EP0668920A1

EP0668920A1 - Methods for converting a, ab, and b blood types to o blood type

Info

Publication number: EP0668920A1
Application number: EP94901346A
Authority: EP
Inventors: Randy M. Berka
Original assignee: Genencor International Inc
Current assignee: Danisco US Inc
Priority date: 1992-11-18
Filing date: 1993-11-08
Publication date: 1995-08-30
Also published as: WO1994011518A2; WO1994011518A3; CA2149641A1; JPH08503137A

Abstract

A process for producing α-N-acetylgalactosaminidase by expressing a DNA sequence derived from human placenta encoding α-N-acetylgalactosaminidase in a recombinant host microorganism transformed with an expression vector containing DNA sequence and isolating transformants having α-N-acetylgalactosaminidase expression. The isolated transformants are then cultured and α-N-acetylgalactosaminidase is purified from the culture media. The recombinantly produced α-N-acetylgalactosaminidase is used to convert A, B and AB blood types into O blood type.

Description

METHODS FOR CONVERTING A, AB, AND B BLOOD TYPES TO O BLOOD TYPE

Field of Invention

The present invention relates to the cloning of a cDNA derived from human placenta mRNA which gene encodes for α-N- acetylgalactosaminidase and the expression of this gene in bacteria, yeast and filamentous fungi. The present invention further relates to the use of this recombinantly produced enzyme to convert type A, AB, and B erythrocytes into 0 erythrocytes and to a kit containing recombinantly produced α-N- acetylgalactosaminidase that can be used for this conversion.

Background of the Invention

Transfusion of blood to patients who have suffered a blood loss has been an accepted medical practice for almost a century. Scientific advances in the transfusion area have made it possible for blood to be stored prior to transfusing the blood into recipients, as well as further processing the blood into specific blood components which can be given to a patient in need of a particular component, thus extending the utility of a collected amount of blood.

For blood transfusions, it is necessary to match the blood type of the recipient during transfusion processes with the same blood type to prevent systemic reactions which can occur when patients are transfused with incompatible blood. For example, recipients with type A blood must be transfused with type A blood, and recipients with type B blood must be transfused with .type B blood. An exception to this is type O blood which can be transfused into type A, B, and AB recipients as well as O recipients. Thus, persons with the O blood type have been termed "universal donors."

In the operation of a blood bank facility, it is necessary to maintain all four different blood types for transfusion purposes. Type O donor blood is used, therefore, in most cases, for type O recipients. However, a majority of blood donors have A, B or AB blood types, and there can exist an excess of these blood types as compared to type O blood. Specifically, it has been desired to convert A, B or AB blood into type O blood for transfusion purposes. Moreover, in times of national disasters and war there is often a lack of specific blood types and thus it would be advantageous to be able to easily convert the needed blood type to type O blood. Furthermore, it would be advantageous if the blood can be readily converted in a disaster or war zone environment with ease and simplicity.

The term "blood group" is applied to a well-defined system of red blood cell antigens controlled by a locus having a variable number of allelic genes such as A, B, and O in the ABO system. Individuals of blood groups A, B, and O express A and H, B and H, and H antigens, respectively, on their erythrocytes. The antigenic molecule consists of one or more straight or branched carbohydrate chains attached to a ceramide or a peptide backbone, which in turn is embedded in the lipid bilayer of the cell membrane. Blood group specificity is determined by the nature and linkage of the monosaccharides at the ends of the carbohydrate chains (see, for example, Figure 1) . For H antigenic activity, the immunodominant sugar fructose is bound in an α-l,2-glycosidic linkage to the penultimate galactose residue. Blood group A or B^,s specificity occurs when N- acetylgalactosamine or galactose is attached in an α-1,3 linkage to the same residue as fucose. Given such structures, it is possible to convert A or B antigenic determinants into H by using specific exoglycosidases to hydrolyze only the N- acetylgalactosamine or galactose linked through a terminal α- glycoside, thus producing group 0 cells.

For instance, the conversion of human B erythrocytes from type B to type 0 with a crude preparation of Clostridial α- galactosidase has been reported by Fujisawa, et al., in Proc. Jap. Acad.. 39:319-324 (1963). Moreover, this enzyme has been purified to homogeneity from Clostridium and similarly used to convert type B blood to type 0 in human erythrocytes as described by Furukawa, et al., in Blood and Tissue Antigens (Aminoff, D., ed.), pp. 415-425, Academic Press, New York (1970).

U.S. Patent No. 4,330,619 relates to the use of α-galactosidase from green coffee beans to convert blood type B erythrocytes into type O cells by equilibrating type B erythrocytes to a pH of 5.7 to 5.8; contacting the erythrocytes with α-galactosidase for a sufficient period of time to convert the B antigen into the H antigen; removing the enzyme from the erythrocytes and reequilibrating the erythrocytes to a pH of 7.2 to 7.4. Similarly, U.S. Patent No. 4,609,627 relates to the conversion of type A erythrocytes to type O erythrocytes by using an α-N- acetylgalactosaminidase obtained from avian liver.

Thus, the exoglycosidases which are capable of removing immunodominant sugar moieties from ABO antigenic determinants are found in a variety of sources such as green coffee beans and avian liver as described above. Moreover, α-galactosidases have been partially purified from a variety of sources including microorganisms, mammalian tissues, plants, watermelons, mung beans, guar and the like. Similarly, the α-N- acetylgalactosaminidases occur in bacteria, mollusks and earthworms and have been obtained from various mammalian sources including human liver. These enzymes, derived from different sources, may act by converting only one substrate or the enzymes may act on a variety of different substrates.

Furthermore, two isoenzymes of α-galactosidase, namely α- galactosidase A and α-galactosidase B have also been isolated from human tissue. See, Kusiak, et al., "Purification and Properties of the Two Major Isoenzymes of α-Galactosidase from Human Placenta," The Journal of Biological Chemistry. Vol. 253, No. 1, (1978) pp. 184-190. Moreover, Dean, et al., in Biochim. Biophys. Res. Co mun. 77 (1977) pp. 1411-1417, identified α- galactosidase purified from a normal human liver as an α-N- acetylgalactosaminidase based upon the abilities of the enzyme to hydrolyze substrates with either α-galactose or α-N- acetylgalactosamine residues, as well as the ability of O-NP-α- GalNAc to competitively inhibit the hydrolysis of 4-MU-α-Gal.

Since this enzyme is produced as a small percentage of the total protein in each of these sources, purification procedures often vary and are quite tedious and time-consuming depending on the source from which the enzyme is derived. For example, α- galactosidase, known to convert the B antigen into the O antigen, is produced in small amounts in green coffee beans. The production of green coffee beans is seasonal, expensive and requires a tropical climate for high yields of production. The recovery of the enzyme, α-galactosidase, from green coffee beans would require large-scale manufacturing with minimal amounts being recovered and, hence, is not economically practical.

Another purification scheme from human placenta tissue is described by Kusiak, et al., supra and requires four different column purification steps which include purification through Concanavalin-A Sepharose, DEAE-Cellulose, Sephadex G-200 and SP- Sephadex.

Moreover, the amount of purified enzyme recovered from the various sources is often minimal and quite often in microgram to milligram quantities. This is exemplified by recovery from the purification scheme described by Kusiak, et al., supra. in which only 18% total recovery was achieved for 10 kg of human placenta. Therefore, it is impractical to use this enzyme in blood banks to convert A, B and AB blood to 0 blood, since much larger quantities of the enzyme are needed for this conversion process.

Recently, it has been discovered that an α-N- acetylgalactosaminidase purified from a human source has a much broader substrate specificity than other exoglycosidases. Namely, this enzyme catalyzes the hydrolysis of not only α-N- acetylgalactosamine, but also the α-linked galactosidase as described by Desnick, et al. , in The Metabolic Basis of Inherited Disease. Scriver, et al., Editors, 6th Edition, McGraw-Hill, New York (1989) . Thus, this enzyme can convert A, B, and AB blood types into type 0 blood. Ideally, the use of recombinant DNA methods to produce α-N-acetylgalactosaminidase would afford a practical method for the production of this enzyme on an industrial scale. This large scale production would thus permit numerous blood banks to convert donor A, B, and AB blood into type 0 blood which could then be used for transfusing any blood type (A, B, AB or 0) recipient.

WO 87/07461 describes the production of guar α-galactosidase by hosts transformed with recombinant DNA methods. The enzyme recovered from the use of the recombinant techniques is suitable for reducing the galactose content of a galacto annan containing 1-6 linked α-D-galactopyranosyl units attached to the main of 1-4 linked β-D-mannopyranosyl units. It is used to prepare foodstuffs, animal feedstuffs or cosmetics in which a galactomannan with reduced galactose content is used therein. Although recombinant techniques have been used to produce guar seed α-galactosidase, the recombinant techniques are specific for the enzymes derived from guar seeds and this enzyme cannot be used in a blood conversion process. Moreover, an exoglycosidase derived from a human source, rather than other microbial and plant sources, would be beneficial since the enzyme is derived from the same mammalian source and co-reactivity among different sources may vary.

Accordingly, it is an object of this invention to clone a human placenta cDNA encoding for α-N-acetylgalactosaminidase and to produce microorganisms which are capable of expressing this enzyme in large quantities in industrial fermentations.

Yet another aspect of the present invention is to permit easy removal of the enzyme N-acetylgalactosaminidase after blood conversion by gene fusing the enzyme with a fusion protein having cellulose-binding properties.

Yet another aspect of the present invention is to alter the amino acid sequence of the gene encoding for the enzyme in such a way to permit the enzyme to function at a pH that is compatible with the blood erythrocytes such that no pH step is needed for the enzyme to act upon the erythrocytes in the conversion process.

Summary of the Invention

Expression of α-N-acetylgalactosamidase in microorganisms such as yeast, filamentous fungi and bacteria has now been obtained and this enzyme can be easily purified in large quantities from cultures containing transformants that can express α-N- acetylgalactosaminidase.

Accordingly, the present invention relates to a process for producing α-N-acetylgalactosaminidase by: a) expressing a DNA sequence derived from human placenta encoding α-N-acetylgalactosaminidase in a recombinant host microorganism, said recombinant host microorganism being transformed with an expression vector containing said DNA sequence; b) isolating transformants having α-N- acetylgalactosaminidase expression; c) culturing said isolated transformants in a culture media to permit growth of said transformants; and d) purifying α-N-acetylgalactosaminidase from said culture media.

Further, the present invention relates to a process for enhancing the expression of α-N-acetylgalactosaminidase in a recombinant host microorganism.

In addition, the present invention relates to transformants having the capacity for α-N-acetylgalactosaminidase expression.

Furthermore, the present invention is directed to a recombinant enzyme composition and a kit containing such composition that can be used to convert A, B, and AB blood types to 0 blood. Brief Description of the Drawings

FIG. 1 is a representation of the blood converting enzymes in blood types A, B, and 0.

FIG. 2 is the deduced amino acid and nucleotide sequence of α-N- acetylgalactosaminidase (Seq. ID No. 4) .

FIG. 3 is a figurative representation of the construction of fungal expression vector pGPT-pyrG:α-gal. The first primer shown in Figure

3 is Seq. ID No. 5 and the second primer shown is Seq. ID No. 6.

FIG. 4 is a figurative representation of the construction of yeast expression vector YEpsecl:α-gal. The first primer shown in Figure

4 is Seq. ID No. 7 and the second primer shown is Seq. ID No. 8.

FIG. 5 is a figurative representation of the construction of bacterial expression vector pNH-STα-gall. The first primer shown in Figure 5 is Seq. ID No. 9 and the second primer shown is Seq. ID No. 10.

Detailed Description of the Invention

The present invention relates to the isolation and characterization of the gene encoding for α-N-acetylgalactosaminidase from human placenta. This gene is then further cloned into expression vectors, which are used to produce transformed microorganisms. These transformed microorganisms, which include bacterial, yeast and fungal species, express α-N-acetylgalactosaminidase, which can then be purified from the fermentation media in large quantities. As used herein, α-N-acetylgalactosaminidase is also recognized in the art as α-galactosidase and hence is also referred to as α-gal within this text. Also contemplated by the present invention is the manipulation of the amino acid sequence of α-N-acetyl-galactosaminidase by alteration of the active site on this enzyme which may lead to a variety of different changes in catalytic conversion. Moreover, manipulation of the amino acid sequence of α-N- acetylgalactosaminidase may result in further changes such as different pH optima, different temperature optima, altered catalytic turnover (Vmax) or altered affinity (Km) . Acceptable methods for such protein engineering are described in US Patent No. 4,760,025, the disclosure of which is incorporated herein by reference. Thus, it is contemplated to alter the amino acid sequence of α-N-acetylgalactosaminidase and recombinantly express the altered enzyme in microorganisms. As used in this application, the phrase "or a modification thereof" when referring to an amino acid or nucleotide sequence is meant to cover the manipulations described above.

Moreover, transformants expressing α-N-acetylgalactosaminidase can have multiple copies of this gene integrated into their genomes with the potential for producing increased amounts of α- N-acetylgalactosaminidase and thus increasing the yield of production of this blood converting enzyme.

Also contemplated by the present invention is the use of a cassette encoding for only the active site domain within α-N- acetylgalactosaminidase, which cleaves only the carbohydrate residues necessary to convert blood types to 0, thereby permitting the enzyme to function without extraneous amino acid sequence.

Further contemplated by the present invention is the attachment of various moieties to the sequence encoding for α-N- acetylgalactosaminidase prior to the insertion of the enzyme into an expression vector. The enzyme would then be altered in such a way to permit removal of the enzyme after use on erythrocytes.

Generally, the present invention involves the screening of human cDNA libraries that contain sequences specific for α-N- acetylgalactosaminidase, the isolation of clones containing a portion of the sequence encoding for this enzyme, isolation of the full cDNA sequence encoding α-N-acetylgalactosaminidase from a human library, the construction of expression vectors and transformation of these expression vectors into appropriate host microorganisms which then express the enzyme. Recovery of the enzyme from the fermentation media in large quantities is then possible and the pure enzyme can be isolated therefrom.

A variety of cDNA libraries can be screened to test for the presence of α-N-acetylgalactosaminidase. However, it is preferable to choose a source where the enzyme has been known to be isolated and also a human library. Most preferably, the following cDNA libraries can be used to screen for the presence of α-N-acetylgalactosaminidase: human placenta, human bone marrow, human breast, human breast carcinoma and human liver made in λgtll. Each library is then amplified to increase the titer of

₉ recombinant phage particles to at least 1 x 10 pfu/ml. Total DNA is isolated from each of the amplified libraries by methods known in the art. After isolation of total DNA from each of the above-described libraries, an aliquot of DNA is then taken and used as a template in a polymerase chain reaction (PCR) to identify which of the libraries contained cDNA sequences which are specific for α-N-acetylgalactosaminidase.

Synthetic oligonucleotide primers are then designed and synthesized based on the cDNA sequence described by Yamauchi, et al. , in "Molecular Cloning of Two Species of cDNAs For Human α-N- Acetylgalactosaminidase and Expression in Mammalian Cells," Biochemical and Biophysical Research Communications _r 170 (1990) pp. 231-237. Any primer that is sufficient to amplify a fragment specific for α-N-acetylgalactosaminidase can be used. However, it is preferable to use the following described primers. The forward primer has the following nucleotide sequence:

5' GAGCAATCCCGGGCCCAGATGGCCCTGTGG 3' (Seq. ID No. 1)

and the reverse primer has the following nucleotide sequence:

5' CTGTCTAGACCCAGCTCCTCACTGCTGGGA 3' (Seq. ID No. 2).

The reaction products for each amplification are then analyzed by agarose gel electrophoresis to determine which of the libraries contained sequences specific for α-N- acetylgalactosaminidase. In this respect, a lane containing only one band means that the specific fragment of α-N- acetylgalactosaminidase was amplified. The amplified fragment is then further excised from the gel, cut with appropriate restriction enzymes and subcloned into a vector cut with the same restriction enzymes.

Any vector can be used to subclone the fragment such as M13mpll, M13mpll FX, M13mpl8 and the like. It is preferable to use M13mpl8. DNA sequencing analysis of the subcloned fragment is then performed to verify that the subcloned fragment is indeed a portion of the α-N-acetylgalactosaminidase cDNA.

Full length cDNA clones encoding α-N-acetylgalactosaminidase are then isolated by using the PCR-amplified fragment as a probe. The fragment is first radio-labeled using techniques described by Maniatis, et al., and used to probe the cDNA library that contains sequences specific for the enzyme using the hybridization methods described by Davis, et al., Advanced Bacterial Genetics, pp. 162-165 (1980) . The plaques that were found to hybridize strongly with the probe are then purified twice by single-plaque isolation and amplified as described above.

Confirmation that the plaques indeed contain α-N- acetylgalactosaminidase cDNA sequences are determined by secondary screening with another synthetic oligonucleotide probe containing a portion of the nucleotide sequence which will hybridize to a portion of the cDNA. Any probe can be utilized for the secondary screening that is specific for α-N- acetylgalactosaminidase, but it is preferable to use a synthetic oligonucleotide which sequence differs from the sequence used in the primary screening. It is most preferable to use a synthetic oligonucleotide probe containing a portion of the nucleotide sequence just upstream of the coding region to ensure that a full length cDNA is obtained. The positive clones containing all of the α-N-acetylgalactosaminidase coding region are then identified and the full length α-N-acetylgalactosaminidase cDNA isolated from these clones and is further ligated into an expression vector.

The full length α-N-acetylgalactosaminidase cDNA can be ligated into a variety of expression vectors such as plasmid vectors, bacteriophage λ vectors, and the like. The essential elements of fungal, yeast and bacterial gene expression systems have two basic components: (1) an expression unit comprising transcriptional, translational, and in some cases secretory signals from the host organism, joined to the DNA sequences which encode the product of interest; and (2) a transformation system for introducing the expression unit and maintaining it in the host organism. The basic components of an expression unit include a strong promotor for efficient transcription, an efficient translation initiation region, DNA sequences coding for the particular gene of interest, DNA sequences encoding a secretory original peptide (where necessary for secretion) and transcription termination and polyadenylation sequences. The expression units are usually assimilated in a plasmid vector which can replicate in Escherichia coli such as pUC18, pBR322 and thereby facilitating construction of the expression vectors and allowing for easy preparation of large quantities of vector DNA. Often the plasmid is modified by the inclusion of a marker gene which can be used for the selection of transformants. Any expression vector can be constructed which permits the ligation of the entire α-N-acetylgalactosaminidase coding region into the vector such that expression of α-N-acetylgalactosaminidase is possible.

Furthermore, the vector should contain a suitable selectable marker to enable detection of the transformed microorganism. A number of selectable markers are available and are well known to those skilled in the art. Such markers generally are selectable by virtue of their ability to complement auxotrophic deficiencies or to confer drug resistance, or by virtue of their ability to utilize a particular growth substrate or synthesize an essential nutrient. A variety of selectable markers that can be used in the expression vectors include but are not limited to: argB, amdS, pyr4, ura3 , pyrG, and trpl .

Transformation of the expression vectors usually requires a host strain that is a mutant strain which lacks or has a nonfunctional gene or genes corresponding to the selectable marker incorporated into the expression vector. For example, if the selectable marker of argB is used, then a specific argr— mutant strain is used as the recipient in the transformation process. A variety of mutant strains can be prepared by a number of techniques known in the art, such as the filtration enrichment technique described by Nevalainen in "Genetic Improvement of Enzyme Production in Industrially Important Fungal Strains," Technical Research Center of Finland, Publication 26 (1985) . Another technique to obtain the mutant strain is to identify the mutants under different growth medium conditions. For example, pyr- mutant strains can be selected by subjecting the strains to fluoroorotic acid (FOA) . Strains with an intact pyrG gene grow in an uridine medium and are sensitive to fluoroorotic acid, and therefore, it is possible to select pyrG mutant strains by selecting for FOA resistance.

After the suitable vector is constructed, it is used to transform various strains. Since the permeability to DNA of the cell wall in fungi and yeast is very low, uptake of the desired DNA sequence, gene or gene fragment is at best minimal. To overcome this problem associated with transformation, the permeability of the cell wall can be increased or the DNA can be shot directly into the cells via a particle gun approach. In the particle gun approach, the DNA to be incorporated into the cells is coated onto micron size beads and these beads are literally shot into cells leaving the DNA therein and leaving a hole in the cell membrane. The cell then self-repairs the cell membrane and the DNA is incorporated in the cell. Besides the shot gun method, there are a number of methods used in the art to increase the permeability of the cell walls in mutant strains of fungus prior to the transformation process. One method involves the addition of alkali metal ions to the fungal cells. Any alkali metal or alkaline earth metal ion can be used in the present invention; however, it is preferable to use either CaCl₂ or lithium acetate, more preferably lithium acetate. The concentration of the ions may vary depending on the ion used, and usually between 0.05 M to 0.4 M concentrations are used. It is preferable to use about a 0.1 M concentration.

Another method that can be used to induce cell wall permeability to enhance DNA uptake in fungi or yeasts is to resuspend the cells in growth medium supplemented with sorbitol and carrier calf thymus DNA. Glass beads are then added to the supplemented medium, and the mixture is vortexed at high speed for about 30 seconds. This treatment disrupts the cell walls, but may kill many of the cells.

Yet another method to prepare strains for transformation involves the preparation of protoplasts from bacterial, yeast and fungi cells. For example, fungal mycelium is a source of protoplasts, so that mycelium can be isolated from the cells. The protoplast preparations are then protected by the presence of an osmotic stabilizer in the suspending medium. These stabilizers include sorbitol, mannitol, sodium chloride, magnesium sulfate, and the like. Usually the concentration of these stabilizers varies between 0.8 M to 1.2 M.

Uptake of DNA into the host mutant strain is dependent upon the calcium ion. Generally between about 10 mM CaCl₂ and 50 mM CaCl₂ is used in the uptake solution. Besides the need for the calcium ion in the uptake solution, other items generally included are a buffering system such as TE buffer (10 mM Tris, pH 7.4; 1 mM EDTA) or 10 mM MOPS, pH 6.0 (morpholinepropanesulfonic acid) and polyethylene glycol (PEG) . The polyethylene glycol acts to fuse the cell membranes, thus permitting the contents of the mycelium to be delivered into the cytoplasm of the fungal mutant strains and the plasmid DNA is transferred to the nucleus. This fusion leaves multiple copies of the plasmid DNA tandemly integrated into the host chromosome. Generally, a high concentration of PEG is used in the uptake solution. Up to 10 volumes of 25% PEG 4000 can be used in the uptake solution. However, it is preferable to use about 4 volumes of PEG. Additives such as dimethyl sulfoxide, heparin, spermidine, potassium chloride, and the like may be added to the uptake solution to aid in the transformation process.

Usually a suspension containing the mutant cells that have been subjected to a permeability treatment or protoplasts at a density of 10 8 to 109/ml, preferably 2 x 108/ml, are used in transformation. These protoplasts or cells are added to the uptake solution, along with the desired transformant vector containing the selectable marker and α-N-galactosaminidase to form a transformation mixture. The mixture is then incubated at 4°C for a period between 10 to 30 minutes. Additional PEG may be added to the uptake solution to further enhance the uptake of the desired cDNA sequence. The PEG may be added in volumes up to 10 times the volume of the transformation mixture, preferably about 9 times. After the PEG is added, the transformation mixture is then incubated at room temperature before the addition of the sorbitol and CaCl₂ solution. The protoplast suspension is then added to molten aliquots of a growth medium.

Yet another method for transforming yeast and fungi protoplasts involves a method known as electroporation. Electroporation involves the application of an electrical pulse to a suspension of protoplasts or cells in a buffer containing DNA. The mechanism of uptake involves formation of transient pores in the plasma membrane allowing access to the cytoplasm by large DNA molecules. This procedure is described in detail by Ward, et al., in "Transformation of Aspergillus awamori and A_^. niger by Electroporation," Experimental Mycology, Vol. 13 (1989), pp. 289- 293.

After transformation with the above plasmids, the transformants are then selected and further grown in the proper culture medium which medium is chosen to promote the growth of the positive transformants. Any culture medium known in the art can be used. However, it is preferable to use a chemically-defined minimal medium which either (a) contains a drug which selects for an antibiotic resistance marker on the transforming DNA plasmid, or (b) lacks the essential nutrient whose synthesis is encoded by a gene on the transforming DNA.

Selected transformants are subsequently cultured in a nutrient medium to promote cell growth and production of α-N- galactosaminidase. After culturing the transformants for an appropriate period of time, the α-N-acetylgalactosaminidase is then purified from the culture media. Since yeast, fungi and bacteria differ in their cellular structure and production, after genetically engineering these strains to express α-N- acetylgalactosaminidase, different purification schemes are needed. For instance, most yeast and fungi express α-N- acetylgalactosaminidase extracellularly and, therefore, purification from the fermentation broth will differ from the procedure when E. coli is used. In the instance when the microorganism E. coli is used, the enzyme is produced in the periplasmic space of E. coli and, therefore, to purify it therefrom, the microorganism is subjected to periplasmic shock treatment to release the enzyme. These techniques are known to those skilled in the art.

Prior to purification from the culture medium, cellular debris is first removed by either centrifugation or filtration. It is preferable to filter the culture medium derived from fungi. The culture medium may then be further concentrated prior to purification of the enzyme therefrom. Known concentration steps such as an amicon filter can be used.

The α-N-acetylgalactosaminidase is purified from the concentrated culture media using basic ion exchange chromatography techniques such as DEAE chromatography, hydroxyapatite chromatography, carboxymethyl cellulose chromatography and the like. Size exclusion chromatography, gel filtration, chromatography using ampholytic displacement, SP-sephadex or butyl agarose may also be used in the purification scheme. Besides, chromatographical purification techniques, salt fractionation and PEG extraction can also be used. When employing this purification technique, α-N-acetylgalacosaminidase is first precipitated using ammonium sulphate or sodium chloride. PEG extraction with water then separates the enzyme into the PEG phase which is then filtered and further purified.

Once the purified enzyme has been obtained it can be used in a process to treat erythrocytes such that A, B and AB blood types can be conveniently converted into O type blood. Blood as it is usually obtained from the donor is mixed with sodium citrate, potassium oxalate or EDTA to prevent coagulation. Usually about 75 mis of anticoagulant is used per 425 mis of blood (425 mis s 1 unit) . Typically after a unit of blood is obtained from a donor, the plasma is removed by centrifugation. Thereafter, the white blood cells are removed by aspiration in an isotonic saline solution, leaving the red blood cells.

In the present invention, the remaining red blood cells can then be equilibrated to a pH of <6, preferably 5.6 to 5.8, using a citrate-phosphate buffer. The equilibrium step provides an optimal pH for enzyme function. The citrate-phosphate buffer contains between 0.02 M to 0.05 M citric acid in addition to dibasic sodium phosphate in a concentration of between 0.05 M to 0.10 M and 0.15 M sodium chloride. The equilibrium is normally effected by suspending the erythrocytes in the buffer solution for a period of at least 5 minutes and preferably no longer than 15 minutes. The old buffer is then removed and fresh buffer is added. This procedure may be repeated at least one more time.

After equilibrating the erythrocytes as described above, α-N- acetylgalactosaminidase is added to the erythrocytes and incubated at 22°C to 37°C, preferably 25°C to 27°C for about 24 hours. The time needed for this conversion varies depending on the amount of enzyme used, as well as the specific activity of the enzyme and can be reduced by increasing the units of enzyme or increasing the specific activity of the enzyme. Generally about 6,000 to 125,000 units of enzyme is used per 300 to 500 is of erythrocytes. By enzyme units is meant that amount of enzyme which will catalyze the transformation of 1 micromole of substrate per minute under standard conditions of temperature, optimal pH and optimal substrate concentration.

The enzyme can be in the form of a free enzyme or bound to a soluble or insoluble support such as dextran, polyethylene glycol, agarose, cellulose and the like.

Following enzyme treatment, the enzyme is then removed from the erythrocytes and the erythrocytes are then re-equilibrated to a pH of 7.2 to 7.4 by washing the same with a buffer and allowing the erythrocytes to remain in contact with the buffer for at least 30 minutes following the last wash. The washing buffer is used to adjust the pH of the erythrocytes to physiological pH and to remove the enzyme. It is preferable to wash the erythrocytes with a buffer such as phosphate buffered saline which contains a concentration of 0.01 M potassium phosphate in the ratio of seven parts dibasic salt to 3 parts monobasic and a concentration of 0.9% sodium chloride. The washing step may be effected as many times as needed to remove the α-N-acetylgalactosaminidase and is generally carried out at 20°C to 26°C. An alternative to washing the enzyme from eyrthrocytes entails column chromatography whereby the erythrocytes are first equilibrated and placed over a DEAE column to remove α-N- acetylgalactosaminidase.

Thereafter, the cells in the H-antigen form can be used for transfusion purposes after being appropriately diluted with a physiologically acceptable medium. These medium are well known in the art and can include, for instance, sterile isotonic saline consisting of 0.9% sodium chloride and sterile isotonic solution containing 0.2% dextrose. Generally, the concentration of the cells in the medium is between 40% and 70%; preferably between 40% and 45%.

Also contemplated by the present invention is the modification of α-N-acetylgalactosaminidase to ensure thorough removal of the enzyme from erythrocyte suspensions. α-N-acetylgalactosaminidase can be immobilized by covalently coupling the enzyme to inert supports, adsorbed to supports or trapped within matrices which permit the rapid removal of α-N-acetylgalactosaminidase after being subjected to erythrocytes. Furthermore, α-N- acetylgalactosaminidase can be manipulated via gene fusion techniques to fuse a cellulose-binding domain of Cellulomonas fimi to the N- or C- terminus of the enzyme. This fusion results in a hybrid protein which has α-N-acetylgalactosaminidase activity and which binds to cellulose. The enzyme can then be easily removed from the erythrocyte suspension by using a variety of forms of cellulose such as paper, powder, cotton, membranes and the like. The enzyme can then be eluted from the cellulose matrix under mild conditions by using distilled water. The use of the cellulose-binding domain and procedures utilized therein to create enzyme immobilization are described by Ong, et al., in "Enzyme Immobilization Using The Cellulose-Binding Domain of A Cellulomonas fimi Exoglucanase," Bio/Technology. Vol. 7 pp. 604- 607 (1989); Ong, et al., "The Cellulose Binding Domains of Cellulases; Tools for Biotechnology," TIBTECH. Vol. 7 (1989) pp. 239-243.

Moreover, kits containing α-N-acetylgalactosaminidase in effective amounts to convert a unit of whole blood having A, B or AB blood types are also contemplated by the present invention. Generally such a kit contains lyophilized or concentrated α-N- acetylgalactosaminidase in an appropriate amount, as well as an appropriate carrier solution which may comprise dilution and washing buffers. Additional excipients as known to those skilled in the art may be added to said kit components. The enzyme in this kit may be modified according to the procedures set forth herein. In order to further illustrate the present invention and advantages thereof, the following specific examples are given, it being understood that the same are intended only as illustrative and should not be interpreted as a limitation of any kind.

Experimental

Example 1: Amplification of cDNA Libraries cDNA libraries constructed in expression vector λgtll were purchased from Clontech Laboratories, Inc. (Palo Alto, California) . Human bone marrow, human breast, human breast carcinoma, human liver and human placenta were then amplified in E. coli strain Y1090hsdi? using the protocol described in Maniatis, et al., Molecular Cloning, A Laboratory Manual (1982, ed) p.294. The amplification procedure consisted of mixing 10 bacteriophages from each library with 600 μl of plating bacteria of E coli strain Y1090λsdi? (ATCC 37197) . After incubating the mixture for 20 minutes at 37°C, the infected culture was plated on a petri dish containing NZCYM agar. The NZCYM agar was made by diluting in 950 ml of deionized water the following compounds: 10 grams of casein hydrolysate (enzymatic) , 5 grams NaCl, 5 grams bact-yeast extract, 1 gram casamino acids and 2 grams MgS0₄-7H₂0 and the pH was adjusted to 7.0 with 5 N NaOH prior to adjusting the volume to 1 liter with deionized water. The agar was then sterilized prior to use. The plates were incubated for 12 hours at 42°C to prevent the formation of lysogens. The plates were then overlayed with 15 ml of SM buffer (5.8 grams NaCl, 2 grams MgS0₄-7H₂0, 1 M Tris chloride, pH 7.5), 5 ml 2% gelatin and deionized water to 1 liter; the buffer was then sterilized prior to use for 2 hours at room temperature and were harvested by the removal of cellular debris by centrifugation at 7,000 x g for 30 minutes at 4°C. The samples were either stored at 4°C in aliquots containing 20-30 μl of chloroform or further titrated on LB agar (10 grams bacto-tyrptone, 5 grams bacto-yeast extract and 10 grams NaCl per liter deionized water, pH 7.0; sterilized prior to use) plates using 3 ml of top agarose containing 40 μl of stock solution of X-gal at a concentration of 20 mg/ml in dimethylformamide and 4 μl of IPTG (isopropylthio- j8-D-galactoside) at a concentration of 200 mg/ml.

Example 2: Total DNA Isolation The bacteriophage DNA was purified by precipitating the phage particles, followed by phenol/chloroform extraction according to the methods described by Maniatis supra.

The bacteriophage suspension was then transferred to a tube and a 0.5 M stock solution of EDTA (pH 8.0) was added until the final concentration in each tube was 20 mM EDTA. Proteinase K was then added such that the final concentration was equivalent to 50 μg/ml. A 10% w/v stock solution of SDS in water was added to each tube such that the final SDS concentration was equivalent to 0.1%. Each tube was inverted several times and incubated at 56°C for about 1 hour. The digestion was then cooled to room temperature and an equal volume of phenol equilibrated with 50 mM Tris (pH 8.0) was added to each digestion tube. The mixture was then inverted several times until a complete emulsion was formed. The phases were then separated by centrifugation at 3,000 x g for 5 minutes at room temperature and the aqueous phase was then transferred to another tube. The aqueous phase was further extracted with a 50:50 mixture of chloroform to phenol and the aqueous phase was recovered and further extracted with an equal volume of chloroform.

To the aqueous phase was added a 3 M stock solution of sodium acetate (pH 7.0) to achieve a final concentration of 0.3 M. The solution was then mixed well and 2 volumes of ethanol was added, the solution was mixed and then frozen on dry ice until it was solid. The tube was then centrifuged at 12,000 x g for 5 minutes at 4°C and the supernatant was discarded leaving a pellet of DNA, which was permitted to then dry under vacuo prior to redissolving the DNA in TE buffer (10 mM Tris-Cl; 1 mM EDTA, pH 7.6).

Example 3: Screening of Libraries Containing Sequences Specific for α-N-Acetylgalactosaminidase cDNA Clones

A 10 μl aliquot was taken from each of the bacteriophage purified total DNA libraries (human bone marrow, human breast, human breast carcinoma, human liver and human placenta) . The polymerase chain reaction (PCR) was used to amplify sequences specific for α-N-acetylgalactosaminidase. Synthetic oligonucleotide primers were synthesized on a

MilliGen/BioSearch® model 8700 DNA synthesizer and were designed based on the cDNA sequence published by Yamauchi, et al., in "Molecular Cloning of Two Species of cDNAs For Human α-N-Acetylgalactosamindase and Expression in Mammalian Cells," Biochemical and Biophysical Research Communications. Vol. 170, pp. 231-237 (1990).

Each PCR reaction mixture contained a 10 μl DNA aliquot (0.25 μg/μl template DNA) taken from the above-described libraries, 53.5 μl aliquot of double distilled sterile water, [10X] reaction buffer from the GeneAmp kit (Perkin Elmer Cetus) , 16.0 μl dNTP mix, 1.25 mM each dNTP, 5 μl of forward primer at a concentration of 20 μM, 5.0 μl of reverse primer at a concentration of 20 μM and 0.5 μl Amplitaq® DNA polymerase. The total reaction mixture was equivalent to 100 μl. The following forward primer was used in the PCR reaction, which was synthesized as described above:

5' GAGCAATCCCGGGCCCAGATGGCCCTGTGG 3'. SEQ ID No:l.

The following reverse primer having the following oligonucleotide sequence was also used in the PCR reaction:

5' CTGTCTAGACCCAGCTCCTCACTGCTGGGA 3'. SEQ ID NO:2.

100 μl of sterile mineral oil was added to the top of each reaction mixture to prevent evaporation during temperature cycling. The reactions were incubated in a Perkin Elmer Cetus DNA Thermal Cycler® programmed to the following settings: (1) 1 cycle at 95°C for 5 minutes, 50°C for 3 minutes, 65°C for 5 minutes;

(2) 35 cycles at 95°C for 2 minutes, 50°C for 2 minutes, 65°C for 7 minutes;

(3) 1 soak cycle at 4°C.

The aqueous phase of each reaction was then extracted with Chloroform to remove traces of the mineral oil. The reaction products were then analyzed by agarose gel electrophoresis. Results of the electrophoresis indicated that only one reaction, that which used the placental cDNA library as the template produced an amplified fragment. This amplified fragment had approximately 466 base pairs.

Example 4: Subcloning of the 466 Base Pair Fragment The 466 base pair fragment of α-N-acetylgalactosaminidase was excised from the agarose gel and digested with Xball and Smal. These sites were contained within the synthetic primers described above. The digested DNA fragment was then subcloned into an M13mpl8 vector cut with the same restriction enzymes. DNA sequencing analysis by the Sanger dideoxy method of the subcloned fragment showed that it was indeed a portion of the α-N-acetylgalactosaminidase cDNA.

Example 5: Isolation of Full Length cDNA Clones Encoding α-N-Acetylgalactosaminidase The purified 466 base pair fragment from the amplified PCR reaction was then used as a probe to screen the placenta cDNA library using the hybridization methods described in Davis, et al.. Advanced Bacterial Genetics (1980) pp.162-165. The fragment was first radiolabeled by nick translation by the methods described in Maniatis, et al., supra.

λ plaques from the human placenta cDNA library were grown on

LB plates with 10 -105 plaques for 1 day. The resulting plates were then cooled at 4°C between 15 minutes to several hours to harden the agar. 82 mm dry nitrocellulose filters (HAWP Millipore) were then placed on the top of the agar cells and were checked to ensure that no air bubbles were formed between the agar and the filters. The filters were left on the plates for 5 minutes to permit adsorption. During the adsorption period, the filters were marked for orientation. The filters were then removed from the plate and treated with 0.5 M NaOH and 1.5 M NaCl for 20 seconds to 5 minutes. The filters were then further treated in 0.5 M Tris (pH 7.5) and 1.5 M NaCl for 20 seconds to 5 minutes. The filters were then further treated with 2X SSPE, blotted and baked at 80°C in a vacuum for 1 1/2 hours.

The filters were then probed using the labeled 466 base pair fragment of α-N-acetylgalactosaminidase. The following hybridization conditions were used: 5X SSPE and 0.3% SDS hybridization buffer, 200 ug/ml denatured salmon sperm DNA, 50% formamide and 10 cpm/ml of the probe. The samples were hybridized at 42°C overnight and washed once at 42°C with 0.2X SSPE and 0.1% SDS for 15 minutes; twice with 0.2X SSPE for 15 minutes and air dried before exposure on Kodak XAR-2 at -70°C.

From approximately 500,000 plaques that were screened, 20 plaques were found to hybridize strongly with the 466 base pair labeled fragment of α-N-acetylgalactosaminidase.

The positive plaques were then purified twice by single-plaque isolation and amplified using the PCR method described above. Eleven of these plaques were rescreened by hybridization using a synthetic oligonucleotide probe which corresponded to sequences upstream of the coding region (Yamauchi, et al., 1990 supra) . The following synthetic oligonucleotide was synthesized and radiolabeled with T4 polynucleotide kinase v32 PdATP and used in this secondary screening:

5' AGACCAGATCTGGTCAGGTCCTCGGAACGT 3' . SEQ ID NO:3.

The hybridization was performed at room temperature overnight in a hybridization buffer which consisted of: 900 mM NaCl, 90 mM Tris-HCI, pH 7.5, 6 mM EDTA, 0.5% NP-40, IX Denhardt's solution, 1 mM sodium pyrophosphate, 1 mM sodium phosphate, 0.1 mM ATP and 200 μg/ml yeast RNA (all concentrations set forth for the above are final concentrations) . After hybridization, the filters were washed twice using 0.5x SSPE and 0.1% SDS at 42°C for 15 minutes; and twice in 0.5x SSC without SDS and the filters were allowed to dry prior to exposing to X-ray film. Seven of the eleven plaques hybridized to this probe.

DNA sequence analysis of the insert subcloned into pUCllδ, using EcoRI from one of these isolates confirmed that a full length α-N-acetylgalactosaminidase cDNA had been cloned and the sequence is set forth in FIG. 2 (SEQ ID NO:4).

Example 6: Construction of Expression Vector pGPT-pyrG:α-gal The construction of the expression vector for filamentous fungi is illustrated in FIG. 3. Plasmid pUC218:α-gal is constructed by ligating the entire α-N-acetylgalactosaminidase coding region into pUC218 after digestion of the vector and insert with Bglll and EcoRI. The DNA is then made single- stranded by infecting M13K07 strains with the pUC218 vector and the single-stranded template is purified therefrom. Basically, the purification procedure involves centrifuging the culture medium for 5 minutes to pellet the cells and the supernatant is then removed and is poured into tubes containing 200 μl 20% (w/v) PEG-6000, 2.5 M NaCl. The supernatant containing the PEG and NaCl is incubated for 15 minutes and then centrifuged once again. The supernatant is aspirated off and neutralized phenol is used to extract the DNA. The phenol is removed by one extraction with 0.5 ml diethylether, 10 μl 3 M sodium acetate, pH 5.5 and 250 μl ethanol. The absorbance of the DNA is measured at 260 nm and the concentration of the template is adjusted to 0.5 μg/ml with TE buffer (10:0.1). The single stranded template is then subjected to site- directed mutagenesis to introduce a BssHII site at the 5' end of the α-N-acetylgalactosaminidase coding region. Basically, this procedure involves the synthesis of an oligonucleotide which is complementary to part of the DNA template but contains an internal mismatch to direct the required mutation, insertion or deletion. The mutagenic primer is annealed with the single-stranded Ml3 template and extended with DNA polymerase I using deoxynucleoside triphosphates. The mutagenic primer used for site-directed mutagenesis has the sequence:

5'ATGTGGCCCCAGGTGCGCGCGCTGGACAATGGGCT 3'. SEQ ID NO:5.

This primer is then annealed with the single-stranded M13 template by reacting 2 μl kinased mutagenic primer (2.5 pmol/μl) , 2 μl template, 1 μl lOx TM buffer and water to a total volume of 10 μl in a tube placed in a beaker of water at 80°C for about 30 minutes. The reaction is then cooled to room temperature. 10 μl of the annealed mixture (set forth above) is placed on ice and 1 μl 10X TM buffer, 1 μl 5 mMrATP, 1 μl 5 mM dNTPs, 1 μl of 100 mM DTT and water to a total volume of 20 μl is added. 10 Units of T4 DNA ligase and 1 unit of Klenow fragment of DNA polymerase I is added and the mixture is incubated at 12°C to 15°C for 4-20 hours. This mixture is then diluted with 30 μl of TE buffer. Similarly, site-directed mutagenesis is also completed to introduce a Xbal site at the 3' end using as a kinased primer:

5 ' TCCCAGCAGTGAGGATCTAGAACATGTGACAGGCTG 3'. SEQ ID NO:6.

following the procedure set forth above. The DNA having the appropriate restriction sites introduced therein is then further digested with BssHII and Xbal and the fragment is subjected to electrophoresis to purify the α-N- acetylgalactosaminidase coding sequence fragment.

Plasmid pGA5', which includes the glaA promotor, glaA coding region and the glaA terminator for Aspergillus niger, is then digested with MluJ and BssHII. A 594 base pair glaA fragment which includes a portion of the promotor, the entire signal peptide and propeptide is then purified by electrophoresis.

Plasmid pGPT-pyrGl which includes the glaA promoter, glaA terminator and pyrG marker for Aspergillus niger is digested with Mlul and Xbal and the large vector segment including the selectable marker, promotor and terminator is purified by electrophoresis.

The pGPT-pyrGl plasmid is constructed according to the procedure described by Berka, et al., "Foreign Protein Secretion in Aspergillus:Current Status," Advances in Gene Technology: Protein Engineering and Production . Proceedings of the 1988 Miami Bio/Technology Winter Symposium, ICSU Press, Vol. 8 (1988), which is incorporated herein by reference. Basically, the expression vector pGPT-pyrGl is constructed using a pBR322 origin of replication and the ampicillin resistance gene; the pyrG gene of Aspergillus nidulans which complements pyrG mutants of Aspergillus niger. Aspergillus awamori as well as Aspergillus nidulans; the Aspergillus awamori glaA promotor; and the Aspergillus niger glaA terminator. The introduction of an unique Bglll site at the end of the promotor and an unique Xbal site just preceding the termination region allows the insertion and expression of foreign genes between these elements.

The α-galactosidase cDNA fragment and the 594 base pair glaA fragment including a portion of the promotor, the entire signal peptide and propeptide is ligated into the digested pGPT-pyrGl plasmid using T4 DNA ligase. The resulting vector pGPT-pyrGα-gal can be used to transform a host strain such as Aspergillus awamori ApepA pyrG. This particular host strain is selected because it produces abundant glucoamylase, it contains a chromosomal deletion in the gene encoding aspergillopepsin A (pepA) as illustrated by Berka, et al., in "Molecular Cloning and Deletion of the Gene Encoding aspergillopepsin A from Aspergillus awamori," Gene 86, pp. 153-162 (1990) and it is an uridine-requiring auxotroph which can be transformed to prototrophy with vectors containing the selectable pyrG marker gene. Example 7: Construction of Expression Vector YEpsecl:α-gal The construction of expression vector YEpsecl:α-gal is illustrated in FIG. 4. Basically this vector is constructed by using the starting pUC218:α-gal plasmid with α-gal cDNA inserted into pUC218 after digestion with the restriction enzymes of Bglll and EcoRI. The plasmid is then made single- stranded by infecting M13K07 with pUC2iδ and purifying the single stranded template according to the procedure set forth in Example 6. The single stranded template is subjected to site-directed mutagenesis, following the procedure set forth in Example 6, to introduce a StuI site at the 5'end of α-N- acetylgalatocaminidase coding region using a kinased primer having the following sequence:

5' ATGTGGCCCAGGTGCAGGCCTTGGACAATGGGCTCC 3'. SEQ ID NO:7.

Similarly site-directed mutagenesis is used to introduce a Hindlll site at the 3' end of the α-N-acetylgalactosaminidase coding region using a kinased primer having the following sequence:

5'TGTCCCAGCAGTGAGAAGCTTGGACATGTACAGGCT 3'. SEQ ID NO:8.

The DNA having the appropriate restriction sites incorporated therein is then digested with StuI plus Hindlll and the cDNA fragment containing the α-gal coding sequence is then purified by electrophoresis according to the procedures of Maniatis, et al. , supra. Plasmid YEpsecl is constructed according to the procedure described by Baldari, et al., "A novel leader peptide which allows efficient secretion on a fragment of human interleukin IB in Saccharomvces cerevisiae," The EMBO Journal. Vol. 6, pp. 229-234 (1987) , which is incorporated herein by reference. The secretion vector YEpsecl is derived from the yeast expression vector pEMBLyex2, which contains two blocks of yeast elements in addition to bacterial sequences necessary for selection and replication of the plasmid in E coli. The first, which determines episomal replication and copy number is derived from plasmid pJBD219 described by Beggs, J.D. in Nature, 275, pp. 104-109 (1978) . Plasmid pJBD219 comprises a 3220-bp NdeJ-StuJ fragment spanning the Ieu2-d, 2μm STB and ori portions of pJBD219. This fragment also includes a small part of the 3' end of the FLP gene of the 2-μm plasmid, which provides a transcription termination and polyadenylation signals to sequence cloning in the polylinker at about 205 bp downstream of the polylinker Hindlll site as described by Sutton, et al., Mol. Cell. Biol. 5, pp. 2770-2780 (1985). The second element is the Hindlll-BamHI fragment from plasmid G2, described by Guarente in Methods in Enzvmology, Vol. 101, pp.. 181-191 (1983) , that carries the ura3 (which is analogous to PyrG) gene and signals which induce transcription into the polylinker during growth on galactose as the carbon source. These transcription signals derive from a hybrid promotor of a fusion between the GAL upstream activation sequence (UAS_G) and the 5' non-translated leader of the yeast CYCI gene, up to position -4 from the ATG translation initiation codon. Translation starts at the first ATG of a fragment inserted in the polylinker. The last step of construction of YEpsecl is the insertion of a synthetic oligonucleotide between the SstJ and Kpnl sites of pEMBLyex2 to give YEpsecl.

Yepsecl is then digested with Smal and Hindlll and thereafter the large vector segment is further purified via electrophoresis. The α-gal cDNA fragment containing StuI and Hindlll sites is then ligated into the purified large vector fragment of YEpsecl using T4 DNA ligase. Plasmid YEpsecl:α- gal is then used to transform various strains of yeast. The host used in the transformation of this expression vector is Saccharomyces cerevisiae S150-2B(leu2-3 leu2-112 ura3-52 trpl- 289 his3-Al cir-, which contains an auxotrophic mutation ura3 .

Example 8: Construction of Expression Vector pNH-STα-gall FIG. 5 illustrates the construction of expression vector pNH- STα-gall. Three cloning vectors are manipulated in such a way as to create the expression vector pNH-STα-gall. The first vector is pUC218:α-gal which has the entire coding sequence for α-N-acetylgalactosaminidase ligated into pUC218 at EcoRI and Bglll restriction sites. The DNA is made single stranded by infecting pUC218 with M1307 and the single stranded template is purified according to the procedure set forth in Example 6. The single stranded template is then subjected to site directed mutagenesis, following the procedure set forth in Example 6, in order to introduce a Nsil site at the 5' end using a kinased primer having the sequence:

5' GACATGTGGCCCAGGATGCATTGCTGGACAATGGGC 3'. SEQ ID No:9.

Further site directed mutagenesis is used to introduce a SphI site at the 3 ' end using a kinased primer having the sequence:

5'CCAGCAGTGAGGAGCATGCACATGTGACAGGCTG 3'. SEQ ID No:10.

The DNA having the appropriate restriction sites incorporated therein is then subjected to digestion with NsiJ and SphI and the α-N-acetylgalactosaminidase fragment is then purified by electrophoretic techniques.

A synthetic oligonucleotide encoding E . coli heat-stable enterotoxin ST II ribosome binding site and signal sequence codons described by Picken, et al., in "Nucleotide Sequence of the Gene for Heat-Stable Enterotoxin II of Escherichia coli," Infection and Immunology, Vol. 42, No. 1 (1983) pp. 269-275, is phosphorylated using T4 kinase and annealed. This fragment is then ligated with T4 DNA ligase into bacteriophage M13mpl8 which has been digested with Xbal and Hindlll to form mplδ- STII. This vector is then digested with Xbal and NsiJ and the 95 base pair fragment containing the ST II signal sequence is purified via electrophoretic techniques. Plasmid pNHlδa is constructed following the procedures outlined by Hasan, et al., in "Control of cloned gene expression by promoter inversion in vivo: construction of improved vectors with a multiple cloning site and the ptac promotor," Gene 56, pp. 145-151 (1987), which is incorporated herein by reference. Generally, construction of the plasmid involved the ligation of the BgrlJ-'ApR-attP-nutL-plac-ptac- attB-N-MCS-Bgrl JJ module from pNH8a with the BajntfJ-ori-Ap^R'- BglJ fragment of pNH55B. Plasmid pNH55B is a derivative of pNH3. This plasmid is then propagated in the λ repressor- producing lysogen C600(λcIδ57cro27S7) .

pNHlδa is then digested with Xball and Sp J and the large fragment is then purified by electrophoresis. Fragments containing the ST II signal sequence codons and the α-N- acetylgalactosaminidase cDNA are ligated into the large pNHlδa purified fragment with T4 DNA ligase to form pNH-STα-gall. Escherichia coli D1210HP (available commercially from Stragene, LaJolla, California) is used as the host for transformation of the plasmid derived from pNH18a. A short heat pulse at 42°C for 10 minutes induces the expression of the int gene product from a λint+κis, KIL (ΔSalI-Xhol ) AΑl prophage in this strain which in turn promotes inversion of the ptac-plac tandem promoter segment on the pNHlδa vector to the "ON" configuration.

Example 9: Isolation of Protoplasts Fungal mycelium is obtained by inoculating 100 ml of YEG (0.5% yeast extract, 2% glucose) in a 500 ml falsk with about 5 x 10⁷ Aspergillus awamori ΔpepApyrG spores. The flask is then incubated at 37°C with shaking for about 16 hours. The mycelium are harvested by centrifugation at 2,750 x g. The harvested mycelium are further washed in a 1.2 M sorbitol solution and resuspended in 5 mg/ml of Novozym® 234 solution, a multicomponent enzyme system containing 1,3-α-glucanase, l,3-3-glucanase, laminarinase, xylanase, chitinase and protease (commercially available from Novo Biolabs, Danbury, Ct.); 5 mg/ml MgS0₄»7H₂0; 0.5 mg/ml bovine serum albumin; 1.2 M sorbitol. The protoplasts are removed from the cellular debris by filtration through Miracloth (Calbiochem Corp.), collected by centrifugation at 2,000 x g and used immediately for transformation as described below.

TRANSFORMATION PROTOCOLS

Transformation of pGPT-pyrG:α-gal

The protoplasts isolated above are then washed twice in 0.7 M KC1 and once in electroporation buffer containing 7 mM sodium phosphate buffer, pH 7.2; 1 mM MgS0₄; 1.2 M sorbitol. 2 x 10 ⁷ protoplasts are resuspended in O.δ ml electroporation buffer. Electroporation is performed in Bio-Rad electroporation cuvettes having an interelectrode distance of 0.4 cm. A 10- minute incubation on ice is allowed before and after delivery of the pulse, the DNA being added immediately prior to the pulse. Following transformation, according to the procedures of Maniatis, et al., protoplasts are then added to molten Aspergillus minimal medium described by Rowlands and Turner in Mol. Gen. Genet.. 126:201-216 (1973) with 2% agar and are plated onto the same medium. Transformed colonies can be observed following 2-5 days incubation at 37°C.

Transformation of Yepsecl:αgal

Yeast spheroplasts are prepared as described by Hutchinson and Hartwell in J. Bacteriology. 94 pp. 1697-1705 (1967) . A fresh logarithmic phase culture (80 ml; 2 x 10 cells per ml) is concentrated to 1/lOth the volume by centrifugation and treated with 1% Glusulase (Endo Laboratories) in 1 M sorbitol for 1 hour at 30°C. Spheroplasts are then washed three times with 1 M sorbitol and resuspended in 0.5 ml of 1 M sorbitol; 10 mM Tris-HCl; 10 mM CaCl₂, pH 7.5. Plasmid DNA is added to a final concentration of 10-20 μg/ml and incubated for 5 minutes at room temperature. 5 ml of 40% polyethylene glycol 4000 in 10 mM Tris-HCl; 10 mM CaCl₂, pH 7.5 is added as described by van Soligen and Plaat in J. Bacteriology. 130 pp. 946-947 (1977) . After 10 minutes, the spheroplasts are sedimented by centrifugation and resuspended in 5 ml of the sorbitol-Tris- CaCl₂ mixture; 0.2 ml aliquots are added to 10 ml of regeneration agar and poured on minimal agar plates. The regeneration agar that can be used is Difco yeast nitrogen base without amino acids, supplemented with 1 M sorbitol, 2% glucose, 2% YEPD and 3% agar. Transformation of pNH-STα-gall

Preparation of transformation-competent E^_ coli cells and transformation protocols are described in Maniatis, et al., supra.

Example 10: Isolation of α-N-acetylgalactosaminidase From Fermentation Media

For yeast and fungal transformants, cells are removed by filtration and/or centrifugation and α-N- acetylgalactosaminidase purified from the culture filtrate/supernatant using the techniques described in Kusiak, et al., supra. or Dean, et al., supra. For E_j_ coli transformants, the cells were subjected to osmotic shock to release the enzymes in the periplasmic space. Following centrifugation to remove cells, α-N-galactosaminidase was isolated from the supernatant fraction, as described further below.

Isolation from E. coli

The fermentation broth is centrifuged to remove extraneous cellular debris and the pellet is placed on ice after the supernatant is removed. All of the subsequent steps are performed at 4°C. The pellets are resuspended in 0.15 ml of an ice-cold solution containing 20% sucrose, 10 mM Tris-Hcl (pH 7.5). 5 λl of 0.5 M EDTA (pH 8.0) is added and incubation on ice is then continued for 10 minutes. The cells are then centrifuged for 5 minutes in the cold and the supernatant is then quickly removed and the pellet is resuspended rapidly with vigorous agitation in 0.1 ml cold distilled water. The mixture is incubated on ice for 10 minutes and then centrifuged again for 5 minutes. The supernatant is then removed and subjected to further purification procedures as outlined for yeasts and fungi set forth below.

The fermentation broth for yeast and fungi is centrifuged and filtered to remove extraneous cell debris. The supernatant is then further purified by either using Concavalin A-Sepharose 4B and carboxymethyl cellulose chromatography as described by Schram, et al., Biochim. Biophvs. Acta. 482, pp. 125-127 (1977) or by hydroxyapatite chromatography and ampholyte displacement chromatography on DEAE-cellulose, as described by Dean, et al., in "Purification of α-Galactosidase A and Its Enzymatic Properties with Glycolipid and Oligosaccharide Substrates," Journal of Biological Chemistry. Vol. 254, No. 20, pp. 9994-10,000 (1979). Basically, this procedure involves concentrating the supernatant in an Amicon Model 52 ultrafiltration cell equipped with a PM-10 membrane. The concentrated solution is then dialyzed versus two changes of 1 mM sodium phosphate buffer, pH 6.5 for 24 hours. The dialyzed enzyme solution is then applied to a hydroxyapatite column that had been equilibrated with 1 mM sodium phosphate buffer, pH 6.5 at 4°C. The enzyme is eluted with 10 mM sodium phosphate buffer, pH 6.5.

The eluted fractions are then applied to a DEAE-cellulose column that had been previously equilibrated with 10 mM phosphate buffer, pH 5.0. The column is then washed two times with distilled water to remove the buffer. The enzyme is then eluted from the column with about 15 ml of a carrier ampholyte solution consisting of 1.0 ml of carrier ampholytes (pH 3 to 5) diluted 1:15 with distilled water. The carrier ampholytes were obtained from LKB Produkter AB. Following application of the ampholyte solution, the column is then washed with distilled water. 0.5 ml fractions are then collected and those containing α-N-acetylgalactosaminidase activity are pooled and then ampholytes removed by Sephadex G-150 column chromatography as previously described.

To determine the presence of α-N-acetylgalactosaminidase transformants also recovered from the purification process, glycosidase activities are monitored at each step of the purification process by the following procedure. The column fractions are tested by incubation of the enzyme preparations in a reaction mixture containing 0.5% bovine serum albumin, 100 mM sodium acetate, pH 4.6, and 12 mM p-nitrophenyl-α- galactoside. The final volume of the reaction mixture is 0.5 ml. The reaction mixture is incubated for 37°C for a period between 10 to 60 minutes and the reaction is stopped by adding 1.0 ml of 0.3 M glycine/NaOH, pH 10.6. The liberated p_- nitrophenol was determined spectrophotometrically at 405 nm

6 -1 - using a molar extinction coefficient of 18. 5 x 10 M 'cm .

Alternatively, the positive transformants can be screened with purified antibodies prepared by injecting α-N- acetylgalactosaminidase into rabbits and purifying the antibody obtained from a blood sample from the rabbit. The transformants are then Western blotted. 125I-labelled protein A is added to the cells, and an autoradiograph is used to detect the positive signals indicating that α-N- acetylgalactosaminidase is being expressed.

Example 11: Enzymatic Conversion of Red Cells Whole blood is obtained containing erythrocytes of the B antigen type. The blood is then centrifuged and the plasma is then removed from the whole cells. The white cells are then removed by aspiration in an isotonic saline solution. The remaining red blood cells are then further washed in isotonic saline solution and suspended in isotonic phosphate-citrate buffer, pH 7.3.

100 ml of packed B erythrocytes are suspended in isotonic phosphate-citrate buffer, pH 5.6-5.8 and 16,000 units containing α-N-acetylgalactosaminidase diluted in 1900 mis of the same buffer are mixed and incubated at 26°C with gentle mixing for 12-24 hours.

The erythrocyte suspension is then washed in phosphate buffered saline containing 0.01 M potassium phosphate pH 7.2 in the ratio of seven parts dibasic salt to 3 parts monobasic and a concentration of 0.9% sodium chloride. The washing is repeated at least five times at 25°C. Hemagglutination assays should confirm that the erythrocytes are in H-antigen form and can be further used for transfusion therapy.

While the invention has been described in terms of various preferred embodiments, the skilled artisan will appreciate that various modifications, substitutions, omissions, and changes may be made without departing from the scope thereof. Accordingly, it is intended that the scope of the present invention be limited solely by the scope of the following claims, including equivalents thereof.

SEQUENCE LISTING

(1) GENERAL INFORMATION:

(i) APPLICANT: GENENCOR INTERNATIONAL, INC.

(ii) TITLE OF INVENTION: Methods For Converting A, AB, and B Blood Types to 0 Blood Type

(iii) NUMBER OF SEQUENCES: 10

(iv) CORRESPONDENCE ADDRESS:

(A) ADDRESSEE: Genencor International, Inc.

(B) STREET: 180 Kimball Way

(C) CITY: So. San Francisco

(D) STATE: CA

(E) COUNTRY: USA

(F) ZIP: 94080

(v) COMPUTER READABLE FORM:

(A) MEDIUM TYPE: Floppy disk

(B) COMPUTER: IBM PC compatible

(C) OPERATING SYSTEM: PC-DOS/MS-DOS

(D) SOFTWARE: Patentin Release #1.0, Version #1.25

(vi) CURRENT APPLICATION DATA:

(A) APPLICATION NUMBER: US 07/977,945

(B) FILING DATE: 18-NOV-1992

(C) CLASSIFICATION:

(viii) ATTORNEY/AGENT INFORMATION:

(A) NAME: Horn, Margaret A

(B) REGISTRATION NUMBER: 33,401

(C) REFERENCE/DOCKET NUMBER: GC205-PCT

(ix) TELECOMMUNICATION INFORMATION:

(A) TELEPHONE: (415) 742-7536

(B) TELEFAX: (415) 742-7217

(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

(ii) MOLECULE TYPE: cDNA

(xi) SEQUENCE DESCRIPTION: SEQ ID NO:l: GAGCAATCCC GGGCCCAGAT GGCCCTGTGG 3

(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

(ii) MOLECULE TYPE: cDNA

(xi) SEQUENCE DESCRIPTION: SEQ ID NO:2: CTGTCTAGAC CCAGCTCCTC ACTGCTGGGA 3

(2) INFORMATION FOR SEQ ID NO:3:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 30 base pairs

(B) TYPE: nucleic acid

(C) STRANDEDNESS: single

(D) TOPOLOGY: linear

(ii) MOLECULE TYPE: cDNA

(Xi) SEQUENCE DESCRIPTION: SEQ ID NO:3: AGACCAGATC TGGTCAGGTC CTCGGAACGT 3

(2) INFORMATION FOR SEQ ID NO:4:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 1840 base pairs

(B) TYPE: nucleic acid

(C) STRANDEDNESS: single

(D) TOPOLOGY: linear

(ii) MOLECULE TYPE: cDNA

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

CAGAGCCCAA CACATACAGC TGATACACGC AGACCAGATC TGGTCAGGTC CTCGGAAGCT 6

GAGTCCAGAG CGATGCTGCT GAAGACAGTG CTCTTGCTGG GACATGTGGC CCAGGTGCTG 12

ATGCTGGACA ATGGGCTCCT GCAGACACCA CCCATGGGCT GGCTGGCCTG GGAACGCTTC 18 CGCTGCAACA TTAACTGTGA TGAGGACCCA AAGAACTGCA TAAGTGAACA GCTCTTCATG 24

GAGATGGCTG ACCGGATGGC ACAGGATGGA TGGCGGGACA TGGGCTACAC ATACCTCAAC 30

ATTGATGACT GCTGGATCGG TGGTCGCGAT GCCAGTGGCC GCCTGATGCC GGATCCCAAG 36

CGCTTCCCTC ATGGCATTCC TTTCCTGGCT GACTACGTTC ACTCCCTGGG CCTGAAGTTG 42

GGTATCTACG CGGACATGGG CAACTTCACC TGCATGGGTT ACCCAGGCAC CACACTGGAC 48

AAGGTGGTCC AGGATGCTCA GACCTTCGCC GAGTGGAAGG TAGACATGCT CAAGTGGGAT 54

GGCTGCTTCT CCACCCCCGA GGAGCGGGCC CAGGGGTACC CCAAGATGGC TGCTGCCCTG 60

AATGCCACAG GCCGCCCCAT CGCCTTCTCC TGCAGCTG5C CAGCCTATGA AGGCGGCCTC 66

CCCCCAAGGG TGAACTACAG TCTGCTGGCG GACATCTGCA ACCTCTGGCG TAACTATGAT 72

GACATCCAGG ACTCCTGGTG GAGCGTGCTC TCCATCCTGA ATTGGTTCGT GGAGCACCAG 78

GACATACTGC AGCCAGTGGC CGGCCCTGGG CACTGGAATG ACCCTGACAT GCTGCTCATT 84

GGGAACTTTG GTCTCAGCTT AGAGCAATCC CGGGCCCAGA TGGCCCTGTG GACGGTGCTG 90

GCAGCCCCCC TCTTGATGTC CACAGACCTG CGTACCATCT CCGCCCAGAA CATGGACATT 96

^■CTGCAGAATC CACTCATGAT CAAAATCAAC CAGGATCCCT TAGGCATCCA GGGACGCAGG 102

ATTCACAAGG AAAAATCTCT CATCGAAGTG TACATGCGGC CTCTGTCCAA CAAGGCTAGC 108

GCCTTAGTCT TCTTCAGCTG CAGGACCGAT ATGCCTTATC GCTACCACTC CTCCCTTGGC 114

CAGCTGAACT TCACCGGGTC TGTGATATAT GAGGCCCAGG ACGTCTACTC AGGTGACATC 120

ATCAGTGGCC TCCGAGATGA AACCAACTTC ACAGTGATCA TCAACCCTTC AGGGGTAGTG 126

ATGTGGTACC TGTATCCCAT CAAGAACCTG GAGATGTCCC AGCAGTGAGG AGCTGGGACA 132

TGTGACAGGC TGTGGTGGCA CCACTGAGCC TAGACCATGG AGCCTTGGCA TGCCCAGGGC 138

AAGTGGGGAG GTTCTCTGCT CCCCAGGCCT GCTCGGTGAC TGACCCCATC ATACCCAAAG 144

TGCAATCTCA CGGCCAGGTT CTATGCCCTG TCCAAGCGTA AACCCTCTTG GAAACTTCTT 150

TTGGGGCAAT TTTCCTGTGG CCTTCCTGGC CTCTACTTCC ATGTGCGCAG CCCCACAGAC 156

GTTGCTGAGC AACTCGCCAG CCTCCTGAGC TCCATGCCCA TCAGGACTCT AGCCTCTGAC 162

CTTGCTGTTG ACTCTGAAAT CAGGATTTGG AAGTTTTCGA ATTAGGAGTA GAGAGATCTG 168

ACCTCTTGCC AGGAATGCCC ATGGATCATG TGATTGGCTT TTCTACCCAT AGAGGGCCTT 174

GCAGCCTGAT ACCACCTGGG AGTGAGGGTC ACAAAGGAGA CCTTGGCTCC CTCAGGTCAC 180 CAATAAACCT GTTCTTTAAT CAAAAAAAAA AAAAAAAAAA 184

(2) INFORMATION FOR SEQ ID NO:5:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 35 base pairs

(B) TYPE: nucleic acid

(C) STRANDEDNESS: single

(D) TOPOLOGY: linear

(ii) MOLECULE TYPE: cDNA

(xi) SEQUENCE DESCRIPTION: SEQ ID NO:5: ATGTGGCCCC AGGTGCGCGC GCTGGACAAT GGGCT 3

(2) INFORMATION FOR SEQ ID NO:6:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 36 base pairs

(B) TYPE: nucleic acid

(C) STRANDEDNESS: single

(D) TOPOLOGY: linear

(ii) MOLECULE TYPE: CDNA

(xi) SEQUENCE DESCRIPTION: SEQ ID NO:6: TCCCAGCAGT GAGGATCTAG AACATGTGAC AGGCTG 3

(2) INFORMATION FOR SEQ ID NO:7:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 36 base pairs

(B) TYPE: nucleic acid

(C) STRANDEDNESS: single

(D) TOPOLOGY: linear

(ii) MOLECULE TYPE: cDNA

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

ATGTGGCCCA GGTGCAGGCC TTGGACAATG GGCTCC 3

(2) INFORMATION FOR SEQ ID NO:8:

(i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 36 base pairs (B) TYPE: nucleic acid

(C) STRANDEDNESS: single

(D) TOPOLOGY: linear

(ii) MOLECULE TYPE: cDNA

(Xi) SEQUENCE DESCRIPTION: SEQ ID NO:8: TGTCCCAGCA GTGAGAAGCT TGGACATGTA CAGGCT 3

(2) INFORMATION FOR SEQ ID NO:9:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 36 base pairs

(B) TYPE: nucleic acid

(C) STRANDEDNESS: single

(D) TOPOLOGY: linear

(ii) MOLECULE TYPE: cDNA

(xi) SEQUENCE DESCRIPTION: SEQ ID NO:9: GACATGTGGC CCAGGATGCA TTGCTGGACA ATGGGC 3

(2) INFORMATION FOR SEQ ID NO:10:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 34 base pairs

(B) TYPE: nucleic acid

(C) STRANDEDNESS: single

(D) TOPOLOGY: linear

(ii) MOLECULE TYPE: cDNA

(xi) SEQUENCE DESCRIPTION: SEQ ID NO:10: CCAGCAGTGA GGAGCATGCA CATGTGACAG GCTG 3

Claims

WHAT IS CLAIMED IS:

1. A process for producing α-N-acetylgalactosaminidase, said process comprising the steps of: a) expressing a DNA sequence derived from human placenta encoding α-N-acetylgalactosaminidase in a recombinant host microorganism, said recombinant host microorganism being transformed with an expression vector containing said DNA sequence; b) isolating transformants having α-N- acetylgalactosaminidase expression; c) culturing said isolated transformants in a culture media to permit growth of said transformants; and d) purifying α-N-acetylgalactosaminidase from said culture media.

2. The process according to Claim 1, wherein said expression vector comprises at least one additional copy of an α-N- acetylgalactosaminidase gene encoding for α-N- acetylgalactosaminidase expression.

3. The process according to Claim 1, wherein said recombinant host microorganism is selected from the group consisting of yeast, bacteria and fungi.

4. The process according to Claim 3, wherein said recombinant host microorganism is a bacteria derived from Escherichia coli.

5. The process according to Claim 3, wherein said recombinant host microorganism is a yeast derived from Saccharomyces cerevisia.

6. The process according to Claim 3, wherein said recombinant host microorganism is a filamentous fungi derived from Aspergillus awamori.

7. The process according to Claim 1, wherein said α-N- acetylgalactosaminidase is purified from said culture media by column chromatography.

8. The process according to Claim 1, wherein said α-N- acetylgalactosaminidase is purified by antibody column chromatography.

9. The process according to Claim 1, wherein said expression vector comprises all of a coding region of an α-N- acetylgalactosaminidase gene and sequences necessary for the α-N-acetylgalactosaminidase gene's transcription and translation.

10. Transformants having α-N-acetylgalactosaminidase expression produced by the process of Claim 1.

11. The process according to Claim 1, wherein the DNA sequence derived from human placenta is modified to introduce moieties into the sequence prior to insertion of the DNA sequence into the expression vector, to permit removal of the enzyme after its use.

12. The process according to Claim 11, wherein the moiety introduced into the sequence is a cellulose-binding domain.

13. A recombinantly produced α-N-acetylgalactosaminidase enzyme having an amino acid and nucleotide sequence corresponding to Seq. ID No. 4 or a modification thereof.

14. A method for converting A, B and AB blood types to O blood type, said method comprising: a) obtaining a unit of A, B or AB blood; b) removing plasma and white blood cells from the blood of step a) , leaving red blood cells; c) optionally equilibrating the pH of remaining red blood cells to pH < 6; and d) treating the red blood cells of step c) with an appropriate amount of recombinantly produced α-N- acetylgalactosaminidase.

15. A method of Claim 14 wherein the recombinantly produced α-N-acetylgalactosaminidase has an amino acid and nucleotide sequence corresponding to Seq. ID No. 4 or a modification thereof.

16. A kit, useful for converting A, B and AB blood types to 0 blood type, said kit comprising: e) an appropriate amount of a recombinantly produced α- N-acetylgalactosaminidase; and f) an appropriate carrier solution.

17. A kit of claim 16 wherein the recombinantly produced α-N- acetylgalactosaminidase has an amino acid and nucleotide sequence corresponding to Seq. ID No. 4 or a modification thereof.