocked for 60 minutes with 5% non-fat dried milk (Marvel - Premier Brands UK Ltd) in 4xSSC, 0.05% Triton X 100. Detection was by sequential incubation in fluoresicein isothiocyanate (FITC - Vector Laboratories) - avidin (DCS - D cell sorter grade) followed biotinylated anit-avidin (Vector Laboratories) , both at 5μg/ml in 4xSSC pH7.0, 0.05% Triton-x 100, 3%BSA, at 37°C for 20 minutes with 3x3 minute washed in 4XSSC/0.05% TRigton between each layer. After 304 layers of avidin, slides were washed in phosphate buffered saline (lxPBS -0.137 mM NaCL, 2.7mM KC1, 8.1 mM Na HPO .2H O,

1.5 mM KH PO - pH 7.4) counter stained with propidium iodide and mounted in 90% glycerol 20 mM Tris pH 8.0, 2.3%

1,4-diazobicyclo- [2.2.2]octane (DABCO] . The DRβ transgene was seen to have integrated at a single site in one of the mouse chromosomes. DRα however, was detected in two different mouse chromosomes, one proximal to the centromere and one distal. DRα transgenics were crossed with CBA's to segregate the two integration events into separate lines. Each was tested by fluorescence antibody cell sorter (FACS) analysis. The distral integration site expressed at a higher level than the proximal integration site.

-1-+-

or-B crosses

DRo.08 transgenic animals were mated .with DRE356 animals to generate progeny carrying both transgenes.

Example 2

DETECTION OF HLA-DR4Dw4 PROTEINS EXPRESSED BY BLOOD CELLS FROM HLA-DR4Dw4 TRANSGENIC MICE.

Blood from an incision in the tail was drawn into a lOOμl heparinised tube (BDH), and immediately blown out into 1ml of ice cold PBS. The sample was then centrifuged (2000rpm for 5 minutes), the pellet resuspended in 200μl of ice cold PBS containing 20Z normal rabbit serum plus 0.2Z sodium azide, and 50μl portions pipetted into LP4 tubes (Luckham). After 15 minutes incubation in a refrigerator, 50μl of pre-titrated monoclonal antibody-containing supernatant diluted (typically 1:5) in PBS + 20% normal rabbit serum + 0.2Z sodium azide was added to the cell suspension. After a further 20-45 minute incubation in the cold, the cells were washed by addition of 450μl of cold PBS + 2% foetal calf serum (FCS) + 0.21 sodium azide followed by centrifugation (2000rpm for 5 minutes) at 4°C. The supernatant was then decanted by suction, and the pellet resuspended by vortexing for 2-3 seconds. 50μl of FITC-conjugated Goat anti-mouse IgG F(c) antibody (Serotec) (preabsorbed for 30 minutes with 502 mouse spleen cell suspension) diluted 1:75 in PBS + 20% rabbit serum + 0.21 sodium azide was added, and the pellet resuspended by vortexing. After 20 minutes incubation in the cold, the cells are washed twice in PBS + 21 FCS + 0.21 sodium azide. Erythrocytes are removed by addition of FACS Lysis Buffer (Becton Dickinson - according to the manufacturers instructions), and the fluorescing cells enumerated on a FACS-SCAN flow cytoraeter (Becton Dickinson).

Figure 5 shows the FACS profiles of blood cells from a mouse containing HLA-DR4Dw4 and transgenes stained with antibodies directed towards HLA-DR (MAb L243 - ATCC HB55) and mouse MHC Class II protein (MAb 10-3.6.2 - ATCC TIB92). It can be seen that similar numbers of cells from HLA-DR4Dw4 transgenic mice are stained with antibodies to mouse and human MHC Class II proteins, respectively.

Example 3

DETECTION OF HLA-DR4Dw4 PROTEINS EXPRESSED BY SPLEEN CELLS FROM TRANSGENIC MICE.

Spleens were excised from transgenic mice, placed in a 50mm diameter sterile plastic petri dish containing 3 ml of cold sterile PBS, and cut into small (approx 3mm) pieces. The pieces were squashed using the plunger of a 2ml plastic syringe (Asik) to release the spleen cells, and a further 7ml of ice cold PBS were added using a 10ml pipette. The suspension was pipetted up and down vigorously ten times, and the mixture filtered through a Cell Strainer (Becton Dickinson) into a 25ml sterile glass beaker to remove debris. The cell suspension was poured into a 15ml sterile plastic conical tube (Falcon), and centrifuged at room temperature (1400rpm for 5 mins). The cell pellet was then resuspended in 3ml PBS, and 2ml of Lympholyte M (Jackson Labs) layered underneath. The cells were centrifuged at 2000rpm for 15 mins at room temperature, and the cells at the interface representing the mononuclear population were then collected and diluted into 10ml ice cold PBS in a 15ml conical tube. The cells were washed twice in PBS before being counted.

For staining, cells were resuspended (10-20x10 /ml) in PBS + 20% rabbit serum + 0.2% sodium azide, and processed as disclosed in Example 2, except that the treatment with FACS Lysis Buffer was omitted.

From the profiles shown in Figure 6, it can be seen that similar numbers of spleen mononuclear cells express HLA-DR molecules and mouse I-A molecules. To further characterise the cells expressing HLA-DR molecules, spleen cells were stained with MAb L243 plus phycoerythrin-conjugated goat anti-mouse IgG (Fc) followed by FITC-conjugated rat-anti-I-A (Serotec clone MCA 46F) or FITC-conjugated rat-anti-CD3 (Serotec MCA 500F) MAbs. The two colour FACS histograms show that the spleen cells expressing I-A molecules also express HLA-DR4Dw4 molecules (population B of Figure 6a), while the HLA-DR4Dw4-negative cells also do not express I-A molecules. Furthermore, mouse T cells (CD3-positive spleen cells) which do not

express mouse MHC Class II molecules, also do not express HLA-DR4Dw4 molecules (Population A of Figure 6b).

Thus, the pattern of expression of the HLA-DR4Dw4 transgenes in lymphoid cells of the spleen is similar to that of endogenous MHC Class II products.

Example 4

IN VITRO RESPONSES OF T CELLS FROM PEPTIDE-IHMUNISED TRANSGENIC MICE.

In order to examine the function of the HLA-DR4Dw4 transgene products, mice were immunised with lOnmol of peptide FHA307-320 (Sequence : Pro-Lys-Tyr-Val-Lys-Gln-Asn-Thr-Leu-Lys-Leu-Ala-Thr-Gly) mixed 1:1 in Complete Freund's Adjuvant (Difco Labs). 0.1ml of the mixture was administered sub-cutaneously in a single site in the flank using a 1ml plastic syringe and 23 gauge needle. Mice were sacrificed 7-10 days following immunisation with peptide, and the enlarged draining ingunal lymph node, and spleen were dissected from each mouse.

Lymph node (LN) cell suspensions were prepared by passing through a cell strainer as disclosed for spleen cells in Example 3. The final suspension was suspended in 1ml of warm tissue culture medium (RPMI-1640, Gibco) supplemented with 5% heat inactivated FCS + 2mM glutamine + lOOU/ml penicillin/lOOg/ml streptomycin + 50μM 2-mercaptoethanol (hereinafter referred to as "culture medium"), and the suspension was then applied to a nylon wool column. The nylon wool column was prepared by adding 0.6g nylon wool to the barrel of a 10ml plastic syringe, and the column autoclaved. The sterile column is washed through with 20ml of warm culture medium approx 1 hour before use, and incubated at 37°C until cells are added. The LN cell suspension was added to the top of the column dropwise, and incubated for 45 mins in a CO- incubator at 37°C. Non-adherent cells were then eluted by adding 20ml of warm culture medium dropwise, and collecting the flow-through into a sterile 30ml plastic universal container. The resulting cells ("LN T cells") were pelleted, resuspended in 1ml of

warm culture medium, counted, and further adjusted to a density of 4xl0 ⁶ /ml.

Spleen mononuclear cell suspensions were prepared as described in Example 3. In some experiments, spleen cells were treated with anti-Thy-1 plus complement to deplete T cells. Specifically, spleen mononuclear cells were suspended (5x10 /ml) in culture medium containing a 1:10,000 dilution of anti-Thy-1 MAb (Serotec) for 30 minutes, washed once in cold PBS, and resuspended (5x10 /ml) in Low-tox rabbit complement (Serotec) for 45 minutes at 37°C. The cells were washed twice with warm culture medium, resuspended in 2ml warm culture medium, counted, and adjusted to a density of 4x10 /ml.

To assess the degree of T cell responses, LN T cells (50μl/well) and spleen cells (50μl/well) are added to lOOμl of medium containing peptides (eg. FHA307-320) at the indicated concentrations in 96-well microtitre plates (Nunc), and incubated for 4 days at 37°C in a humidified atmosphere of 5% CO. in air. Responding T cells are quantified by addingAlCi/ml 3H-thymidine (Amersham) for 6 hours, and the plates harvested and counted on a Beta-plate counter (LKB). Results are expressed as the mean (+SEM) of triplicate cultures.

Peptide FHA307-320 is known to bind to HLA-DR4Dw4 and induce proliferation in primed human T cells of appropriate specificity. Figure 7 shows the responses of LN T cells obtained from 10 individual mice (5 HLA-DR4Dw4-transgenic mice; 5 non-transgenic littermates) immunised with FHA307-320. It can be seen that addition of FHA307-320 to primed LN T cells from HLA-DR4Dw4-transgenic mice produces a dose-dependent proliferative response, whereas primed LN T cells from non-transgenic littermates do not proliferate when FHA307-320 is added.

Example 5

EFFECT OF ANTI-HLA-DR MONOCLONAL ANTIBODIES ON RECOGNITION OF FHA307-320 BY PRIMED T CELLS FROM HLA-DR4DU4-TRANSGENIC MICE.

To confirm that the proliferation of primed LN T cells from transgenic mice observed on addition of FHA307-320 is stimulated by HLA-DR4Dw4, monoclonal antibodies able to block the interaction of T cells with HLA-DR4Dw4 were added. Spleen cells and primed LN T cells were prepared as described for Example 4, and a stimulatory concentration of FHA307-320(ιθμM) ^w as added. Honoclonal antibody L243 which binds HLA-DR4Dw4 molecules, but not mouse MHC Class II molecules vas also added, and the cultures incubated, harvested and counted as described above. The results are shown in Figure 8, from which it can be seen that addition of MAb L243 completely inhibits proliferation of primed LN T cells from transgenic mice. In contrast, a control monoclonal antibody (MKD6 - ATCC HB3) which would not react with HLA-DR4Dw4 molecules has no inhibitory activity, showing that the effect is specific to HLA-DR4Dw4.

Example 6

EFFECT OF HLA-DR4Dff4-SPECIFIC BLOCKING PEPTIDES ON RECOGNITION OF FHA307-320 BY PRIMED T CELLS FROM HLA-DR4DV4-TRANSGENIC MICE.

A further proof that the proliferative response of primed LN T cells to FHA307-320 is due to the presence of HLA-DR4Dw4 in transgenic mice is the demonstration that the response is blocked by peptides which bind to HLA-DR4Dw4 molecules and are able to displace the stimulating FHA307-320 peptide. Primed LN T cells were prepared as described for Example 4.

In these experiments, spleen cells were prepared as for Example 4, and "fixed" with gluteraldehyde prior to addition of peptides. Specifically, anti-Thy-1 plus complement-treated spleen mononuclear cells (as prepared in Example 4 above) were washed in PBS, and resuspended at 5x10 cells/ml for 30 seconds in 0.5% gluteraldehyde. An equal volume of 0.4 mM lysine in PBS is added for 3 minutes, and the cells pelleted and washed twice in culture medium. The fixed cells were resuspended to 4x10 /ml in serum-free tissue culture medium (RPMI-1640 medium, Gibco), and lOOμl added to lOOμl of

serum-free RPMI1640 medium containing lμM FHA307-320 and blocking peptides at concentrations between 0.1-100μM in 96-well microtitre culture plates. After 2 hours incubation, the cells were pelleted, the medium removed by suction, and 200μl of culture medium added. The cells were again pelleted, the medium removed, and a further 200μl of medium added. After pelleting the cells and removing the medium once more, 2x10 primed LN T cells were added in 200μl of warm culture medium. Cultures are then treated as described in Example 4.

A number of HLA-DR4Dw4-binding peptides are able to inhibit the presentation of FHA307-320 to primed LN T cells from transgenic mice. In contrast, peptides able to compete for binding to mouse MHC Class II molecules (I-Ak ; I-Ek) do not prevent activation of primed LN T cells by FHA307-320. This will confirm that presentation of FHA307-320 is mediated solely by binding to HLA-DR4Dw4 in transgenic mice.

Example 6

INHIBITION OF IN VIVO RESPONSES IN HLA-DR4DW4-TRANSGENIC MICE BY HLA-DR-ANTAGONIST PEPTIDES

The ability of antagonist peptides to inhibit binding of stimulatory peptides to HLA-DR4Dw4, and subsequent activation of T cells in vivo can be tested in HLA-DR4Dw4-transgenic mice essentially according to the method of Adorini et al (1988; Nature 334:623). Transgenic mice are immunised with a mixture of 10 nmoles of FHA307-320 mixed with an effective amount (eg 1-500 nmoles) of the antagonist peptide in 1001 of a 1:1 mixture of Freund's Complete Adjuvant and PBS. After 7-14 days, mice are killed, and spleen and LN T cell suspensions made and assayed for reactivity to 1M FHA307-320 as described in Example 4. HLA-DR4Dw4 antagonist peptides are able to inhibit the priming of LN T cells in transgenic mice, demonstrating that they are able to antagonise the function of HLA-DR4Dw4 in vivo.

INHIBITION OF IN VIVO DELAYED TYPE HYPERSENSITITY RESPONSE IN HLA-DR4DW4 TRANGENIC MICE BY HAL-DR-ANTAGONISTS PEPTIDES

The ability of antagonist peptides and non-peptides to inhibit binding of stimulatory peptides to HLA-DR4Dw4, and subsequent activation of T cells in HLA-DR4Dw4 transgenic mice can also be tested by the following method. HLA-DR transgenic mice are immunised with a mixture of an effective amout of FHA307-320 (100 ng in this example) in 109 mictrolitres of a 1:1 mixture of Freund's Complete adjuvant and PBS. After 7-14 days, approximately 30 microlitrues of a 100 micromolar solution of FHA307-320 is injected sub-plantar into one hind foofpad, and the resultant immune response determined by measuring foot swelling wit hmicronic calipers. HLA-DR4Dw4-binding peptides administered either by admixing with the sub-plantar challenge injection, or given by other standard route, or by slow-release depots such as osmotic mini-pumps can be shown to inhibit the swelling induced by FHA307-320, showing that they are able to antagonise the function of HLA-DR4Dw4 in vivo (Figure 8) .

SEQUENCE LISTING

(1) GENERAL INFORMATION

(i) APPLICANT: ZENECA Limited

(ii) TITLE OP INVENTION: POLYPEPTIDES

(iii) NUMBER OF SEQUENCES: 10

(iv) CORRESPONDENCE ADDRESS:

(A) ADDRESSEE: Group Patents Services Department

(B) STREET: Bessemer Road

(C) CITY: Welwyn Garden City

(D) STATE: Hertfordshire

(E) COUNTRY: United Kingdom

(F) ZIP: GB-AL7 1HD

(v) COMPUTER READABLE FORM:

(A) MEDIUM TYPE: DISKETTE, 3.5 INCH, 1.44 mB storage

(B) COMPUTER: IBM PS/2

(C) OPERATING SYSTEM: PC-DOS 3.20

(D) SOFTWARE: ASCII from WPS-PLUS

(vi) CURRENT APPLICATION DATA:

(A) APPLICATION NUMBER:

(B) FILING DATE:

(vii) PRIOR APPLICATION DATA:

(A) APPLICATION NO. GB 9315303.9

(B) FILING DATE: 23-Jul-1993

(2) INFORMATION FOR SEQ ID NO:l:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 30 base pairs

(B) TYPE: nucleic acid

(C) STRANDEDNESS: single

(D) TOPOLOGY: linear

(xi) SEQUENCE DESCRIPTION: SEQ ID NO:l:

GTCGAATTCT CCAGCATGGT GTGTCTGAAG 30

(2) INFORMATION FOR SEQ ID NO:2:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 30 base pairs

(B) TYPE: nucleic acid

(C) STRANDEDNESS: single

(D) TOPOLOGY: linear

(xi) SEQUENCE DESCRIPTION: SEQ ID NO:2:

CATGTCGACT TCAGCTCAGG AATCCTGTTG 30

(2) INFORMATION FOR SEQ ID NO:3:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 10 base pairs

(B) TYPE: nucleic acid

(C) STRANDEDNESS: single

(D) TOPOLOGY: linear

(xi) SEQUENCE DESCRIPTION: SEQ ID NO:3:

AATTAGATCT 10

(2 ) INFORMATION FOR SEQ ID NO : 4 :

( i) SEQUENCE CHARACTERISTICS :

(A) LENGTH : 10 base pairs

(B) TYPE : nucleic acid

(C) STRANDEDNESS: single

(D) TOPOLOGY: linear

(xi) SEQUENCE DESCRIPTION: SEQ ID NO:4:

TCTAGATTAA 10

(2) INFORMATION FOR SEQ ID NO:5:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 23 base pairs

(B) TYPE: nucleic acid

(C) STRANDEDNESS: single

(D) TOPOLOGY: linear

(xi) SEQUENCE DESCRIPTION: SEQ ID NO:5:

AAGAACATGT GATCATCCAG GCC 23

(2) INFORMATION FOR SEQ ID NO:6:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 24 base pairs

(B) TYPE: nucleic acid

(C) STRANDEDNESS: single

(D) TOPOLOGY: linear

(xi) SEQUENCE DESCRIPTION: SEQ ID NO:6:

CCCCAGTGCT TGAGAAGAGG CTCA 24

(2) INFORMATION FOR SEQ ID NO:7:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 23 base pairs

(B) TYPE: nucleic acid

(C) STRANDEDNESS: single

(D) TOPOLOGY: linear

(xi) SEQUENCE DESCRIPTION: SEQ ID NO:7:

GAGATGCATC CAGCAATAAG GAG 23

(2) INFORMATION FOR SEQ ID NO:8:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 23 base pairs

(B) TYPE: nucleic acid

(C) STRANDEDNESS: single

(D) TOPOLOGY: linear

(xi) SEQUENCE DESCRIPTION: SEQ ID NO:8:

TGCACTGTGA AGCTCTCACC AAC 23

(2) INFORMATION FOR SEQ ID NO:9:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 24 base pairs

(B) TYPE: nucleic acid

(C) STRANDEDNESS: single

(D) TOPOLOGY: linear

(xi) SEQUENCE DESCRIPTION: SEQ ID NO:9:

ACTGTGAAGC TCTCACCAAC CCCG

(2) INFORMATION FOR SEQ ID NO:10:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 14 amino acids

(B) TYPE: amino acid

(C) STRANDEDNESS: single

(D) TOPOLOGY: linear

(xi) SEQUENCE DESCRIPTION: SEQ ID NO:10:

Pro Lys Tyr Val Lys Gin Asn Thr Leu Lys Leu Ala Thr Gly