CA2217173C - Isolation of a cdna clone encoding a subunit of the alternaria alternata major allergen alt a 1, and production of recombinant protein thereof - Google Patents

Abstract

Alternaria alternata is recognized as an important source of fungal aeroallergens. The major allergen of this mold, Alt a 1 (previously known as Alt a I, Alt a-29) is a dimer of disulfide-linked subunits that migrate in SDS-PAGE under reducing conditions at apparent Mrs of 14,500 and 16,000. IgE antibodies to this protein are present in the sera of >90% of A. alternata-sensitive individuals. This invention comprises an isolated cDNA sequence from A. alternata encoding an Alt a 1 subunit, from which recombinant allergen has been produced.

Description

BACKGROUND OF THE INVENTION

FIELD OF THE INVENTION

This invention relates to an allergen of Alternaria alternata, and specifically to a DNA sequence encoding an allergenic protein of Alternaria alternata.
DISCUSSION OF THE PRIOR ART

Allergies to airborne antigens are estimated to afflict 10-20% of Europeans and North Americans (Gergen et al, J Allergy Clin Immunol 80: 669-679, 1987). Aeroallergens may consist of plant pollen- or fungal spore-derived protein, or airborne detritus from their biological sources, i.e. cat dander or dust mites.

The current accepted therapeutic treatment for allergies is to desensitize the patient by injection with extracts prepared from the specific allergen to which the patient is reactive. Ideally, hyposensitization reduces the allergic response through production of IgG antibodies capable of competing for allergen with effector cell-bound IgE, as well as by induction of suppressor T cell populations (Bousquet et al, in Progress in Allergy and Clinical Immunology, Pichler et al, eds.; Hogrefe and Huber, Toronto, pp.377-382, 1988).

Initially, allergen extracts used for desensitization were relatively crude mixtures of protein and carbohydrate extracted from the source organism or material by physicochemical methods. More recently, purification techniques have been applied to some allergens, yielding partially purified products with improved efficacy (Solvaggio et al, J Allergy Clin Immunol 92: 217-222, 1993; Malling et al, Allergy 41: 507-519, 1986). However, commercially prepared allergen extracts are often of inconsistent quality.
Impurities or irritants in the extracts are potentially capable of inducing specific IgE production, resulting in risk of anaphylaxis.

By using recombinant DNA techniques to isolate allergen cDNA clones, allergens may be produced in organisms engineered for large scale protein expression, and conveniently purified with greater fidelity than in the original source. Purified recombinant allergens are of utility in the immunodiagnosis of allergy, as coating antigens for solid phase assays of atopic sera from allergic individuals, i.e. enzyme-linked immunosorbent (ELISA) or radioallergosorbent (RAST) assays. Knowledge of the DNA
sequences of such allergens may be applied to synthesize short peptide fragments, thus providing a means of localizing IgE-binding or T cell stimulatory epitopes to specific regions of the protein. Desensitization of allergic patients using short peptides comprising such T or B cell epitopes may also be of use as immunotherapeutic reagents administered as part of a desensitization protocol. Numerous recent publications have described the cloning of allergens from plant or animal sources, usually by immunoscreening of cDNA libraries with human atopic IgE (Valenta et al, Science 253: 557-560, 1991;
Sidoli et al, J Biol Chem 268: 21819-21825, 1993; Shen et al, Clin Exp Allergy 23: 934-940, 1993; Sone et al, Biochem Biophys Res Commun 199: 619-625, 1994), and at least two

2 patents have also been filed, one for a ryegrass pollen allergen (Theerakulpisut et al, PCT Application WO 92/03550, filed March 5, 1992) and one for a ribosomal P2 protein allergen from Cladosporium herbarum (Zhang and Vijay, U.S.

Patent No. 5,556,953 which issued on September 17, 1996).
GENERAL DESCRIPTION OF THE INVENTION

Progress in the characterization and purification of fungal aeroallergens is complicated by considerable variability with respect to allergen protein profile between strains, as well as difficulties in the recovery of specific spore populations (Vijay et al, Int Arch Allergy Appl Immunol 74: 256-261, 1984; Steringer et al, Int Arch Allergy Appl Immunol 84: 190-197, 1987; Solvaggio et al, J Allergy Clin Immunol 92: 217-222, 1993). Because of this variability between isolates of a given species, fewer major allergens (allergens to which IgE is produced in greater than 50% of individuals allergic to a given species) have been characterized in fungi.

Examination of sera from individuals allergic to aeroallergens has determined that Alternaria alternata (A.
alternata) represents an important source of fungal aeroallergens (Aas et al, Allergy 35:443-451, 1980; Yee and Bahna, J Allergy Clin Immunol 77: 200, 1986). Initial attempts to characterize the major allergenic components of A.

alternata extracts yielded protein fractions whose components bound IgE in a majority of A. alternata-allergic patients (Yunginger et al, J Allergy Clin Immunol 66: 138-147, 1980), but were of heterogeneous composition, consisting of several

3 IgE-binding proteins (Vijay et al, in Epitopes of Atopic Allergens. Proceedings of Workshop, XIVth Cong Eur Acad Allergol Clin Immunol, Berlin, Sept. 1989, Sehon et al, eds.;
Brussels, UCB Institute of Allergy,pp 14-17, 1990). Recent immunoblot analyses of A. alternata extracts probed with pooled atopic sera have enabled more precise characterization of individual allergens within fractionated extracts. Of these, greater than 90% of atopic sera containing A.
alternata-specific IgE reacted positively against a major allergen purified by several groups and originally designated Alt a-29 (Curran et al, Int Arch Allergy Immunol 102: 267-275, 1993), Alt I (Matthiesen et al, Allergologie 12: 21, 1989;
Matthiesen et al, J Allergy Clin Immunol 89: 241, 1992) or Alt a Bd 29K (Deards and Montague, Mol Immunol 28: 409-415, 1991).

Further purification and immunoblot analyses demonstrated that this allergen, as isolated by the present inventors, is a disulfide-linked dimer of subunits with apparent relative molecular masses of 14,500 and 16,000 (Curran et al, Int Arch Allergy Immunol 102: 267-275, 1993) and nearly identical N-terminal sequences (SEQ ID NO:1, SEQ ID NO:2) that are also 85% homologous to the corresponding N-terminal residues of the isolate of Matthiesen et al. (J Allergy Clin Immunol 89: 241, 1992; Vijay et al, J. Allergy Clin Immunol 91: 826-828, 1993).
Based on the striking N-terminal homology between these two protein isolates, these allergens are now considered to represent the same protein, and are collectively termed Alt a 1 according to the recently revised system of allergen

4 nomenclature (IUIS/WHO Allergen Nomenclature Subcommittee, Bull WHO 72: 797-806, 1995).

An object of the present invention is to provide the nucleic acid sequence that encodes a subunit of the allergen Alt a 1.

An object of the present invention is to-characterize the cloned DNA sequence of the major allergen, Alt a 1, isolated from Alternaria alternata.

Another object of the invention is to provide a means of producing the allergen using recombinant techniques.
Thus according to one aspect, the invention relates to a nucleic acid sequence isolated from Alternaria alternata that encodes a subunit of the major allergen Alt a 1.

More specifically, the invention relates to a nucleic acid sequence encoding a subunit of the major allergen of Alternaria alternata Alt a 1 comprising the open reading frame of clone rb5l (ATCC 97754), from bases 64-534 (amino acids 1-157) of SEQ ID NO: 3:

Met Gln Phe Thr Thr Ile Ala Ser Leu Phe

5 10 Ala Ala Ala Gly Leu Ala Ala Ala Ala Pro Leu Glu Ser Arg Gln Asp Thr Ala Ser Cys Pro Val Thr Thr Glu Gly Asp Tyr Val Trp Lys Ile Ser Glu Phe Tyr Gly Arg Lys Pro Glu Gly Thr Tyr Tyr Asn Ser Leu Gly Phe Asn Ile Lys Ala Thr Asn Gly Gly Thr Leu Asp Phe Thr Cys Ser Ala Gln Ala Asp Lys Leu Glu Asp His Lys Trp Tyr Ser Cys Gly Glu Asn Ser Phe Met Asp Phe Ser Phe Asp Ser Asp Arg Ser Gly Leu Leu Leu Lys Gln Lys Val Ser Asp Asp Ile Thr Tyr Val Ala Thr Ala Thr Leu Pro Asn Tyr Cys Arg Ala Gly Gly Asn Gly Pro Lys Asp Phe Val Cys Gln Gly Val Ala Asp Ala Tyr Ile Thr Leu Val Thr Leu Pro Lys Ser Ser ***

BRIEF DESCRIPTION OF THE DRAWINGS

The invention will be described in greater detail with reference to the following drawings, wherein:

Figure 1 is a graphic reproduction of an agarose gel of PCR-amplified inserts from cDNA clones selected using rabbit anti-Alt a 1 IgG antibodies;

Figure 2 is a Southern hybridization of PCR-amplified inserts from cDNA clones selected using rabbit anti-Alt a 1 IgG antibodies;

6 Figure 3 is a hydrophobicity analysis of the peptide sequence translated from the open reading frame of isolated A.
alternata Alt a 1 cDNA (SEQ ID NO:3);

Figure 4 is a secondary structure analysis of the peptide sequence translated from the open reading frame of -isolated A. alternata Alt a 1 cDNA (SEQ ID NO:3);

Figures 5 is an agarose gel showing a size analysis of Alt a 1 mRNA;

Figure 6 is a Northern blot showing a size analysis of Alta a 1 mRNA;

Figure 7 is a comparison of non-reduced (top row) and reduced (bottom row) recombinant (lanes 1) and A.
alternata-derived Alt a 1 allergen (lanes 2) in Western blots stained with Coomassie blue R250 (D), or immunoreacted with rabbit anti-Alt a 1 (A), human atopic IgE (B), or non-atopic IgE (C);

Figures 8 and 9 illustrate the reactivity of serum IgEs from A. alternata-sensitive individuals to reduced natural or recombinant Alt a 1 allergen, respectively.

DESCRIPTION OF PREFERRED EMBODIMENT
Total RNA isolation from A. alternata A. alternata isolate 34-016 was cultured on liquid synthetic revised tobacco medium at 22 C as described by Vijay et al (Grana 28: 53-61, 1989). Mycelia were harvested on days 4, 6 and 10 by washing with ice-cold PBS, drying and storing at -70 C. The mycelial material was suspended at approximately 1 ml per gram of mycelial tissue in 4.7 M
guanidine isothiocyanate containing 1% (3- mercaptoethanol and

7 homogenized with a Polytron homogenizer for 2-3 intervals of two min each. After removal of debris by centrifugation at 10,000 x g for 10 minutes, the supernatant was extracted twice with phenol/chloroform (1:1, v/v), and precipitated with 0.2 M

potassium acetate (pH 4.0) and ethanol at -20 C overnight. .
RNA pelleted at 14,000 x g for 20 min at 4 C was resuspended in 0.2 - 0.3% of the original volume of guanidine isothiocyanate solution, layered on a cushion of 5.7 M CsC12 and subjected to overnight ultracentrifugation at 30,000 rpm with a Beckman SW50.1 rotor at 18 C. The RNA pellets were resuspended in RNase-free water, quantitated by spectrophotometric determination of A260, and equal amounts of the time point samples were pooled.

Establishment of an A. alternata- specific cDNA library in a.gtll A cDNA expression library was constructed by a commercial firm (Clontech Laboratories, CA) in Igtll vector with Poly A+
RNA isolated from total RNA by oligo dT cellulose chromatography (Aviv and Leder, Proc. Natl. Acad. Sci. USA 69:

1408-1412, 1972). First strand cDNA molecules were synthesized using oligodeoxythymidine (dT15) as primer and reverse transcriptase, followed by DNA polymerase I and RNAse H to synthesize the second strand DNA. Potential EcoRI sites of DNA molecules were methylated prior to ligation of EcoRI

adaptors. EcoRI digestion of the cDNA and vector was followed by dephosphorylation of the vector with calf intestinal alkaline phosphatase, and ligation of the cDNA into the vector

8 arms. Packaging extracts were used to reconstitute functional phage particles containing the ligated recombinant genomes.
Pxoduction of rabbit polyclonal, antiserum against A..
alteririata Alt a 1 allergen Partially purified SephadexTM G-100 fraction G3 of A.
alternata extract was prepared as previously described by Vijay et al (U.S. Patent No. 4,387,091, 1983). -To prepare hyperimmunE antiserum against proteins in this fraction, which includes Alt a 1, New Zealand albino male rabbits were immunized subcutaneously at aach of 5 dorsal s-ites with 0.2 ml of a homogenized mixture of 0.5 mg of antigen in 0.5 ml of saline and 0.5 ml of complete Freund's adjuvan't. Boosts were carried out at intervals of two weeks, and the initial test bleed was carried out at day 5Q, and at 42 day intiervals thereafter. The terminal bleeds werc-, carried out at day 226, and sera from rabbits with high titers were pooled. saturated ammonium sulfate solution was aclded dropwise to an equal volume of serum over a 1 h period on ice with slow stirring.
After stirring further for 30 min, the precipitate was pelleted at 7000 x g for 30 min at 4 C, washed with 1.75 M
ammonium sulfate, respun, and the pellet was resuspended in 10 mrt potassium phosphate buffer (pH 7.2-7.4). Dialysis was carried out over a three day period with 5 changes of 2 liters each, in cellulose dialysis membrarie with a molecular weight cutoff of 6000-8000. The dialyzed material was centrifuged at 7000 x g for 30 mzn at 4 C, and the-pxotEin content of the supernatant was measured at ODz... After adjustment of protein concentration, the hyperimmune serum (hereafter called rabbit

9 anti - Alt a 1, even though the serum is not monospecific) was aliquoted and stored frozen at -80 C.

Immunoscreening of the library using rabbit anti-Alt a 1 polyclonal antiserum Screening of the library was carried out using standard protocols (Huynh et al, in Glover D (ed): DNA
Cloning: A Practical Approach. Oxford, IRL Press, vol 1, pp 49-78, 1985). Phage from the unamplified library were plated at 10,000 pfu per 100 mm plate and grown for 3.5 h at 42 C.

Dry nitrocellulose membranes previously impregnated with 10 mM
isopropyl-D-D-galactopyranoside (IPTG) were overlaid on the plates and incubated at 37 C for 1.5 h. Duplicate membranes were also applied and incubated for an additional 2 h.

Membranes were blocked with 1% BSA in Tris-buffered saline (50 mM Tris-HC1 pH 7.5, 150 mM NaCl) overnight at 4 C.

Polyclonal rabbit anti-Alt a 1 was used in immunoscreening and immunoblots at 1:800 dilution in TBST (TBS
containing 0.05% Tween-20 ). This antiserum was preabsorbed by incubation with dry nitrocellulose membranes impregnated with lysates of IPTG-induced lysogens of Igtll in E. coli Y1090. Membranes were washed for 3 x 5 min each in TBST and incubated for 1 h with alkaline phosphatase (AP)-conjugated goat anti-rabbit IgG antibody (diluted 2000X with TBST) at room temperature. After washing as before, staining was carried out using nitroblue tetrazolium (NBT) and 5-bromo-4-chloro-3-indolyl phosphate (BCIP) in AP buffer (100 mM Tris-HC1 pH 9.5, 100 mM NaCl, 5 mM MgC12).

Polymerase Chain Reaction (PCR) and DNA Sequencing of Isolated Clones Immunoselected phage plaques were resuspended and purified by several rounds of sparse plating and immunoscreening until all plaques on a membrane stained positively. Purified phage plaque lysates (10 pl) heated to 99 C for 5 min were subjected to PCR using 10 pmol each of Igtll forward and reverse primers, 0.2 mM dNTPs, and 5 U of Taq polymerase in a 50 Nl volume of PCR buffer (10 mM Tris-HC1 pH 8.3, 50 mM KC1, 1.5 mM MgC12). The reactions were heated to 95 C for 5 min and subjected to 30 PCR cycles of 95 C for 36 sec, 51 C for 36 sec and 72 C for 90 sec. Aliquots of PCR
products were ligated into the plasmid pCR II (Invitrogen) and plasmid mini-prep DNAs from transformants of the construct were sequenced using the ALF laser fluorescence system and fluorescein-conjugated M13-40 and reverse primers (Pharmacia).
Southern and Northern Hybridizations To assess whether positive clones immunoselected using rabbit anti-Alt a 1 contained sequences homologous to the Alt a 1 N-terminus, digoxigenin (DIG)-labeled ICHA7 probe (SEQ ID NO:6) was prepared using the DIG Oligo Labeling kit (Boehringer-Mannheim). ICHA7 is an 8-fold degenerate oligonucleotide designed from the N-terminal peptide sequence of Alt a 1 (Curran et al, Int Arch Allergy immunol 102:
267-275, 1993). PCR-amplified inserts from immunoselected positives were run in 0.8% agarose gels, transferred to nylon membrane, and incubated with probe at 5 pmol/ml in hybridization solution (1% blocking agent, 5X SSC, 0.1%
N-lauroyl sarcosine, 0.02% SDS) for 3 h at 35 C, after a 1 h prehybridization. Washes were for 2 x 5 min with (5X SSC, 0.1% SDS) at room temperature, followed by 2 x 10 min washes using (2X SSC, 0.1% SDS) also at room temperature. Membrane5 were blocked for 30 min at room temperature with 1% Blocking Reagent (Boehringer-Mannheim) in TBS. AP-conjugated anti-DIG
Fab fragments (Boehringer Mannheim) were applied at 1:10,000 dilution in blocking solution, followed by washing for 2 x 15 min each with TBST containing 0.2% Tween-20, a brief incubation in AP buffer containing 50 mM MgCl 2 and an incubation with Lumi-Phos 530 chemiluminescent detection reagent (Boehringer-Mannheim) for 1 min. After incubating between transparency pages for 30 min at 37 C, blots were exposed to X-ray film for 5-15 min.

DIG-labeled rb51 probe used in Northern analyses of A. alternata total RNA was prepared by PCR in reactions as described above, except for the replacement of 0.2 mM TTP with 0.13 mM TTP and 0.07 mM digoxigenin-11-dUTP. Total RNA was extracted from mycelia of A. alternata isolate 34-016 grown in Revised Tobacco Medium for 3-20 days. Harvested mycelia were frozen and ground to powder in liquid nitrogen. RNA
extraction was carried out by the method of Chomczynski and Sacchi (Chomczynski and Sacchi, Anal Biochem 162: 156-159, 1987) using 1 ml Trizol (Life Technologies) per 30 mg of mycelia. RNAs were fractionated in 1.2% formaldehyde-agarose gels and transferred to nylon membrane. Labeled rb5l probe was hybridized to the blot overnight at 50 C. Washes were for 2 x 5 min with (2X SSC, 0.1% SDS) at room temperature and 2 x min using (0.1X SSC, 0.1% SDS) at 68 C. Further processing of these blots was carried out as described above.

Recombinant Alt a 1 Expression and Immunoblot Analyses 5 Plasmid pCR II containing the rb5l insert was =
cleaved with NheI and XbaI and subcloned directly into AvrII-cleaved and dephosphorylated Pichia pastoris expression vector pPIC9. The orientation of inserts from individual transformants in pPIC9 was established by restriction

10 endonuclease digestion using SacI and correctly oriented constructs were linearized with DraI prior to transfection of spheroplasts. Spheroplasts were prepared, transformed with 10 pg of linearized construct and plated using reagents and protocols provided with the Pichia Expression Kit (Invitrogen).

To prepare spheroplasts, a single colony of GS115 was grown in 10 ml of YPD (1% yeast extract, 2% peptone, 2%
glucose) overnight at 30 C. From this culture, 10 N1 were inoculated into 200 ml of YPD and grown at 30 C overnight until the OD6M reached 0.2. Cells were pelleted at 3000 rpm in the HN-S centrifuge at room temperature for 5 min, and resuspended in 20 ml sterile water. After re-pelleting, the cells were resuspended in 20 ml of SED (0.95 M sorbi-tol, 22.5 mM EDTA pH 8.0, 50 mM DTT). The cells were spun again and resuspended in 20 ml of 1 M sorbitol. After pelleting, the cells were next suspended in 20 ml SCE (1 M sorbitol, 1 mM
EDTA, 10 mM citrate pH 5.8) and divided into two aliquots of 10 ml each. The first aliquot received 7.5 ul of Zymolyase followed by incubation at 37 C. Time point samples of 200 pl were diluted into 800 N1 of 5% SDS and measured spectrophotometrically at OD8. to determine the optimal time required for spheroplasting.

The other 50 ml aliquot of cells was subjected to-Zymolyase for 20 min to achieve 70% spheroplasting. The cells were spun at 2000 rpm in the HN-S centrifuge for 10 min, gently washed with 10 ml of 1M sorbitol and respun. The spheroplasts were again washed with 10 ml of CaS (1 M

sorbitol, 10 mM Tris-HC1 pH 7.5, 10 mM CaC12), respun and resuspended in 0.6 ml of CaS.

To transform the spheroplasts, the 10 pl restriction digests were added to 100 pl of spheroplasts and incubated at room temperature for 10 min. PEG/CaT solution (20% PEG, 10 mM

Tris-HC1 pH 7.5, 10 mM CaC12) was added (1 ml) and incubation continued for another 10 min at room temperature. The spheroplasts were spun at 2000 rpm for 10 min at room temperature, and the supernatant was aspirated off. Cells were resuspended in 150 pl of SOS medium (1 M sorbitol, 0.3X YPD, 10 mM CaCl2) and incubated at room temperature for 20 min, followed by addition of 850 N1 1 M sorbitol and plating of 100 pl. To plate, 100 N1 of reaction was added to 10 ml molten RD agarose [1 M sorbitol, 1% glucose, 1.34% YNB, 0.00004% biotin, 0.005% amino acid mixture (glu, met, lys, leu, ile), 0.7% agarose] and poured onto a RDB plate (same composition as RD agarose except 0.7% agarose is replaced by 2% agar). Viability was tested by plating 100 pl of spheroplasts on a RDHB (RDB also containing 0.004% his) plate.

Controls without spheroplasts and without DNA were also plated.

Transformants appeared 4-6 days after plating and incubating at 30 C. Individual colonies (12 for each of 4 trials) were replica plated on MD [2% agar, 1% glucose, 1.34%
yeast nitrogen base (YNB), 0.00004% biotin] and MM (2% agar, 0.5% methanol, 1.34% YNB, 0.00004% biotin) plates to establish methanol utilization (mut) phenotypes of the colonies. To assess levels of expression of recombinant Alt a 1 in individual transformants, 3 ml cultures were grown in BMGY (1%
yeast extract, 2% peptone, 0.01 M potassium phosphate pH 6.0, 1% glycerol, 1.34% YNB, 0.00004% biotin) and cells were pelleted and transferred to BMMY (BMGY containing 0.5%
methanol instead of glycerol) one day later. After 2 days of growth at 30 C, aliquots of culture were removed and the cells were pelleted. A nitrocellulose membrane was spotted with 3 pl of each culture supernatant and allowed to dry. The dot blot was then immunostained using rabbit anti-Alt a 1 antibodies as described previously. Large cultures of a mut+

transformant expressing high levels of Alt a 1 were grown and induced in liquid MM medium supplemented with 1% casamino acids and 2.5% methanol. Culture medium was concentrated using an Amicon ultrafiltration unit with YM10 membrane, diluted with sterile water, and concentrated again. After centrifugation at 30,000 x g for 20 min, the supernatant was used in immunoblotting studies. The resulting solution was concentrated approximately 40 fold from the original culture medium.

Atopic sera from A. alternata-sensitive individuals were obtained locally; additional plasma samples were obtained commercially (PlasmaLab International). All sera scored 19%
or greater in RAST assays carried out in our laboratory. Total protein in concentrated conditioned medium from Pichia pastoris Alt a 1 transformant and A. alternata 34-016 crude extract was assayed using the method of Bradford (Anal Biochem 72: 248-254, 1976). Samples corresponding to 500 pg of extract protein or 27 pg of conditioned media protein were loaded into preparative 12% SDS-PAGE gels (Laemmli, Nature 227: 680-685, 1970) and blotted onto 0.2 pm nitrocellulose membrane in Towbin transfer buffer containing 15% methanol (Towbin et al, Proc Natl Acad Sci USA 76: 4350-4354, 1979). After blocking with 1% BSA in TBS, blots were screened in a Bio-Rad manifold.

Antisera or plasma were applied to blots at 1:4 in TBST and incubated overnight at room temperature. Immunoblots were washed with TBST for 3 x 5 min and incubated with AP-conjugated goat anti-human IgE at 1:1000 dilution for 2 h, also at room temperature. Washing and staining with NBT and BCIP were as described above.
Sequence Analyses Analyses of hydrophobicity (Kyte and Doolittle, J
Mol Biol 157: 105-132, 1982), secondary structure (Chou and Fasman, Biochemistry 13: 222-245, 1974) and amino acid content were carried out using Prosis software (Hitachi). The BLAST
algorithm was used for homology searches of GenBank and EMBL
databases (Altschul et al, J Mol Biol 215: 403-410, 1990).
Analyses of sequence block motifs were carried out using PROSITE (Bairoch, Nuc1 Acids Res 21: 3097-3103, 1993).
Southern analysis of PCR-amplified inserts from rabbit anti-G3 IgG-immunoselected clones High density immunoscreening of the A. alternata-specific cDNA library using rabbit anti-Alt a 1 hyperimmune.
serum yielded nineteen positive clones from a total of 305,000 pfu screened. The staining intensities of these positives varied considerably, suggesting that the clones represented either Alt a 1 clones of differing sizes, or sequences corresponding to more than one distinct gene product. None of the clones reacted positively when immobilized plaques were immunostained with pooled human atopic IgE from A. alternata-sensitive individuals.

Because lack of IgE reactivity of the clones could have resulted from epitopic masking of a small C-terminal Alt a 1 peptide sequence by the large N-terminal 5-galactosidase moiety of Xgtll fusion proteins, inserts from all nineteen positives were amplified by PCR, and Southern blots of the PCR
products were probed using ICHA7 (SEQ ID NO:6), a 23-mer oligonucleotide with 8-fold degeneracy designed from the Alt a 1 N-terminus (Curran et al, Int Arch Allergy Immunol 102:
267-275, 1993). This probe was hybridized to the PCR-amplified inserts at low stringency, owing to the size of the oligo and its degeneracy. As shown in Fig. 1, the sizes of the immunoselected inserts ranged from approximately 570-1880 bp, of which a large proportion were 800-900 bp in length. The majority of the clones bound the probe (Fig.2), although hybridization to clones rb25, rb5l and rb6l was more intense than to any of the others. In contrast, clones rb3l, rb82 and rb53 were unable to bind the probe, either because they were too short (570-640 bp) to include Alt a 1 N-terminal peptide sequences, or they represented a different gene product. Strongly immunoreactive clones rb9l, rb24, rb2l and rb81 also failed to bind the probe despite their large sizes (840, 860, 1030 and 1060 bp, respectively).

DNA sequencing of immunopositive clone rb5l On the basis of its relatively strong hybridization with ICHA7 (SEQ ID NO:6), clone rb5l (ATCC 97754) was selected for DNA sequencing. The 660 bp insert contains an open reading frame (ORF) of 471 bp, from positions 64 to 534 (SEQ
ID NO:3). A sequence exhibiting strong homology to SEQ ID
NO:6, which encodes amino acids near the Alt a 1 N-terminus, is located at positions 166-188. Searches of gene banks using the BLAST algorithm (Altschul et al, J Mol Biol 215: 403-410, 1990) failed to reveal significant sequence identity with genes encoding known proteins or allergens. Of other immunopositive clones that have been partially sequenced, eleven contain coding sequences that are common to rb51 (rbOl, rb12, rb14, rb23, rb25, rb4l, rb42, rb52, rb53 rb6l and rb82), although in at least one instance (rb42), 5' sequences that bridge Igtll ~-galactosidase and Alt a 1 coding sequences are distinct from those of rb51 (SEQ ID NO:3). In rb42, the absence of termination codons within the 5' untranslated sequences accounts for greatly enhanced 5-galactosidase fusion protein expression and hence, stronger immunoreactivity of rb42 clones relative to rb5l. Two clones (rb52 and rb7l) .CA 02217173 1997-11-25 contain known rRNA sequences, which likely accounts for their longer length. Clone rb25, a strongly immunoreactive and ICHA7-binding clone of 1400 bp, also contains distinct sequences of unknown identity. Clones rb81 and rb2l are homologous and contain distinct sequences with limited =
homology to plant endochitinases. Four clones have not yet been investigated. Therefore, we conclude that most of the immunopositive clones obtained contain some, if not the entire coding sequences of Alt a 1.

Table 1 Amino Acid Composition and pI of Alt a 1 and Translated rb5l ORF

Amino acid Alt a 1* Rb51 ORF
(residues/100) (residues/100) Aspartate/asparagine 12.18 11.4 Threonine 9.78 8.92 Serine 8.64 8.92 Glutamate/glutamine 7.75 7.00 Proline 7.20 3.82 Glycine 8.77 8.28 Alanine 9.73 10.83 Cystine 1.01 NA
Cysteine NA 3.18 Valine 6.03 4.46 Methionine 0.65 1.27 Isoleucine 3.03 3.18 Leucine 7.71 7.64 Tyrosine 4.53 5.10 Phenylalanine 5.37 5.73 Histidine 0.51 0.64 Lysine 5.44 5.73 Arginine 2.00 2.55 Tryptophan ND 1.27 pI 4.15 4.59 *data from Curran et al, Int Arch Allergy Immunol 102:

267-275, 1993.

Translation of rb5l ORF and comparison with Alt a 1 N-terminal peptide sequences The translated rb5l ORF encodes a peptide of 157 amino acids, with a calculated molecular mass of 16,960 (SEQ

ID NO:3). Comparisons of amino acid analyses between rb5l ORF
and Alt a 1 show considerable similarity (Table 1), except for the amino acid proline. However, conditions of hydrolysis previously used in the amino acid analysis of Alt a 1 (Curran et al, Int Arch Allergy Immunol 102: 267-275, 1993) may have resulted in the over estimation of proline in that analysis.
The predicted isoelectric point of rb5l is within 0.5 units of purified Alt a 1 (Table 1). A sequence homologous to one of the two expected N-terminal peptide sequences of Alt a 1 (SEQ
ID NO:1, SEQ ID NO:2) is located at residues 29-48 of the rb51 ORF, and matches the sequence of Curran et al at 16 or 17 of positions. The only dissimilar residues between rb5l ORF
and Alt a 1 are at positions 29, 30, 40 and 46. These variations in amino acid sequence indicate that either A.
alternata cells express multiple isoforms of rb5l, or that Alt 20 a 1 is prone to the development of slightly different polymorphic forms. Isoelectric focusing of purified Alt a 1 has revealed a major component with a pI of 4.15, and minor components with pIs of 4.25 and 4.40 (Curran et al, Int Arch Allergy Immunol 102: 267-275, 1993). Minor differences between the amino acid distributions of Alt a 1 and the translated rb5l ORF may also reflect the presence of more than one isoform in the sample subjected to amino acid analysis (Curran et al, Int Arch Allergy Immunol 102: 267-275, 1993).

However, if multiple isoforms of Alt a 1 were expressed in A.
alternata, cDNAs to such isoforms should also have been selected as distinct clones using rabbit anti-Alt a 1. A
comparison of three clones (rb5l, rb6l, rb42) whose DNA

sequences at the Alt a 1 N-terminus are known failed to reveal differences that could have accounted for the amino acid alterations expected at these specific positions. It remains possible that other clones may contain isoform-specific sequence deviations at these or other sites.

It is now accepted that protein isolates originally designated Alt a I and Alt a-29 represent the same allergen.
Despite a lack of sequence data, it is also possible that Alt a Bd 29K (Deards and Montague, Mol Immunol 28: 409-415, 1991) and Alt a I1563 (Paris et al, int Arch Allergy App1 Immunol 92:

1-8, 1990; Paris et al, J Allergy Clin Immunol 88: 902-908, 1991), disulfide-linked dimers of 29-31 kDa with pIs of 4.2, may also represent the same protein. However, the subunit sizes of Alt a I1563 are 19 and 20 kDa, considerably larger than those of Alt a 1, and this protein has been stated as containing 20% carbohydrate (Paris et al, J Allergy Clin Immunol 88: 902-908, 1991). Analysis of the translated rb5l ORF using PROSITE (Bairoch, Nucl Acids Res 21: 3097-3103, 1993) failed to reveal potential N-linked glycosylation sites.
Although there is no obvious consensus sequence for isolated 0-glycosylation sites (Wilson et al, Biochem J 275: 529-534, 1991), proline at -1 relative to the serine or threonine acceptor, or at -6 or +3 in the absence of a charged residue at -1, has been identified as a potential, but not absolute predictor of 0-linked glycosylation (Wilson et al, Biochem J
275: 529-534, 1991; O'Connell et al, Biochem Biophys Res Commun 180: 1024-1030, 1991). An examination of rb5l ORF
reveals 3 sites that fulfill these criteria, with acceptors at residues 44, 53 and 157. However, previous attempts to deglycosylate Alt a 1, either enzymatically with endoglycosidase F or 0-glycosidase, or chemically using TFMS, invariably failed to reduce the Mr of the protein (Curran et al, Int Arch Allergy Immunol 102: 267-275, 1993), suggesting that Alt a 1 is non-glycosylated.

Hydrophobicity and secondary structure prediction of rb5l ORF
A hydrophobicity analysis of the rb51 ORF was carried out according to the method of Kyte and Doolittle, using a window of 6 amino acids (Kyte and Doolittle, J Mol Biol 157: 105-132, 1982). The longest region of strong hydrophobicity in the rb5l ORF is located at the N-terminus, from amino acids 10-23 (Fig. 3). This region coincides with the longest a-helical domain observed in the secondary structure prediction illustrated in Fig. 4. As amino acids 1-28 were not previously detected in microsequenced Alt a 1 N-terminal peptides, it is likely that the first 28 amino acids represents a cleaved signal peptide that directs Alt a 1 for export from the cell, as significant amounts of Alt a 1 are secreted into culture broth of A. alternata (Curran et al, Int Arch Allergy Immunol 102: 267-275, 1993).

Northern analysis of rb5l transcripts in A. alternata total RNA

To assess the size of rb51 transcripts in A.
alternata, total RNA was extracted from mycelial samples taken at various times post-inoculation and analyzed by Northern blot probed with DIG-labeled rb5l. Total RNA was extracted with little or no degradation, as is evident from the integrity of the rRNA bands in Figure 5. However, we consistently observed a substantial proportion of insoluble nucleic acid in samples from 13, 17 and 20 day cultures, likely caused by the presence of contaminating DNA. Rb51 is shown as an abundant transcript of approximately 700 bases (F;_g. 6), indicating that the rb5l clone is similar in size to the full length Alt a 1 transcript. Peak levels of rb5l transcripts were observed in 5 day samples, which decreased rapidly from 7 to 10 days post-inoculation.

Expression and IgE-binding capability of recombinant Alt a 1 protein in immunoblots To express recombinant Alt a 1 protein in large quantity, rb5l sequences were cleaved with NheI and XbaI and ligated into pPIC9, a Pichia pastoris secreted expression vector (Invitrogen). This vector features a S. cerevisiae a-factor prepro peptide sequence located upstream from its multiple cloning site. It was anticipated that the insertion of rb5l into the AvrII site in this vector would be in the appropriate reading frame for correct expression of the recombinant allergen, although the expressed protein would contain 20 additional N-terminal amino acids followed by the Alt a 1 N-terminal signal peptide (SEQ ID NO:5). Spheroplasts were transformed with this construct, and transformants were screened for their methanol utilization phenotype and for production of secreted protein reactive with rabbit anti-Alt a 1 sera. A single transformant of mut+ phenotype was isolated that secreted recombinant Alt a 1 protein into conditioned medium. Aliquots of concentrate (1 pg) were compared with an equal amount of Alt a 1 that was extracted from A. alternata, and further purified by ultrafiltration and gel filtration in Sephadex G-100. Both natural and recombinant allergens bound rabbit anti-Alt a 1 as well as human atopic IgE (Fig. 7, columns A and B, respectively). The recombinant (Fig. 7, lanes 1) and natural (FIg. 7, lanes 2) allergens were compared under both non-reducing and reducing conditions in immunoblots. The recombinant Alt a 1 protein, like the natural allergen, associated as a dimer in culture medium as shown by the 29 kDa band in non-reducing immunoblots (Fig. 7, top row). The dimer also appears to show greater reactivity with IgE than do reduced monomers, indicating that some IgE antibodies in atopic sera may be specific for epitopes generated by subunit interactions.

Under reducing conditions (Fig. 7, bottom row), the recombinant Alt a 1 migrated as a doublet of bands with slightly less mobility than those of the natural allergen (Fig. 7, columns A, B, D) even though both subunits of the recombinant allergen are translated from the same Alt a 1 DNA

sequence. A third band is sometimes visible below this doublet, although it is usually fainter than the other two.
These findings imply that post-translational processing of the subunits likely occurs during or following their association as a dimer. These processing functions may be intrinsic to the subunit, or may be carried out by other secreted proteins such as extracellular proteases. However, P. pastoris carries out relatively little protein secretion of its own and likely secretes a significantly different profile of proteins than would A. alternata. Therefore, it remains possible that dimerized Alt a 1 subunits may catalyze their own processing events. Differential processing of the subunits also appears to occur despite the presence of additional N-terminal amino acids in Pichia-secreted recombinant Alt a 1. At present, it is uncertain whether P. pastoris is capable of cleaving the endogenous Alt a 1 signal peptide as well as the yeast a-factor signal peptide derived from the pPIC9 Pichia expression vector. Recombinant and natural Alt a 1 bands were of similar intensity and purity in Coomassie blue R250-stained blots (Fig. 7, column D). Thus, the production of Alt a 1 from culture medium of Pichia pastoris offers a convenient source for the purification of the recombinant allergen.
Reactivity of recombinant Alt a 1 with serum IgE from 57 A.

alternata-sensitive individuals The reactivity of recombinant Alt a 1 with serum IgE
from 57 A. alternata-sensitive individuals was compared with crude A. alternata 34-016 extract in immunoblots run under reducing conditions in Figs. 8 and 9. Both Alternaria-derived and recombinant Alt a 1 subunits interact relatively weakly with most of the commercially obtained plasma samples (lanes 1-17), although visible staining of the fastest migrating band of the recombinant doublet is evident in 14 of the 17 lanes.

Lanes immunostained with locally obtained IgE tended to show stronger and more consistent reactivity with both bands of the recombinant doublet than those incubated with plasma samples (lanes 18-57). In total, 46 of 57 serum and plasma IgEs tested reacted positively with recombinant Alt a 1(82$), indicating that the recombinant protein may be of use in the immunodiagnosis of allergy to A. alternata. The reactivity of the sera to recombinant Alt a 1 appeared to correlate with that of the natural allergen in crude extracts of A. alternata isolate 34-016, although immunostaining of the natural allergen was especially weak in lanes 41-57. This can be primarily attributed to limitations in the amounts of crude extract that can be loaded into gels without gel distortion, as well as to the relative insensitivity of the immunodetection methods utilized.

These results confirm that Alt a 1 represents a major allergen of A. alternata and also demonstrates the potential benefit of recombinant Alt a 1 as an immunodiagnostic reagent with greater sensitivity than crude extracts of A. alternata. The ease with which relatively pure Alt a 1 has been obtained will facilitate the development of specific ELISA and RAST assays or monoclonal antibodies.
Hydrophobicity and secondary structure predictions of Alt a 1 offer clues to locating antigenic and allergenic sites on Alt a 1, as antigenic determinants on proteins have been predicted to reside at hydrophilic sites that will be exposed to the aqueous medium and readily accessible for binding to B cells (Hopp and Woods, Proc Nat1 Acad Sci USA 78: 3824-3828, 1981).

T cell epitopes are predicted to be amphipathic, with separated hydrophobic and hydrophilic surfaces, as these epitopes are presented t=o T cells as a processed antigen bound =to histocompatibility antigens on the surface of antigen presenting cells (APC) (DeLisi and Berzofsky, Proc Nat1 Acad Sci USA 82: 7048-7052, 1985). Alt a 1 appears to be a predominantly hydrophilic protein that contains relatively little a-helical sequence.

SEQUENCE LISTING
(1) GENERAL INFORMATION

(i) APPLICANT: De Vouge, Michael W.
Vijay; Hari M.

(ii) TITLE OF INVENTION:
ISOLATION OF A cDNA CLONE ENCODING A SUBUNIT OF THE
ALTERNARIA ALTERNATA MAJOR ALLERGEN, Alt a 1, AND
PRODUCTION OF RECOMBINANT PROTEIN THEREOF

(iii) NUMBER OF SEQUENCES: 6 (iv) CORRESPONDENCE ADDRESS:
(A) Address: George A. Seaby Seaby and Maclean (B) Street: 880 Wellington Street, Suite 708.
(C) City: Ottawa (D) Province: Ontario (E) Country: Canada (F) Postal Code: K1R 6K7 (v) COMPUTER READABLE FORM:
(A) Medium Type: 3.5 inch diskette (B) Computer:. IBM PC compatible (C) Operation System: PC-DOS/MS-DOS
(D) Software: ASCII. Text (vi) CURRENT APPLICATION DATA:
(A) Application Number:
(B) Filing Date:
(C) Classification:

(vii) ATTORNEY/AGENT INFORMATION:
(A) Name: Ge,orge A. Seaby (B) Registration.Number:
(C) Reference/Docket Number:

(viii) TELECOMMUNICATION INFORMATION:
(A) Telephone: (613) 232-5815 (B) Fax: (613) 232-5831 (2) INFORMATION FOR SEQ ID NO: 1:

(i). SEQUENCE CHARACTERISTICS:
(A) Length: 20amino acids (B) Type: protein (C) Strandedness:
(D) Topology: lihear (ii) MOLECULE TYPE: Alternative N-terminus of Alt a 2 subunit (iii) FEATURE:
(A) Name/Key:
(B) Location: 1...20 (iv) SEQUENCE DESCRIPTION: SEQ ID NO: 1 Ala Asp Pro Val Thr Thr Glu Gly Asp Tyr Val Val Lys Ile Ser Glu Phe Tyr Gly Arg (2) INFORMATION FOR SEQ ID NO: 2:
(i) SEQUENCE CHARACTERISTICS:
(A) Length: 20 amino acids (B) Type-: protein (C) Strandedness:
(D) Topology: linear (ii) MOLECULE TYPE: Alternative N-terminus of Alt a 1 subunit (iii) FEATURE:
(A) Name/Key:
(B) Location: 1...20 (iv) SEQUENCE DESCRIPTION: SEQ ID NO: 2 Ala Asp Pro Val Thr Thr Glu Gly Asp Tyr Val Val Lys Ile Ser Glu Phe Phe Gly Arg (2) INFORMATION FOR SEQ ID NO: 3:
(i) SEQUENCE CHARACTERISTICS:
(A) Length: 660 base pairs (B) Type: nuclei_c acid (C) Strandedness: single ,(D) Topology: linear (ii) MOLECULE TYPE: cDNA and its encoded protein (iii) FEATURE:
(A) Name/Key: open reading frame of rb5l clone encoding Alt a 1 allergen Location: 64...534 (iv) SEQUENCE DESCRIPTION: SEQ ID NO: 3:

Met Gln Phe Thr Thr Ile Ala Ser Leu Phe Ala Ala Ala Gly Leu Ala Ala Ala Ala Pro Leu Glu Ser Arg Gln Asp Thr Ala Ser Cys Pro Val Thr Thr Glu Gly Asp Tyr Val Trp Lys Ile Ser Glu Phe Tyr Gly Arg Lys Pro Glu Gly Thr Tyr Tyr Asn.Ser Leu Gly Phe Asn Ile Lys Ala Thr Asn G1y Gly Thr Leu Asp Phe Thr Cys Ser Ala Gin Ala Asp Lys Leu Glu Asp His Lys Trp Tyr Ser Cys Gly Glu Asn Ser Phe Met Asp Phe Ser Phe Asp Ser Asp Arg Ser Gly Leu Leu Leu Lys Gln Lys Val Ser Asp Asp Ile Thr Tyr Val Ala Thr Ala Thr Leu Pro Asn Tyr Cys Arg Ala Gly Gly Asn Gly Pro Lys Asp Phe Val Cys Gln Gly Val Ala Asp Ala Tyr Ile Thr Leu GTC ACT CT~C CCC AAG AGC TCT TAA GCGATAGTTG GTCGAGCATA 557 Val Thr Leu Pro Lys Ser Ser ***

(2) INFORMATION FOR SEQ ID NO: 4:

(i) SEQUENCE CHARACTERISTICS: (A) Length: 242 base pairs (B). Type: nucleic acid (C) Strandedness: single (D) Topology: linear (ii) MOLECULE TYPE: cDNA and its encoded protein (iii) FEATURE:
(A) Name/Key: partial.5' sequence of rb42 clone encoding Alt a 1 allergen Location: 1...242 (iv) SEQUENCE DESCRIPTION: SEQ ID NO: 4:

Met Gin TTC ACCACC ATC'GCC TCT CTC TTC GCC GCC GCT GGC CTT GCC GCT 90 Phe Thr Thr Ile Ala Ser Leu Phe Ala Ala Ala Gly Leu Ala Ala Ala Ala Pro Leu Glu Ser Arg Gln Asp Thr Ala Ser Cys Pro Val Thr Thr Glu Gly Asp Tyr Val Trp Lys Ile Ser Glu Phe Tyr Gly Arg Lys Pro Glu Gly Thr Tyr Tyr Asn Ser Leu Gly Phe Asn Ile AAG GCC ACC AAC GGG GG. ... 242 Lys Ala Thr Asn Gly Gly ...
(2) INFORMATION FOR SEQ ID NO: 5:
(i) SEQUENCE CHARACTERISTICS:
(A) Length: 177 amino acids (B) Type: amino acid (C) Strandedness:
(D) Topology: linear (ii) MOLECULE TYPE: recombinant Alt a 1 protein (iii) FEATURE:
(A) Name/Key: amino acids contributed by pPIC9 DNA
sequence Location: 1..9 (B) Name/Key: amino acids contributed by 5' untranslated sequences of Alt a 1 cDNA
Location: 10...20 (C) Name/Key: amino acids corresponding-to Alt a 1 coding sequence Location: 21...177 (iv) SEQUENCE DESCRIPTION: SEQ ID NO: 5:

Glu Ala Glu Ala Tyr Val Glu Phe Pro Ser Leu Val Glu Ser Gin 10 1~5 Ile Asn Phe Ile Lys Met Gln Phe Thr Thr Ile Ala Ser Leu Phe Ala Ala Ala Gly Leu Ala Ala Ala Ala Pro Leu Glu Ser Arg Gln Asp Thr Ala Ser Cys Pro Val Thr Thr Glu Gly Asp Tyr Val Trp Lys Ile Ser Glu Phe Tyr Gly Arg Lys Pro Glu Gly Thr Tyr Tyr Asn Ser Leu Gly Phe Asn Ile Lys Ala Thr Asn Gly Gly Thr Leu Asp Phe Thr Cys Ser Ala Gln Ala Asp Lys Leu Glu Asp His Lys Trp Tyr Ser Cys Gly Glu Asn Ser Phe Met Asp Phe Ser Phe Asp Ser Asp Arg Ser Gly Leu Leu Leu Lys Gln Lys Val Ser Asp Asp Ile Thr Tyr Val Ala Thr Ala Thr Leu Pro Asn Tyr Cys Arg Ala Gly Gly Asn Gly Pro Lys Asp Phe Val Cys Gln Gly Val Ala Asp Ala Tyr Ile Thr Leu Val Thr Leu Pro Lys Ser Ser (2) INFORMATION FOR SEQ ID NO: 6:

(i) SEQUENCE CHARACTERISTICS:
(A) Length: 20 bases (B) Type: nucleic acid (C) Strandedness: single (D) Topology: linear (ii) MOLECULE TYPE: hybridization oligonucleotide (iii) FEATURE:
(A) Name/Key: degenerate oligonucleotide .designed from Alt a-29 N-terminus Location: 1...23 (iv) SEQUENCE DESCRIPTION: SEQ ID NO: 6

Claims

WE CLAIM:

1. A nucleic acid molecule encoding a subunit-of the major allergen of Alternaria alternata, Alt a 1 comprising the open reading frame of clone rb51(ATCC 97754), from bases 64-534 (amino acids 1-157) of SEQ ID NO: 3: