CA2211417A1 - Recombinant .alpha.-galactosidase enzyme - Google Patents

Abstract

This invention relates to recombinant enzyme for use in the removal of B
antigens from the surface of cells in blood products. Specifically, this invention is directed to a recombinant coffee bean .alpha.-galactosidase enzyme, a recombinant vector which encodes coffee bean .alpha.-galactosidase, methods of cloning and expressing recombinant coffee bean .alpha.-galactosidase and a method of removing B antigens from the surface of cells in blood products utilizing recombinant coffee bean .alpha.-galactosidase.

Description

CA 02211417 1997-07-2~
W096/23869 PCT~S96/01212 RECOMBINANT ~-GALACTOSIDASE ENZYME

5Statement of Government Interest This invention was made with government support under NMRDC Grant Number N00014-90-J-1638. As such, the government has certain rights in the invention.

FIELD OF THE I~v~llON
This invention relates to a recombinant enzyme for use in the removal of type B antigens from the surface of cells in blood products, thereby converting type B blood products to type O blood products and type AB blood products to type A blood products without otherwise affecting the structure and function of the cells in the blood products.
This invention further relates to methods of cloning and expressing said recombinant enzyme. More particularly, this invention is directed to a recombinant coffee bean ~-galactosidase enzyme, a recombinant vector which encodes coffee bean ~-galactosidase, methods of cloning and expressing said recombinant ~-galactosidase enzyme, the use of said recombinant ~-galactosidase enzyme to cleave galactose sugar residues, most particularly ~1,3-linked galactose residues, which are responsible for blood group B
specificity, and a method of removing type B antigens from the surface of cells in type B and AB blood products using said recombinant coffee bean ~-galactosidase enzyme by contacting said enzyme with blood products so as to remove the terminal moiety of the B-antigenic determ;nAnt from the surface of cells (for example, erythrocytes~ in said blood products. The recombinant coffee bean ~-galactosidase enzyme of this invention provides a readily available and cost-efficient enzyme which can be used in the removal of type B antigens from the surface of cells in type B and AB
blood products. Treatment of type B blood products with the recombinant enzyme of this invention provides a source of cells free of the B antigen, which blood products are -CA 02211417 1997-07-2~
W 096/23869 PCTrUS96/01212 thereby rendered useful in transfusion therapy in the same manner as O type blood products.

BACKGROUND OF THE INVENTION
As used herein, the term "blood products" includes whole blood and cellular components derived from blood, including erythrocytes (red blood cells) and platelets.
There are more than thirty blood group (or type) systems, one of the most important of which is the ABO
system. This system is based on the presence or absence of antigens A and/or B. These antigens are found on the surface of erythrocytes and on the surface of all endothelial and most epithelial cells as well. The major blood product used for transfusion is erythrocytes, which are red blood cells containing hemoglobin, the principal function of which is the transport of oxygen. Blood of group A contains antigen A on its erythrocytes. Similarly, blood of group B contains antigen B on its erythrocytes.
Blood of group AB contains both antigens, and blood of group 0 contains neither antigen, but does contain a structure known as H antigen.
The blood group structures are glycoproteins or glycolipids and considerable work has been done to identify the specific structures making up the A and B determinants or antigens. It has been found that the blood group specificity is determined by the nature and linkage of monosaccharides at the ends of the carbohydrate chains. The carbohydrate chains are attached to a peptide or lipid backbone which is embedded in the lipid bi-layer of the membrane of the cells. The most important (immuno-dominant or immuno-determinant) sugar has been found to be N-acetylgalactosamine for the type A antigen and galactose for the type B antigen.
Blood of group A contains antibodies to antigen B.
Conversely, blood of group B contains antibodies to antigen A. Blood of group AB has neither antibody, and blood group O has both. A person whose blood contains either (or both) CA 02211417 1997-07-2~

of the anti-A or anti-B antibodies cannot receive a transfusion of blood cont~;n;ng the corresponding incompatible antigen(s). If a person receives a transfusion of blood of an incompatible group, the blood transfusion 5 recipient's antibodies coat the red blood cells of the transfused incompatible group and cause the transfused red blood cells to agglutinate, or stick together. Transfusion reactions and/or hemolysis (the destruction of red blood cells) may result therefrom.
In order to avoid red blood cell agglutination, transfusion reactions and hemolysis, transfusion blood type is cross-matched against the blood type of the transfusion recipient. For example, a blood type A recipient can be safely transfused with type A blood which contains compatible antigens. Because type 0 blood contains no A or B antigens, it can be transfused into any recipient with any blood type, i.e., recipients with blood types A, B, AB or O.
Thus, type 0 blood is considered "universal", and may be used for all transfusions. Hence, it is desirable for blood banks to maintain large quantities of type O blood.
However, there is a paucity of blood type O donors.
Therefore, it is useful to convert types A, B and AB blood to type O blood in order to maintain large quantities of universal blood products.
In an attempt to increase the supply of type O
blood, methods have been developed for converting certain type A, B and AB blood to type O blood (containing only the H antigen structure). For example, U.S. Patent No.
4,330,619 entitled "Enzymatic Conversion of Red Cells for Transfusion" issued May 18, 1982 to Goldstein ("the '619 Patent"), which is incorporated herein by reference, is directed to a process for converting type B erythrocytes to the H antigen type (or type O) and type AB erythrocytes to type A utilizing a!-galactosidase enzyme. The process for converting B and AB erythrocytes which is described in the '619 Patent includes the steps of equilibrating B or AB
erythrocytes, contacting the equilibrated erythrocytes and CA 022ll4l7 l997-07-2~
W 096/23869 PCTrUS96/01212 purified ~-galactosidase for a period of time sufficient to convert the B antigen in the erythrocytes to the H antigeln, removing the ~-galactosidase enzyme from the erythrocytes and re-equilibrating the erythrocytes. U.S. Patent No.
4,427,777 entitled "Enzymatic Conversion of Red Cells for Transfusion" issued January 24, 1984 to Goldstein, which Patent is incorporated herein by reference, is directed to compositions free of B antigens wherein B antigens are removed from the compositions utilizing ~-galactosidase and the process described in the '61~ Patent.
a-galactosidase enzymes are characterized (and thereby named) by their ability to cleave a-linked galactose sugar groups. In isolating or identifying these enzymes, their activity is assessed in the laboratory by evaluating cleavage of synthetic substrates which mimic the sugar groups cleaved by the enzymes, with p-nitrophenyl ~-D-galactopyranoside derivatives of the target sugar groups being commonly used. Synthetic substrates are useful in enzyme identification and isolation procedures (the quantitative cleavage of these synthetic substrates can be used to readily distinguish (and thereby identify) enzymes isolated from different sources). However, these synthetic substrates and other oligosaccharide substrates are structurally simple and small-sized, and mimic only a portion of the natural glycoproteins and glycolipid structures (glycoconjugates) which are of primary concern, those being the B antigens on the surface of cells.
~ -galactosidase enzymes from a number of sources have been purified, sequenced, cloned and expressed. (See, for example, Fellinger et al., Yeast, Vol. 7, pp. 463-473 (1991) (expression of guar a-galactosidase); Yagi et al~, Archives Biochem. and Biophysics, Vol. 280, pp. 61-67 (1990) (purification of Ehrlich ascites tumor cells and fluid ~-galactosidase); Bahl et al., Meth. EnzYmol., Vol. XXVIII, pp. 728-743 (1972) (purification of AsPerqillus niger ~-galactosidase); and Courtois et al., Meth. EnzYmol., Vol. VIII, pp. 565-571 (1966) (purification of coffee bean CA 022ll4l7 l997-07-2~
W 096/23869 PCTrUS96/01212 ~-galactosidase)). However, not all ~-galactosidase enzymes are appropriate for use in removing B antigens from the surface of cells in blood products.
In determining whether an enzyme is appropriate for use in removing B antigens from the surface of cells, one must consider the following enzyme characteristics:
substrate specificity, specific activity or velocity of the substrate cleavage reaction, and pH optimum. Substrate specificity is measured in the Km value, which measures the binding constant or affinity of an enzyme for a particular substrate. The lower a Km value, the more tightly an enzyme binds its substrate. The velocity of an enzyme cleavage reaction is measured in the Vmax, the reaction rate at a saturating concentration of substrate. A higher Vmax indicates a faster cleavage rate. The ratio of these two parameters, Vmax/Km, is a measure of the overall efficiency of an enzyme in reacting with (cleaving) a given substrate.
A higher Vmax/Km indicates greater enzyme efficiency. For successful and clinically applicable removal of B antigens from the surface of cells, the enzyme must be sufficiently active at or above a pH at which the cells being treated that can be maintained, that being pH 5.6 (or above) for red cells. Therefore, the pH optimum and activity profile of an appropriate enzyme must still provide reasonable enzyme activity at this pH.
The pH optimum of Ehrlich cell ~-galactosidase enzyme centers near 4.5, irrespective of substrate (see Yagi et al., Archives Biochem. and Biophysics, Vol. 280, pp.
61-67 (1990)). The pH optimum or Ehrlich cell ~-galactosidase has been found to be 4.5 for water-soluble fluorogenic substrates and oligosaccharides (see Dean et al., J. Biol. Chem., Vol. 254, pp. 10006-10010 (1979)). In contrast, the pH optimum of coffee bean ~-galactosidase for the fluorogenic substrate PNP-~-Gal is 6.0, indicating that the coffee bean enzyme exhibits significant activity at or above a pH at which cells are treated for removal of B
antigens.

CA 02211417 1997-07-2~
W 096123869 PCTrUS96/01212 Coffee bean ~-galactosidase enzyme shows a Vmax/km value of 236 at pH 6.0 toward PNP-~-gal, whereas ~-galactosidases isolated from human cells (see Dean and Sweeley, J. Biol. Chem., Vol. 254, pp. 10006-10010 (1979)) or Ehrlich ascites tumor (see Yagi et al., Archives Biochem.
and Biophysics, Vol. 280, pp. 61-67 (1990)) only have a Vmax/Km value of between 7.59 and 9.9 at pH 4.5-4.6 (using 4-Me-~-gal or PNP-~-gal). Furthermore, in the comparison of substrate specificity of ~-galactosidases from coffee bean and Ehrlich ascites tumor cells, Yagi et al. found that coffee bean ~-galactosidases showed much higher Vmax/km values toward oligosaccharide substrates such as raffinose (450 fold), gal~1,3 gal (180 fold) and galactor~nn~n (30 fold). Based on this study, they concluded that coffee bean ~-galactosidase showed a relatively broad substrate specificity, suggesting that it is suited for cleaving many kinds of terminal ~-galactosyl linkages. Of all the ~-galactosidases studied, the one obtained from coffee bean demonstrates the highest activity in removing terminal ~1,3-linked galactose residues from glycoconjugates. This makes the coffee bean enzyme a most appropriate enzyme in the study and performance of enzymatic blood conversion.
Prior to the present invention, it was necessary to purify the ~-galactosidase enzyme from a coffee bean source, a process which is time consuming and can be expensive. Because coffee bean ~-galactosidase can be used to convert B and AB blood products, a need has arisen to develop a coffee bean ~-galactosidase enzyme source which is more readily available. In addition, a need has arisen to develop a coffee bean ~-galactosidase enzyme useful in type B and AB blood product conversion, the production of which enzyme is cost-efficient. Further, it is desirable to provide a more readily available and controlled source of enzyme, that source being a cloned and expressed enzyme, which would provide an enzyme source which is more consistent and which is readily purified at less cost.
Additionally, a recombinant, cloned enzyme would allow for -CA 022ll4l7 l997-07-2~
W 096123869 PCTrUS96/01212 specific protein sequence modifications, which can be introduced in order to generate an enzyme with further optimized specific activity, substrate specificity and pH
range.
It is therefore an object of this invention to provide recombinant coffee bean ~-galactosidase enzyme for use in the removal of B antigens from the surface of cells in blood products.
It is another object of this invention to provide recombinant coffee bean a-galactosidase enzyme for use in the removal of B antigens from the surface of cells in blood products wherein said enzyme is readily available and may be manufactured on a cost-efficient basis.
It is a further object of this invention to provide methods of cloning and expressing recombinant coffee bean ~-galactosidase enzyme useful in the removal of B
antigens from the surface of cells in blood products.
It is another object of this invention to provide a recombinant vector containing a nucleotide sequence encoding coffee bean ~-galactosidase enzyme useful for expressing recombinant coffee bean ~-galactosidase enzyme or for modifying said enzyme through recombinant methods.
It is a still further object of this invention to provide a recombinant coffee bean ~-galactosidase enzyme which is capable of removing terminal ~1,3-linked galactose residues from substrates and glycoconjugates.
It is yet another object of this invention to provide a method of removing B antigens from the surface of cells in blood products using recombinant coffee bean ~-galactosidase enzyme.

SU~ RY OF THE INrVENTION
This invention is directed to a recombinant coffee bean ~-galactosidase enzyme capable of cleaving ~1,3-linked glycoside linkages on cells. This invention is further directed to a recombinant vector containing a nucleotide sequence encoding coffee bean ~-galactosidase.

_ _ CA 02211417 1997-07-2~

W 096123869 PCT~US96101212 Additionally, this invention is directed to a method of producing coffee bean ~-galactosidase, and to a method of removing B antigens from the surface of cells which method comprises contacting cells with recombinant coffee bean ~-galactosidase enzyme for a period of time sufficient to remove the B antigens from the surface of the cells.

BRIEF DESCRIPTION OF THE DRAWINGS
The above brief description, as well as further objects and features of the present invention, will be more fully understood by reference to the following detailed description of the presently preferred, albeit illustrative/
embodiments of the present invention when taken in conjunction with the accompanying drawings wherein:
Figure 1 represents the nucleotide and deduced amino acid sequence of full-length cDNA encoding coffee bean ~-galactosidase;
Figure 2 represents a comparison of sequence homology of ~-galactosidase from coffee bean, guar (CYamoPsis tetraqonoloba), human placenta, yeast (Saccharomyces cerevisiae) and fungi (Aspergillus niaer) as aligned using the computer program PROSIS and manual arrangement;
Figure 3 represents immunoprecipitation with polyclonal antibody of cloned coffee bean ~-galactosidase expressed in vitro in rabbit reticulocyte lysate and wheat germ extract, as analyzed by SDS-PAGE and autoradiographed;
and Figure 4 represents Western blot analysis of recombinant coffee bean ~-galactosidase expressed in transfected sf9 insect cells using antibody against purified coffee bean ~-galactosidase.
Figure 5 represents the sequence of the plasmid p~F-BZ around the 5' cloning site (EcoRI) of the insert.
The two arrows indicate the signal cleavage sites as indicated by N-terminal sequencing of the secreted enzyme.

CA 02211417 1997-07-2~
W 096/23869 PCTrUS96101212 Figure 6 represents the chromatogram from the cation exchange chromatography purification of the recombinant ~-galactosidase enzyme produced by the Pichia pastoris expression system.
Figure 7 represents the SDS-PAGE analysis of the chromatography purified recombinant a-galactosidase enzyme produced by the Pichia pastoris expression system. Lane 1, supernatant of Pichia pastoris culture; lane 2, unbound fraction from the column; lane 3 r fraction #55; lane 4, fraction #65; lane 5, fraction #80; lane 6, fraction #150;
lane 7, native ~-galactosidase enzyme; and lane 8, size markers (kDa).

DETAILED DESCRIPTION OF THE INVENTION
This invention is directed to a recombinant coffee bean ~-galactosidase enzyme capable of cleaving ~1,3-linked glycoside linkages on cells. The recombinant coffee bean ~-galactosidase enzyme of the invention has a molecular weight of about 42 kDa, and has about 80% amino acid sequence homology with guar a-galactosidase enzyme. This invention is further directed to a recombinant vector containing a nucleotide sequence which encodes coffee bean ~-galactosidase.
In addition, this invention is directed to a method of producing coffee bean ~-galactosidase, and to a method of removing B antigens from the surface of cells which method comprises contacting cells with a recombinant coffee bean ~-galactosidase enzyme for a period of time sufficient to remove the B antigens from the surface of the cells.
Group B erythrocytes may be treated with ~ ~-galactosidase isolated from coffee beans to cleave the terminal ~1,3-linked galactose residues responsible for ~ blood group B specificity in order to convert the group B
erythrocytes serologically to group O erythrocytes. Hence, it is desirable to have readily available purified coffee bean ~-galactosidase. In order to produce purified CA 02211417 1997-07-2~
W 096/23869 PCTrUS96101212 ~-galactosidase enzyme in large quantities and improve its enzymatic properties, the inventors have isolated the cDNA
clone for coffee bean ~-galactosidase.
As discussed hereinabove, ~-galactosidase has been purified from several sources. However, only coffee bean ~-galactosidase cleaves ~1,3-linked galactose residues responsible for blood group B specificity. Hence, only coffee bean ~-galactosidase can be used to convert type B
blood products to type O blood products, and type AB blood products to type A blood products.
The full length cDNA which encodes coffee bean ~-galactosidase is represented in SEQ ID NO:1 and Figure 1~
A DNA vector containing a sequence encoding coffee bean ~-galactosidase was deposited under the Budapest Treaty with the American Type Culture Collection, Rockville, Maryland, on September 8, 1993, tested and found viable on September 14, 1993, and catalogued as ATCC #75556.
Methods which are well known to those skilled in the art can be used to construct expression vectors containing the coffee bean ~-galactosidase coding sequence, with appropriate transcriptional/translational signals for expression of the enzyme in the corresponding expression systems. Appropriate org~n; ~r~, cell types and expression systems include: cell-free systems such as a rabbit reticulocyte lysate system, prokaryotic bacteria, such as E.
coli, eukaryotic cells, such as yeast, insect cells, mammalian cells (including human hepatocytes or Chinese hamster ovary (CH0) cells), plant cells or systems, and animal systems including oocytes and transgenic animals.
The entire coffee bean ~-galactosidase coding sequence or functional fragments of functional equivalents thereof may be used to construct the above expression vectors for production of functionally active enzyme in the corresponding expression system. Due to the degeneracy of the DNA code, it is anticipated that other DNA sequences which encode substantially the same amino acid sequence may be used. Additionally, changes to the DNA coding sequence CA 02211417 1997-07-2~
W096/23869 PCT~S96/01212 which alter the amino acid sequence of the coffee bean ~-galactosidase enzyme may be introduced which result in the expression of functionally active enzyme. In particular, amino acid substitutions may be introduced which are based on similarity to the replaced amino acids, particularly with regard to the charge, polarity, hydrophobicity, hydrophilicity, and size of the side chains of the amino acids.
Once a recombinant coffee bean ~-galactosidase enzyme is cloned and expressed, said enzyme can be used to remove B antigens from the surface of cells in blood products. Type B antigens can be removed from the surface of erythrocytes by contacting the erythrocytes with the recombinant coffee bean ~-galactosidase enzyme of the invention for a period of time sufficient to remove the B
antigens from the surface of the erythrocytes.

ExamPle In order to assess the relative abilities of ~-galactosidase enzyme isolated from different sources to remove el,3-linked galactose residues from red cells, l00 ml of type B red blood cells were treated with isolated ~-galactosidases from yeast (S. cerevisiae), fungi (A.
niger), guar (C. tetraqonoloba) and coffee bean~ The treatment conditions and digestion results are provided below in Table I, below. Digestion of terminal sugars (~1,3-linked galactose residues) was determined by assessing reduction or elimination of the agglutination of treated cells in the presence of polyclonal anti-B antibody. No detectable change in agglutination indicated no digestion~

CA 02211417 1997-07-2~
W 096/23869 PCTrUS96/01212 TABLE I
Source of Digestion ~-Galactosidase Conditions Results Yeast 90 U/ml RBC No Digestion 26~C, 17 hours pH 5.6 10 Fungi 1600 U/ml RBC No Digestion 26~C, 6 hours pH 5.6 Guar 20 U/ml RBC* No Digestion 26~C, 2.25 hours pH 5.6 Coffee Bean 90 U/ml RBC Complete 26~C, 90 minutes Digestion pH 5.6 *Digestion with coffee bean enzyme under the same conditions removes at least 95% of B antigens.
None of the a-galactosidases isolated from sources other than coffee bean showed any significant activity in removing the terminal a-linked galactose residues from the red blood cell surfaces. In contrast, coffee bean a-galactosidase demonstrated high activity in removing terminal al,3-linked galactose residues from glycoconjugates on the red cell surfaces. Hence, coffee bean ~-galactosidase is appropriate for use in enzymatic blood group B type conversion.

Peptide Sequencing of a-Galactosidase Purified from Coffee Beans In order to sequence the coffee bean a-galactosidase, coffee bean a-galactosidase was purified to apparent homogeneity from green coffee beans. The procedure used for purification of a-galactosidase from coffee beans was developed and optimized in the laboratory to provide pure a-galactosidase enzyme (demonstrating a single band on SDS polyacrylamide gel electrophoresis) with optimal yields under large scale conditions. This provided sufficient - =
CA 022ll4l7 l997-07-2~
W 096/23869 PCTrUS96/01212 material to be used for treatment of type B blood products to remove B antigens on a clinical scale. The following procedure was utilized: Green Santos beans were frozen in either liquid nitrogen or in a -70~C freezer to facilitate grinding in a Waring blender. The crude powder obtained was ~ homogenized with H2O (at a ratio of 4 liters H2O per kg ground beans) and allowed to stand overnight at room temperature. The following day repeated homogenization was necessary to achieve a rich first extract, which was then expressed (under pressure) through multiple layers of cheesecloth. The homogenate was subjected to a second extraction as described above. The pH of the combined extracts was adjusted to 4.0 with acetic acid and then 0.5%
diatomaceous earth (Super Cell filter aid, available from Cellulo Company, a division of Gosmer Enterprises, Cranford, NJ) was added. The formed heavy precipitate was removed with a filter funnel. The filtrate was then processed through the chromatography steps as indicated in Table II.
After steps 1, 2, 4 and 5 in Table II, the collected main fraction was concentrated and then equilibrated for the next step. This was accomplished with a Pellicon Cassette System using a 10,000 MW cutoff membrane. PCS buffer pH 5.6 had the following composition:
58 mM dibasic sodium phosphate, 21 mM citric acid, 77 mM
sodium chloride. Concentrate phosphate-citrate buffer was prepared by titrating 50 mM citric acid with 100 mM dibasic sodium phosphate pH 3.7. Sepharose divinylsulfone galactose was prepared according to the procedure described by Ersson et al., Biochem. et Biophys. Acta, Vol. 494, pp. 51-60 (1977), which is incorporated herein by reference. PBE94 (Polybuffer exchanger 4) is available from Pharmacia, Inc~
Other similar anion exchange resins, which are well known in the art and available commercially, can be used in place of DE53 in the above procedure, particularly other DEAE
(diethylaminoethyl) resins.

CA 022ll4l7 l997-07-25 W 096/23869 PCTrUS96/01212 ...
q ~I ~U
;~ ~ C
q O O ~ ~ ~ O O O

o L G

_- I ,o ~D
Q_ = D C =

D ~ C
~ S~ ~w~vc î ~vv T T
.

"
u c .
J Llq -- 2 ~Jq ~ ~
~ O
_r ~ o = ~ q; = _ -- = O al = t_~

l~q ~Iq ~ o O O O
r~ O ~ O O O O
~ r' O
~1 ~ ', 'J ~ "~
-- ~ rq _ O
rq _ _ C -,,, -. , U~ " --_ ~ rq ~t Lq ~ q ~q CA 02211417 1997-07-2~
W096/23869 PCT~S96/01212 Alternative methods for purification of ~-galactosidase from coffee beans have been reported by Haibach et al., Biochem. et Biophys. Res. Comm., Vol. 181, No. 3, pp. 1564-1571 (1991) and Harpaz et al., Biochem et Biophys. Acta, Vol. 341, pp. 213-221 (1974), however, these purification procedures have been found to provide lower overall yields of enzyme from coffee beans on sufficient scale for clinical use than that described above. The enzyme activity of coffee bean ~-galactosidase was assayed according to the procedure described by Kuo et al., Enzyme Microb. Technol., Vol. 5, pp. 285-290 (1983)), which is incorporated herein by reference, using a final concentration of PNP-a-Gal of 1.25 mM.
Microsequencing of the unblocked mature enzyme provided an amino-terminal sequence (N-pep) of 19 residues~
In order to obtain additional peptide sequences including internal sequences, 0.2 mg of the purified ~-galactosidase was treated with 2 mg of cyanogen bromide in 70~ formic acid for 24 hours at room temperature in the dark. The peptides were isolated by reverse phase HPLC. Two peptide sequences, 2-pep and 3-pep, were then determined by automated gas phase microsequencing. The sequences of the three peptides (N-pep, 2-pep and 3-pep) are indicated in Figure 1.
Figure 1 represents the full length cDNA encoding coffee bean ~-galactosidase. Three peptide sequences, N-pep, 2-pep and 3-pep, which were obtained from purified coffee bean ~-galactosidase, are underlined below the deduced amino acid sequence. The first 15 amino acids comprise a putative signal peptide which is cleaved during biosynthesis. Therefore, the mature coffee bean ~-galactosidase enzyme is comprised of the amino acids 16-378 of Figure 1. The potential N-linked glycosylation site is double-underlined at amino acid residues 160-162.
The polyadenylation signal (AA'Tl~AA) at the position ntl361-1366 is boxed.
The oligonucleotides, CB1, CB4 through CB9, are shown with arrows to indicate 5' to 3' direction. Based on CA 02211417 1997-07-2~

W O 96/23869 PCTrUS96/01212 the sequence of the peptide, N-pep, the CBl* was designed as 5'ACA(CT)CCA(T)CCA(T)ATGGNTGGAA. Accordingly, the CB4*, based on the sequence of the peptide, 3-pep, has the sequence 5'-TGT(A)GGT(GA)GTNAGG(CA)ACG(A)TACAT. CB1 bears the least codon degeneracies in the peptide sequence of N-pep as determined by the computer program "Primer"
(Scientific and Educational Software, Inc.). Because CNBr cleaves at the C-terminal side of the methionine residue, the codon for methionine was included at the 3' end of oligonucleotide CB4. Because there have not been any reported genes cloned from the coffee plant, the preference codons used in the designing of the oligonucleotides were chosen based on those of other plants listed in the codon usage table.
Molecular Cloning of the Full Length cDNA Encoding Coffee Bean ~-Galactosidase In order to perform molecular cloning of the full length cDNA encoding coffee bean ~-galactosidase, total RNA
was prepared from 2 grams of dried green coffee beans by using the Extract-A-plant RNA Isolation kit (ClonTech) according to the manufacturer's procedure. The quality of the isolated RNA was confirmed by denaturing agarose gel electrophoresis. The messenger RNA was purified from the total RNA by using an oligo-dT column (ClonTech).
In order to isolate the specific cDNA encoding the ~-galactosidase, cDNA using isolated coffee bean mRNA was prepared according to the stAn~rd procedure known in the art for reverse transcription. A mixture of oligo dT and random primer was used as the reverse transcription primer in the reaction to avoid the 3' bias when oligo-dT is used alone. The cDNA then provided the template in a PCR
reaction for 35 cycles, 94~C 1 minute, 50~C 2 minutes and 72~C 3 minutes. In the presence of oligonucleotides CB1 and CB4, this PCR procedure produced a fragment of approximately l.lkb. This fragment, designated BZ, was cloned directly into the pCRII vector (Invitrogen) for further analysis.
The sequencing data indicated that BZ was highly homologous CA 022ll4l7 l997-07-2~
W 096/23869 PCTnUS96/01212 with guar ~-galactosidase. (BZ corresponds to sequence nucleotides (nt) 168 to 1234 in Figure 1). Furthermore, its deduced amino acid sequence matched the peptide sequences obtained from purified coffee bean ~-galactosidase, providing evidence for its authenticity as coffee bean ~-galactosidase cDNA.
In order to obtain the full-length cDNA coding for the coffee bean ~-galactosidase, the technique of 5' and 3' RACE was applied (see Frohman et al., Proc. Natl. Acad. Sci.
USA, Vol. 85, pp. 8998-9002 (1988)). For 5' RACE, the oligonucleotide, CB8, was first used as a primer for the reverse transcription from coffee bean mRNA. The following procedure was then followed for 5' RACE as in the kit manufacturer's protocol (Bethesda Research Laboratories (BRL)). The second oligonucleotide, CB6, together with the universal primer was used to amplify the 5' region upstream of the coding sequence by PCR. Since no distinctive DNA
band was visible on the agarose gel after two PCR
amplifications, the PCR product mixture was cloned into the pCRII vector and screened by hybridizing the colonies with the radioactively labeled l.lkb fragment BZ (sequence ntl68-1234 in Figure 1). The positive colonies were picked for plasmid preparation. The sequencing of the plasmid indicated that the DNA fragment, designated S'BZ, obtained by the 5' RACE technique contained a 240bp overlap (the sequence between CB1 and CB6) with the BZ, and about a 170bp further upstream sequence which includes the N-terminus of the mature enzyme and the putative signal peptide sequence.
To isolate the 3' sequence, downstream from the fragment BZ, the cDNA was reverse-transcribed from coffee bean mRNA by using a primer, PI, which has the sequence 5'-GACTCGAGTCGACATCGA-(T)17. PCR amplification was then carried out with a specific primer CB7 (sequence nt940-957 in Figure 1) and an adapter primer, PII. PII has the same sequence as PI except that it lacks seventeen thymidine residues at its 3' end. The PCR product was analyzed on 1~
low melting point agarose gel and a distinctive band of CA 02211417 1997-07-2~
W O 96/23869 PCTfUS96/01212 about 500 bp long was visualized. The fragment designated 3' BZ, was isolated and cloned into the pCRII vector for sequencing. The sequence data indicated that the 5' region of the fragment, 3'BZ, is identical to the 3' region (sequence from nt940-1234) of the BZ. An in-frame stop codon TGA was localized at ntl236-1238, confirming that the peptide sequence, 3-pep, obtained from the purified coffee bean ~-galactosidase represents the C-terminal sequence of the protein.
The three DNA fragments, 5'BZ, BZ and 3'BZ, were linked together by PCR to reconstitute the full length clone for coffee bean ~-galactosidase. Oligonucleotide CB9 was made, which corresponded to sequence nt77-94 shown in Figure 1. cDNA was synthesized from coffee bean mRNA using the 3' RACE technique as previously described. A 1.35kb fragment was amplified by PCR using two primers CB9 and PII, and was then cloned into the pCR II vector. The plasmids thus generated contained the 1.35kb insert in both orientations.
The plasmid pCR-BZ6 has the insert downstream of an SP6 promoter, which was used in in vitro expression. A second plasmid pCR-BZ7 containing the opposite insert orientation was used for subcloning the ~-galactosidase cDNA into a baculovirus expression vector. The sequence of the 1.35kb product matched with the corresponding sequences from the three separate fragments, 5'BZ, BZ and 3'BZ, confirming the authentic sequence of the coffee bean ~-galactosidase cDNA
shown in Figure 1.

Characterization of the Coffee Bean ~-Galactosidase cDNA Clone Next, the coffee bean ~-galactosidase cDNA clone was characterized. The sequence shown in the Figure 1 encodes a protein having a molecular weight of 42kDa, which closely approximates the size of the purified coffee bean ~-galactosidase as estimated on SDS-PAGE. Three peptide sequences, N-pep, 2-pep and 3-pep, which were derived from purified enzyme, are underlined in Figure 1. These sequences matched the deduced amino acid sequences. This CA 02211417 1997-07-2~

confirms that the cDNA clone isolated from coffee bean RNA
encodes ~-galactosidase. Since the mature protein starts at leucine, amino acid residue 16 in Figure 1, the first 15 residues may serve as a signal sequence which is removed 5 following translation. There is an in-frame stop codon (TGA) at position nt66-68, suggesting that the first downstream ATG at position ntlO2-104 is the initiation codon for in vivo synthesis of ~-galactosidase precursor. The sequence at residues 160-162 (boxed) is the only possible 10 site for N-glycosylation. However, purified coffee bean c~-galactosidase does not bind to ConA Sepharose, indicating that there is no or minimal glycosylation of c~-galactosidase enzyme synthesized in coffee beans.
The first plant ~-galactosidase cDNA was cloned 15 from guar (see Overbeeke et al., Plant Molecular Biology, Vol. 13, pp. 541-550 (1989)). Guar ~-galactosidase encodes a protein of 411 residues haviny a molecular weight of 45 kDa. Although both ~-galactosidases from guar and coffee bean show comparable activities toward the synthetic 20 substrate PNP-~-gal, their specificities toward oligosaccharide chains are very different. Guar ~-galactosidase primarily cleaves a 1,6 glycoside linkages, whereas coffee bean ~-galactosidase cleaves a! 1,3 and 1,4 linkages. Thus, the guar ~-galactosidase is unable to 25 cleave significant amounts of terminal ~1,3-linked galactose residues from the cell surface of B group red cells. Figure 2 represents the sequence homology of ~-galactosidase from different sources. The aminc acid sequences of ~-galactosidase from coffee bean (coffee), CyamoPsis 30 tetraqonoloba (guar), human placenta (human), Saccharomyces cerevisiae (yeast) and Asperqillus niger (Aspergillus) were aligned by using the computer program PROSIS (Hitachi Software Engineering Corp., Ltd.) and manual arrangement.
The gaps are created in order to show maximum similarity.
35 The numbers above the sequences indicate the relative position of each amino acid sequence. The sequences of yeast and AsPerqillus niqer ~-galactosidases are truncated CA 02211417 1997-07-2~

at the C-terminus, (indicated by *), removing 38 and 103 residues respectively. Identical or conservatively substituted amino acid residues (five out of six or more at the same position) are boxed according to the equivalent amino acid list. 1: A,S,T,P and G; 2: N,D,E and Q; 3: H,R
and K; 4: M,L,I and V; and 5: F,Y and W.
By using the protein analysis program PROSIS
(Hitachi Software Engineering Corp., Ltd.~, it was determined that the deduced amino acid sequences of the coffee bean cDNA clone and the guar a-galactosidase cDNA
clone share approximately 80% overall homology, even though their signal peptide sequences bear little similarity. The coffee bean a-galactosidase also shows amino acid sequence homology with a-galactosidase from human (59%), yeast (58%) and Asperqillus niqer (52%).
As shown in Figure 2, when the deduced amino acid sequences of a-galactosidase from these five different sources are aligned, many residues are well conserved in all of these a-galactosidases. However, these sequences share little, if any, homology with a-galactosidase isolated from E. coli (see Liljestrom et al., Nucleic Acids Res., Vol. 15, pp~ 2213-2220 (1987)). Recently, two more ~-galactosidase cDNAs have been reported from Klebsiella Pneumoniae (see Hama et al., J. Biol. Chem., Vol. 267, pp. 18371-18376 (1992)) and a Streptococcus mutant (see Aduse-Opoku et al., J. Gen. Microbiol., Vol. 137, pp. 2271-2272 (1991)).
However, they do not bear any sequence homology with known a-galactosidases at the amino acid level.

In Vitro ExPression of Clone Coffee Bean a-Galactosidase In order to perform in vitro expression of cloned coffee bean ~-galactosidase in a transcription-translation coupled system (TNT system, Promega), two plasmids were used: pCR-BZ6, which contains a-galactosidase cDNA
downstream from the SP6 promoter, and the vector pCRII as a control. The protein(s) were expressed in both rabbit reticulocyte lysate and wheat germ extract and then CA 022ll4l7 l997-07-2~
W O 96/23869 PCTrUS96/01212 immunoprecipitated by the polyclonal antibody raised against purified coffee bean ~-galactosidase. All the samples were analyzed by a SDS-PAGE and autoradiographed as shown in Figure 3. Figure 3 represents in vitro expression and immunoprecipitation of cloned coffee bean ~-galactosidase.
Approximately 2 mg of plasmids (pCR and pCR-BZ6) were added to 50 ml mixture of TNT rabbit reticulocyte lysate in the presence of 35S-methionine and SP6 DNA polymerase according to the Promega recommended protocol. After incubation at 30~C for 90 minutes, the samples (5 ml of each reaction) were loaded onto a 12~ gel SDS-PAGE (lanes 2 and 3).
Immunoprecipitation was carried out by incubating the same expression mixtures (20 ml) with antisera (1 ml) raised against purified coffee bean ~-galactosidase. The immunoprecipitated samples were analyzed in lanes 4 and 5.
Lanes ~ through 9 show the results of the same experiments except that TNT wheat germ extract was used instead of rabbit reticulocyte lysate. The arrow at right indicates the expressed ~-galactosidase. The molecular weight standard (lane 1) is shown at left.
When the rabbit reticulocyte lysate was used, multiple proteins (molecular weights ranging from 25 kDa to 42 kDa) were expressed from ~-galactosidase cDNA and recognized by the antibody (lanes 3 and 5). In contrast, only one predominant band was visualized on the gel (lanes 7 and 9) by using wheat germ extract. This band, as indicated by an arrow, migrated as a protein of approximate molecular weight 42 kDa, the predicted size based on the a-galactosidase cDNA sequence. The apparent discrepancy of the translation patterns in these two extracts may possibly be due to the fact that since the ~-galactosidase cDNA was isolated from coffee bean, its expression may be more optimized in the wheat germ extract. The multiple bands observed in the rabbit reticulocyte lysate might result from alternative initiation or premature termination. The results show that the 42 kDa protein expressed in vitro is coffee bean ~-galactosidase.

CA 022ll4l7 l997-07-2~

W O 96/23869 PCTrUS96/01212 Functional Expression of Coffee Bean a-Galactosidase in Insect Cells Coffee bean ~-galactosidase was then functionally expressed in insect cells. Many eukaryotic proteins have been expressed in insect cells infected with recombinant baculovirus (see King et al., The Baculovirus Expression System: A Laboratory Guide, Chapman ~ Hall, New York (1992)). Since insect cells carry out the post-translational processing events that occur in eukaryotic cells, including glycosylation and signal peptide cleavage, proteins produced by such baculovirus expression systems are similar to their natural counterparts in biological activity, structure and antigenicity.
Coffee bean ~-galactosidase cDNA was subcloned from the plasmid pCR-BZ7 into the unique NotI/BamHI sites of a baculovirus expression vector pVL 1392 (PharMingen), generating the plasmid pVL-BZ. Expression of ~-galactosidase cDNA was thus under the control of a strong viral promoter (polyhedrin promoter). The plasmid pVL-BZ
was co-transfected into sf9 insect cells with baculoGold DNA
(PharMingen), a lethal deletion of the virus DNA, according to the procedure suggested by the manufacturer. Through homologous recombination with a complementary sequence in the plasmid, viable virus cont~;n;ng the ~-galactosidase cDNA was thus reconstituted inside the insect cells and released into the medium. The transfection supernatant (1 ml) was then added to fresh sf9 cells (2 x 106). After incubation at 27~C for three days, the supernatant was harvested and used for virus amplification one more time in order to obtain a high titer of virus.
In order to detect the expression of ~-galactosidase in the transfected sf9 cells, a Western blot was carried out by using antibody against purified coffee bean ~-galactosidase. Figure 4 represents expression of recombinant coffee bean ~-galactosidase in insect cells (sf9). The supernatants and cells were collected after the second amplification and analyzed by Western blot with polyclonal antibody against purified coffee bean -CA 02211417 1997-07-2~
W096/23869 PCT~S96101212 ~-galactosidase. The supernatant and cells from pVL-BZ
transfection are shown in lanes l and 2, respectively. The supernatant (lane 3) and cells (lane 4) from wild-type virus transfection were used as a negative control. Lane 5 is ~-galactosidase purified from coffee bean. Molecular weight standards are listed at left.
As shown in Figure 4, the coffee bean ~-galactosidase was expressed in the sf9 cells transfected with the plasmid pVL-BZ (lane 2) but not in cells transfected with wild-type virus (lane 4). Its migration on the gel was similar to that of purified enzyme (lane 5). In addition, secretion of the expressed protein in the culture supernatant (lane l) was not detected.
The ~-galactosidase activity was tested by directly incubating pVL-BZ transfected sf9 cells with l.25 lluvi PNP-a-gai (pH 6.5j. During the incubation, the PNP-~-galactosidase substrate enters the cells by diffusion.
After incubation at 37~C for an hour the reaction was stopped by adding l ml of borate buffer (pH 9.8) and absorbance at 405 nm was measured. The total proteins in the reaction were precipitated by addition of trichloroacetate and measured by Bio-Rad Protein Assay (Bio-Rad). The average activity of ~-galactosidase expressed in the insect cells was approximately 300 uni~s, one unit being defined as l nmol of substrate hydrolyzed per hour at 37~C. The endogenous activity in the wild-type virus infected cells was undetectable under such conditions.
The recombinant coffee bean ~-galactosidase enzyme of the invention can be used to remove terminal galactose residues from the non-reducing end of carbohydrate chains (from polysaccharides and glycoconjugates), particularly those galactose residues which are ~l,3-linked.
The recombinant coffee bean ~-galactosidase enzyme of the invention can be used to convert type B erythrocytes to type O and type AB erythrocytes to type A. Specifically, the recombinant coffee bean ~-galactosidase enzyme of the invention is put into contact with cells having B

CA 022ll4l7 l997-07-2~
W 096/23869 PCTrUS96/01212 antigenicity for a period of time sufficient to remove the B antigens from the surface of the cells. More specifically, erythrocytes having B antigenicity are washed and then equilibrated in isotonic phosphate-citrate-sodium 5 chloride (PCS) at pH 5.5 and 5.6, sequentially, c~-galactosidase is added at a concentration of from 75 to 200 U/ml, and the mixture is incubated at 26~C. More preferably, the ~-galactosidase is added at a concentration of 200 U/ml and the mixture is incubated at 26~C for 2.25 10 hours. Most preferably, the final hematocrit of the erythrocytes is 65 to 75g~. In addition, cells transformed with a recombinant vector which encodes coffee bean ~x-galactosidase can be cultured and coffee bean ~-galactosidase can be recovered from the culture, which 15 coffee bean ~-galactosidase can then be used to remove B
antigens from the surface of cells and blood products.

Expression of Coffee Bean ~-Galactosidase in Pichia pastoris Coffee bean ~-galactosidase cDNA was subcloned in 20 the EcoRI site of Pichia pastoris expression vector pPIC9 (Invitrogen Corp., San Diego, CA) generating the plasmid p~F-BZ with the sequence around the 5' cloning site of the insert as shown in Figure 5. The expression of ~x-galactosidase was under the control of the methanol 25 inducable promoter AOX1. The expressed protein was secreted into the culture media via the signal sequence of yeast ~
mating factor. After transformation of Pichia pastoris, one transformant with the highest ~-galactosidase activity was selected for large-scale production in a fermentor.
30 Approximately eleven copies of the ~-galactosidase was incorporated into the Pichia pastoris genome based upon a dot blot experiment. The level of ~-galactosidase secreted into the Pichia pastoris culture media reached 400 mg/L of culture. After removal of the Pichia pastoris cells from 35 the fermentation culture, the recombinant ~-galactosidase in the supernatant was purified by cation exchange chromatography. The chromatogram from the purification is -CA 02211417 1997-07-2~
W 096/23869 PCTrUS96/01212 presented in Figure 6. The chromatography purified enzyme was then subjected to SDS-PAGE analysis.
N-terminal sequencing of the purified protein ~ suggested that the signal sequence was cleaved at two positions as indicated by the two arrows in Figure 5, generating two forms of ~-galactosidase with Phe or Leu as the N-terminal residue in approximately equal molarity.
Characterization of the native coffee bean ~-galactosidase and the recombinant coffee bean ~-galactosidase indicated that despite the mixture of N-terminal amino acid residues, the recombinant ~-galactosidase exhibited essentially the same molecular size, specific activity, pH optima and kinetic parameters as the native enzyme.

Although the invention herein has been described with reference to particular embodiments, it is to be understood that these embodiments are merely illustrative of various aspects of the invention~ Thus, it is to be understood that numerous modifications may be made in the illustrative embodiments and other arrangements may be devised without departing from the spirit and scope of the invention.

W 096/23869 PCTrUS96/01212 SEOUENCE LISTING
(1) GENERAL INFORMATION
(i) APPLICANT: ALEX ZHU AND JACK GOLDSTEIN
(ii) TITLE OF INVENTION: RECOMBINANT
ALPHA-GALACTOSIDASE ENZYME AND CDNA ENCODING SAID
ENZYME
(iii~ NUMBER OF SEQUENCES: 11 (iv) CORRESPONDENCE ADDRESS:
(A) ADDRESSEE: AMSTER, ROTHSTEIN & EBENSTEIN
(B) STREET: 90 PARK AVENUE
(C) CITY: NEW YORK
(D) STATE: NEW YORK
(E) COUNTRY: U.S.A.
(F) ZIP: 10016 (v) COh~U'l'~K READABLE FORM:
(A) MEDIUM TYPE: 3. 5 INCH 1. 44 Mb STORAGE
DI~K~
(B) COh~U'1'~;K: IBM PC COMPATIBLE
(C) OPERATING SYSTEM: MS-DOS
(D) SOFTWARE: ASCII
(vi) CURRENT APPLICATION DATA:
(A) APPLICATION NUMBER: NOT YET ASSIGNED
(B) FILING DATE: NOT YET ASSIGNED
(viii) ATTORNEY/AGENT INFORMATION:
(A) NAME: CRAIG J. ARNOLD
(B) REGISTRATION NUMBER: 34,287 (C) REFERENCE/DOCKET NUMBER: 63475/71 (iX) TELECOMMUNICATION INFORMATION:
(A) TELEPHONE: (212) 697-5995 (B) TELEFAX: (212) 286-0854 Or 286-0082 (C) TELEX: TWX 710-581-4766 (2) INFORMATION FOR SEQ ID NO: 1 (i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 1409 (B) TYPE: NUCLEIC ACID
(C) STRANDEDNESS: DOUBLE
(D) TOPOLOGY: LINEAR
(ii) MOLECULE TYPE:
(A) DESCRIPTION: OLIGONUCLEOTIDE
(iii) HYPOTHETICAL: NO
(iv) ANTI-SENSE: NO

CA 02211417 1997-07-2~

(vi ) ORIGINAL SOURCE:
(A) ORGANISM: GREEN COFFEE BEAN
(B) INDIVIDUAL ISOLATE: ALPHA-GALACTOSIDASE
( xi ) SEQUENCE DESCRIPTION: SEQ ID NO: 1 Met Val Lys Ser Pro Gly Thr Glu Asp Tyr Thr Arg Arg Ser Leu Leu Ala Asn Gly Leu Gly Leu Thr Pro Pro Met Gly Trp Asn Ser Trp Asn His Phe Arg Cys Asn Leu Asp Glu Lys Leu Ile Arg Glu Thr Ala Asp Ala Met Val Ser Lys Gly Leu Ala Ala Leu Gly Tyr Lys Tyr Ile Asn Leu Asp Asp Cys Trp Ala Glu Leu Asn Arg Asp Ser Gln Gly Asn Leu Val Pro Lys Gly Ser Thr Phe Pro Ser Gly Ile Lys Ala Leu Ala Asp Tyr Val His Ser Lys Gly Leu Lys Leu Gly Ile Tyr Ser Asp Ala Gly Thr Gln Thr Cys Ser Lys Thr Met Pro Gly Ser Leu Gly His Glu Glu Gln Asp Ala Lys Thr Phe Ala Ser Trp Gly Val Asp Tyr Leu Lys Tyr Asp Asn Cys Asn Asn Asn Asn Ile Ser Pro Lys Glu Arg Tyr Pro I le Met Ser Lys Ala Leu Leu Asn Ser Gly Arg Ser I le Phe Phe -CA 022ll4l7 l997-07-2~

W 096/23869 PCTrUS96/01212 Ser Leu Cys Glu Trp Gly Glu Glu Asp Pro Ala Thr Trp Ala Lys Glu Val Gly Asn Ser Trp Arg Thr Thr Gly Asp Ile Asp Asp Ser Trp Ser Ser Met Thr Ser Arg Ala Asp Met Asn Asp Lys Trp Ala Ser Tyr Ala Gly Pro Gly Gly Trp Asn Asp Pro Asp Met Leu Glu Val Gly Asn Gly Gly Met Thr Thr Thr Glu Tyr Arg Ser His Phe Ser Ile Trp Ala Leu Ala Lys Ala Pro Leu Leu Ile Gly Cys Asp Ile Arg Ser Met Asp Gly Ala Thr Phe Gln Leu Leu Ser Asn Ala Glu Val Ile Ala Val Asn Gln Asp Lys Leu Gly Val Gln Gly Asn Lys Val Lys Thr Tyr Gly Asp Leu Glu Val Trp Ala Gly Pro Leu Ser Gly Lys Arg Val Ala Val Ala Leu Trp Asn Arg Gly Ser Ser Thr Ala Thr Ile Thr Ala Tyr Trp Ser Asp Val Gly Leu Pro Ser Thr Ala Val Val Asn Ala Arg Asp Leu Trp Ala His Ser Thr Glu Lys Ser Val Lys Gly Gln Ile Ser Ala Ala Val Asp Ala His Asp Ser Lys Met Tyr Val Leu Thr Pro Gln ***

CA 02211417 1997-07-2~

W 096/23869 PCTrUS96/01212 -29-(3) INFORMATION FOR SEQ ID NO: 2 (i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 378 (B) TYPE: AMINO ACID
(ii) MOLECULE TYPE:
(A) DESCRIPTION: PROTEIN
(iii) HYPOTHETICAL: NO
(vi) ORIGINAL SOURCE:
(A) ORGANISM: GREEN COFFEE BEAN
(B) INDIVIDUAL ISOLATE: ALPHA-GALACTOSIDASE
(xi) SEQUENCE DESCRIPTION: SEQ ID NO: 2 Met Val Lys Ser Pro Gly Thr Glu Asp Tyr Thr Arg Arg Ser Leu Leu ~la Asn Gly Leu Gly Leu Thr Pro Pro Met Gly Trp Asn Ser Trp Asn His Phe Arg Cys Asn Leu Asp Glu Lys Leu Ile Arg Glu Thr Ala Asp Ala Met Val Ser Lys Gly Leu Ala Ala Leu Gly Tyr Lys Tyr Ile ~sn Leu Asp Asp Cys Trp Ala Glu Leu Asn Arg Asp Ser Gln Gly Asn Leu ~al Pro Lys Gly Ser Thr Phe Pro Ser Gly Ile Lys Ala Leu Ala Asp ~yr Val His Ser Lys Gly Leu Lys Leu Gly Ile Tyr Ser Asp Ala Gly Thr Gln Thr Cys Ser Lys Thr Met Pro Gly Ser Leu Gly His Glu Glu Gln Asp Ala Lys Thr Phe Ala Ser Trp Gly Val Asp Tyr Leu Lys Tyr Asp Asn Cys Asn Asn Asn Asn Ile Ser Pro Lys Glu Arg Tyr Pro Ile Met Ser Lys Ala Leu Leu Asn Ser Gly Arg Ser Ile Phe Phe Ser Leu CA 02211417 1997-07-2~
W 096/23869 PCTrUS96/01212 Cys Glu Trp Gly Glu Glu Asp Pro Ala Thr Trp Ala Lys Glu Val Gly Asn Ser Trp Arg Thr Thr Gly Asp Ile Asp Asp Ser Trp Ser Ser Met Thr Ser Arg Ala Asp Met Asn Asp Lys Trp Ala Ser Tyr Ala Gly Pro Gly Gly Trp Asn Asp Pro Asp Met Leu Glu Val Gly Asn Gly Gly Met ~hr Thr Thr Glu Tyr Arg Ser His Phe Ser Ile Trp Ala Leu Ala Lys ~la Pro Leu Leu Ile Gly Cys Asp Ile Arg Ser Met Asp Gly Ala Thr Phe Gln Leu Leu Ser Asn Ala Glu Val Ile Ala Val Asn Gln Asp Lys Leu Gly Val Gln Gly Asn Lys Val Lys Thr Tyr Gly Asp Leu Glu Val Trp Ala Gly Pro Leu Ser Gly Lys Arg Val Ala Val Ala Leu Trp Asn ~rg Gly Ser Ser Thr Ala Thr Ile Thr Ala Tyr Trp Ser Asp Val Gly ~eu Pro Ser Thr Ala Val Val Asn Ala Arg Asp Leu Trp Ala His Ser ~hr Glu Lys Ser Val Lys Gly Gln Ile Ser Ala Ala Val Asp Ala His Asp Ser Lys Met Tyr Val Leu Thr Pro Gln (4) INFORMATION FOR SEQ ID NO: 3 (i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 411 (B) TYPE: AMINO ACID
(ii) MOLECULE TYPE:
(A) DESCRIPTION: PROTEIN
(iii) HYPOTHETICAL: NO
(vi) ORIGINAL SOURCE:
(A) ORGANISM: CYAMOPSIS TETRAGONOLOBA
(B) INDIVIDUAL ISOLATE: ALPHA-GALACTOSIDASE

CA 02211417 1997-07-2~
W096/23869 PCT~S96101212 (xi) SEQUENCE DESCRIPTION: SEQ ID NO: 3 Met Ala Thr His Tyr Ser Ile Ile Gly Gly Met Ile Ile Val Val Leu ~eu Met Ile Ile Gly Ser Glu Gly Gly Arg Leu Leu Glu Lys Lys Asn ~rg Thr Ser Ala Glu Ala Glu His Tyr Asn Val Arg Arg Tyr Leu Ala Glu Asn Gly Leu Gly Gln Thr Pro Pro Met Gly Trp Asn Ser Trp Asn His Phe Gly Cys Asp Ile Asn Glu Asn Val Val Arg Glu Thr Ala Asp ~la Met Val Ser Thr Gly Leu Ala Ala Leu Gly Tyr Gln Tyr Ile Asn ~eu Asp Asp Cys Trp Ala Glu Leu Asn Arg Asp Ser Glu Gly Asn Met Val Pro Asn Ala Ala Ala Phe Pro Ser Gly Ile Lys Ala Leu Ala Asp Tyr Val His Ser Lys Gly Leu Lys Leu Gly Val Tyr Ser Asp Ala Gly Asn Gln Thr Cys Ser Lys Arg Met Pro Gly Ser Leu Gly His Glu Glu ~ln Asp Ala Lys Thr Phe Ala Ser Trp Gly Val Asp Tyr Leu Lys Tyr ~sp Asn Cys Glu Asn Leu Gly Ile Ser Val Lys Glu Arg Tyr Pro Pro Met Gly Lys Ala Leu Leu Ser Ser Gly Arg Pro Ile Phe Phe Ser Met Cys Glu Trp Gly Trp Glu Asp Pro Gln Ile Trp Ala Lys Ser Ile Gly Asn Ser Trp Arg Thr Thr Gly Asp Ile Glu Asp Asn Trp Asn Ser Met ~hr Ser Ile Ala Asp Ser Asn Asp Lys Trp Ala Ser Tyr Ala Gly Pro ~ly Gly Trp Asn Asp Pro Asp Met Leu Glu Val Gly Asn Gly Gly Met ~hr Thr Glu Glu Tyr Arg Ser His Phe Ser Ile Trp Ala Leu Ala Lys W 096/23869 PCTrUS96/01212 Ala Pro Leu Leu Val Gly Cys Asp Ile Arg Ala Met Asp Asp Thr Thr His Glu Leu Ile Ser Asn Ala Glu Val Ile Ala Val Asn Gln Asp Lys ~eu Gly Val Gln Gly Lys Lys Val Lys Ser Thr Asn Asp Leu Glu Val ~rp Ala Gly Pro Leu Ser Asp Asn Lys Val Ala Val Ile Leu Trp Asn Arg Ser Ser Ser Arg Ala Thr Val Thr Ala Ser Trp Ser Asp Ile Gly Leu Gln Gln Gly Thr Thr Val Asp Ala Arg Asp Leu Trp Glu His Ser Thr Gln Ser Leu Val Ser Gly Glu Ile Ser Ala Glu Ile Asp Ser His Ala Cys Lys Met Tyr Val Leu Thr Pro Arg Ser (5) INFORMATION FOR SEQ ID NO: 4 (i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 429 (B) TYPE: AMINO ACID
(ii) MOLECULE TYPE:
(A) DESCRIPTION: PROTEIN
(iii) HYPOTHETICAL: NO
(vi) ORIGINAL SOURCE:
(A) ORGANISM: HUMAN PLACENTA
(B) INDIVIDUAL ISOLATE: ALPHA-GALACTOSIDASE
(xi) SEQUENCE DESCRIPTION: SEQ ID NO: 4 Met Gln Leu Arg Asn Pro Glu Leu His Leu Gly Cys Ala Leu Ala Leu ~rg Phe Leu Ala Leu Val Ser Trp Asp Ile Pro Gly Ala Arg Ala Leu Asp Asn Gly Leu Ala Arg Thr Pro Thr Met Gly Trp Leu His Trp Glu Arg Phe Met Cys Asn Leu Asp Cys Gln Glu Glu Pro Asp Ser Cys Ile Ser Glu Lys Leu Phe Met Glu Met Ala Glu Leu Met Val Ser Glu Gly CA 02211417 1997-07-2~
W 096/23869 PCTrUS96/01212 ~rp Lys Asp Ala Gly Tyr Glu Tyr Leu Cys I le Asp Asp Cys Trp Met ~la Pro Gln Arg Asp Ser Glu Gly Arg Leu Gln Ala Asp Pro Gln Arg Phe Pro His Gly Ile Arg Gln Leu Ala Asn Tyr Val His Ser Lys Gly Leu Lys Leu Gly Ile Tyr Ala Asp Val Gly Asn Lys Thr Cys Ala Gly Phe Pro Gly Ser Phe Gly Tyr Tyr Asp Ile Asp Ala Gln Thr Phe Ala ~sp Trp Gly Val Asp Leu Leu Lys Phe Asp Gly Cys Tyr Cys Asp Ser ~eu Glu Asn Leu Ala Asp Gly Tyr Lys His Met Ser Leu Ala Leu Asn Arg Thr Gly Arg Ser Ile Val Tyr Ser Cys Glu Trp Pro Leu Tyr Met Trp Pro Phe Gln Lys Pro Asn Tyr Thr Glu Ile Arg Gln Tyr Cys Asn His Trp Arg Asn Phe Ala Asp Ile Asp Asp Ser Trp Lys Ser Ile Lys ~er Ile Leu Asp Trp Thr Ser Phe Asn Gln Glu Arg Ile Val Asp Val ~la Gly Pro Gly Gly Trp Asn Asp Pro Asp Met Leu Val Ile Gly Asn Phe Gly Leu Ser Trp Asn Gln Gln Val Thr Gln Met Ala Leu Trp Ala Ile Met Ala Ala Pro Leu Phe Met Ser Asn Asp Leu Arg His Ile Ser Pro Gln Ala Lys Ala Leu Leu Gln Asp Lys Asp Val Ile Ala Ile Asn ~ln Asp Pro Leu Gly Lys Gln Gly Tyr Gln Leu Arg Gln Gly Asp Asn ~he Glu Val Trp Glu Arg Pro Leu Ser Gly Leu Ala Trp Ala Val Ala Met Ile Asn Arg Gln Glu Ile Gly Gly Pro Arg Ser Tyr Thr Ile Ala Val Ala Ser Leu Gly Lys Gly Val Ala Cys Asn Pro Ala Cys Phe Ile CA 02211417 1997-07-2~

W 096/23869 PCTrUS96/01212 Thr Gln Leu Leu Pro Val Lys Arg Lys Leu Gly Phe Tyr Glu Trp Thr Ser Arg Leu Arg Ser His Ile Asn Pro Thr Gly Thr Val Leu Leu Gln ~eu Glu Asn Thr Met Gln Met Ser Leu Lys Asp Leu Leu ~6) INFORMATION ~OR SEQ ID NO: 5 (i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 432 (B) TYPE: AMINO ACID
(ii) MOLECULE TYPE:
(A) DESCRIPTION: PROTEIN
(iii) HYPOTHETICAL: NO
(vi) ORIGINAL SOURCE:
(A) ORGANISM: SACCHAROMYCES CEREVISIAE
(B) INDIVIDUAL ISOLATE: ALPHA-GALACTOSIDASE
(xi) SEQUENCE DESCRIPTION: SEQ ID NO: 5 Met Phe Ala Phe Tyr Phe Leu Thr Ala Cys Ile Ser Leu Lys Gly Val ~he Gly Val Ser Pro Ser Tyr Asn Gly Leu Gly Leu Thr Pro Gln Met ~ly Trp Asp Asn Trp Asn Thr Phe Ala Cys Asp Val Ser Glu Gln Leu Leu Leu Asp Thr Ala Asp Arg Ile Ser Asp Leu Gly Leu Lys Asp Met Gly Tyr Lys Tyr Ile Ile Leu Asp Asp Cys Trp Ser Ser Gly Arg Asp ~er Asp Gly Phe Leu Val Ala Asp Glu Gln Lys Phe Pro Asn Gly Met ~ly His Val Ala Asp His Leu His Asn Asn Ser Phe Leu Phe Gly Met ~yr Ser Ser Ala Gly Glu Tyr Thr Cys Ala Gly Tyr Pro Gly Ser Leu Gly Arg Glu Glu Glu Asp Ala Gln Phe Phe Ala Asn Asn Arg Val Asp Tyr Leu Lys Tyr Asp Asn Cys Tyr Asn Lys Gly Gln Phe Gly Thr Pro CA 022ll4l7 l997-07-2~
W O 96/23869 PCTrUS96/01212 ~lu Ile Ser Tyr His Arg Tyr Lys Ala Met Ser Asp Ala Leu Asn Lys 165 17û 175 ~hr Gly Arg Pro I le Phe Tyr Ser Leu Cys Asn Trp Gly Gln Asp Leu Thr Phe Tyr Trp Gly Ser Gly Ile Ala Asn Ser Trp Arg Met Ser Gly Asp Val Thr Ala Glu Phe Thr Arg Pro Asp Ser Arg Cys Pro Cys Asp Gly Asp Glu Tyr Asp Cys Lys Tyr Ala Gly Phe His Cys Ser Ile Met ~sn Ile Leu Asn Lys Ala Ala Pro Net Gly Gln Asn Ala Gly Val Gly ~ly Trp Asn Asp Leu Asp Asn Leu Glu Val Gly Val Gly Asn Leu Thr Asp Asp Glu Glu Lys Ala His Phe Ser Met Trp Ala Met Val Lys Ser Pro Leu Ile Ile Gly Ala Asn Val Asn Asn Leu Lys Ala Ser Ser Tyr Ser Ile Tyr Ser Gln Ala Ser Val Ile Ala Ile Asn Gln Asp Ser Asn ~ly Ile Pro Ala Thr Arg Val Trp Arg Tyr Tyr Val Ser Asp Thr Asp ~lu Tyr Gly Gln Gly Glu Ile Gln Met Trp Ser Gly Pro Leu Asp Asn Gly Asp Gln Val Val Ala Leu Leu Asn Gly Gly Ser Val Ser Arg Pro Met Asn Thr Thr Leu Glu Glu I le Phe Phe Asp Ser Asn Leu Gly Ser Lys Lys Leu Thr Ser Thr Trp Asp Ile Tyr Asp Leu Trp Ala Asn Arg ~al Asp Asn Ser Thr Ala Ser Ala Ile Leu Gly Arg Asn Lys Thr Ala ~hr Gly Ile Leu Tyr Asn Ala Thr Glu Gln Ser Tyr Lys Asp Gly Leu (7) INFORMATION FOR SEQ ID NO: 6 ( i ) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 441 CA 022ll4l7 l997-07-2~
W 096/23869 PCTrUS96/01212 (B) TYPE: AMINO ACID
(ii) MOLECULE TYPE:
(A) DESCRIPTION: PROTEIN
(iii) HYPO~ CAL: NO
(vi) ORIGINAL SOURCE:
(A) ORGANISM: ASPERGILLUS NIGER
(B) INDIVIDUAL ISOLATE: ALPHA-GALACTOSIDASE
(xi) SE~UENCE DESCRIPTION: SEQ ID NO: 6 Met Ile Gln Gly Leu Glu Ser Ile Met Asn Gln Gly Thr Lys Arg Ile 1 5 10 15~eu Leu Ala Ala Thr Leu Ala Ala Thr Pro Trp Gln Val Tyr Gly Ser ~le Glu Gln Pro Ser Leu Leu Pro Thr Pro Pro Met Gly Phe Asn Asn Trp Ala Arg Phe Met Cys Asp Leu Asn Glu Thr Leu Phe Thr Glu Thr Ala Asp Thr Met Ala Ala Asn Gly Leu Arg Asp Ala Gly Tyr Asn Arg ~le Asn Leu Asp Asp Cys Trp Met Ala Tyr Gln Arg Ser Asp Asn Gly ~er Leu Gln Trp Asn Thr Thr Lys Phe Pro His Gly Leu Pro Trp Leu ~la Lys Tyr Val Lys Ala Lys Gly Phe His Phe Gly Ile Tyr Glu Asp Ser Gly Asn Met Thr Cys Gly Gly Tyr Pro Gly Ser Tyr Asn His Glu Glu Gln Asp Ala Asn Thr Phe Ala Ser Trp Gly Ile Asp Tyr Leu Lys ~eu Asp Gly Cys Asn Val Tyr Ala Thr Gln Gly Arg Thr Leu Glu Glu ~lu Tyr Lys Gln Arg Tyr Gly His Trp His Gln Val Leu Ser Lys Met ~ln His Pro Leu Ile Phe Ser Glu Ser Ala Pro Ala Tyr Phe Ala Gly ~hr Asp Asn Asn Thr Asp Trp Tyr Thr Val Met Asp Trp Val Pro Ile CA 02211417 1997-07-2~
W 096/23869 PCTrUS96/01212 Tyr Gly Glu Leu Ala Arg His Ser Thr Asp Ile Leu Val Tyr Ser Gly Ala Gly Ser Ala Trp Asp Ser Ile Met Asn Asn Tyr Asn Tyr Asn Thr ~eu Leu Ala Arg Tyr Gln Arg Pro Gly Tyr Phe Asn Asp Pro Asp Phe ~eu Ile Pro Asp His Pro Gly Leu Thr Ala Asp Glu Lys Arg Ser His Phe Ala Leu Trp Ala Ser Phe Ser Ala Pro Leu Ile Ile Ser Ala Tyr Ile Pro Ala Leu Ser Lys Asp Glu Ile Ala Phe Leu Thr Asn Glu Ala ~eu Ile Ala Val Asn Gln Asp Pro Leu Ala Gln Gln Ala Thr Leu Ala ~er Arg Asp Asp Thr Leu Asp Ile Leu Thr Arg Ser Leu Ala Asn Gly ~sp Arg Leu Leu Thr Val Leu Asn Lys Gly Asn Thr Thr Val Thr Arg Asp Ile Pro Val Gln Trp Leu Gly Leu Thr Glu Thr Asp Cys Thr Tyr Thr Ala Glu Asp Leu Trp Asp Gly Lys Thr Gln Lys Ile Ser Asp His ~le Lys Ile Glu Leu Ala Ser His Ala Thr Ala Val Phe Arg Leu Ser ~eu Pro Gln Gly Cys Ser Ser Val Val Pro Thr Gly Leu Val Phe Asn ~hr Ala Ser Gly Asn Cys Leu Thr Ala ~8) INFORMATION FOR SEQ ID NO: 7 (i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 18 (B) TYPE: NUCLEIC ACID
(C) STRANDEDNESS: SINGLE
(D) TOPOLOGY: UNKNOWN
(ii) MOLECULE TYPE:
(A) DESCRIPTION: OLIGONUCLEOTIDE
(iii) HYPOTHETICAL: NO

CA 02211417 1997-07-2~

(iv) ANTI-SENSE: NO
(vi) ORIGINAL SOURCE:
(A) ORGANISM: GREEN COFFEE BEAN
(B) INDIVIDUAL ISOLATE: ALPHA-GALACTOSIDASE
(xi) SEQUENCE DESCRIPTION: SEQ ID NO: 7 (9) INFORMATION FOR SEQ ID NO: 8 (i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 20 (B) TYPE: NUCLEIC ACID
(C) STRANDEDNESS: SINGLE
(D) TOPOLOGY: UNKNOWN
(ii) MOLECULE TYPE:
(A) DESCRIPTION: OLIGONUCLEOTIDE
(iii) HYPOTHETICAL: NO
(iv) ANTI-SENSE: NO
(vi) ORIGINAL SOURCE:
(A) ORGANISM: GREEN COFFEE BEAN
(B) INDIVIDUAL ISOLATE: ALPHA-GALACTOSIDASE
(xi) SEQUENCE DESCRIPTION: SEQ ID NO: 8 (10) INFORMATION FOR SEQ ID NO: 9 (i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 18 (B) TYPE: NUCLEIC ACID
(C) STRANDEDNESS: SINGLE
(D) TOPOLOGY: UNKNOWN
(ii) MOLECULE TYPE:
(A) DESCRIPTION: OLIGONUCLEOTIDE
(iii~ HYPOTHETICAL: NO
(iv) ANTI-SENSE: NO
(vi) ORIGINAL SOURCE:
(A) ORGANISM: GREEN COFFEE BEAN
(B) INDIVIDUAL ISOLATE: ALPHA-GALACTOSIDASE
(Xi) SEQUENCE DESCRIPTION: SEQ ID NO: 9 CA 02211417 1997-07-2~
W096/23869 PCT~S96/01212 (11) INFORMATION FOR SEQ ID NO: 10 (i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 66 (B) TYPB: NUCLEIC ACID
(C) STRANDEDNESS: DOUBLE
(D) TOPOLOGY: UNKNOWN
(ii) MOLECULE TYPE:
(A) DESCRIPTION: OLIGONUCLEOTIDE
(iii) HYPOTHETICAL: NO
(iv) ANTI-SENSE: NO
(vi) ORIGINAL SOURCE:
(B) INDIVIDUAL ISOLATE: SIGNAL SEQUENCE OF ALPHA
MATING FACTOR AND N-TERMINAL OF
ALPHA-GALACTOSIDASE
(xi) SEQUENCE DESCRIPTION: SEQ ID NO: 10 (12) INFORMATION FOR SEQ ID NO: 11 (i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 22 (B) TYPE: AMINO ACID
(ii) MOLECULE TYPE:
(A) DESCRIPTION: PEPTIDE
(iii) HYPOTHETICAL: NO
(vi) ORIGINAL SOURCE:
(B) INDIVIDUAL ISOLATE: SIGNAL SEQUENCE OF ALPHA
MATING FACTOR AND N-TERMINAL OF
ALPHA-GALACTOSIDASE
(xi) SEQUENCE DESCRIPTION: SEQ ID NO: 11 Val Set Leu Glu Lys Lys Arg Glu Ala Glu Ala Tyr Val Glu Phe Leu Ala Asn Gly Leu Gly Leu

Claims (15)

WE CLAIM:

1. A recombinant coffee bean .alpha.-galactosidase produced by a Pichia pastoris cell transformed with a vector comprising a nucleic acid encoding coffee bean .alpha.-galactosidase.

2. The recombinant coffee bean .alpha.-galactosidase of Claim 1, wherein said nucleic acid encoding coffee bean .alpha.-galactosidase enzyme is contained in the vector deposited under ATTC No. 75556.

3. The recombinant coffee bean .alpha.-galactosidase of Claim 1, wherein said nucleic acid encoding coffee bean .alpha.-galactosidase enzyme is contained in Figure 1.

4. A Pichia pastoris expression vector comprising a nucleic acid encoding coffee bean .alpha.-galactosidase enzyme.

5. The expression vector of Claim 4, wherein said nucleic acid encoding coffee bean .alpha.-galactosidase enzyme is contained in the vector deposited under ATTC No.
75556.

6. The expression vector of Claim 4, wherein said nucleic acid encoding coffee bean .alpha.-galactosidase enzyme is contained in Figure 1.

7. A Pichia pastoris cell transformed with a vector comprising a nucleic acid encoding coffee bean .alpha.-galactosidase enzyme.

8. The cell of Claim 7, wherein said nucleic acid encoding coffee bean .alpha.-galactosidase enzyme is contained in the vector deposited under ATTC No. 75556.

9. The cell of Claim 7, wherein said nucleic acid encoding coffee bean .alpha.-galactosidase enzyme is contained in Figure 1.

10. A method for producing recombinant coffee bean .alpha.-galactosidase comprising culturing a Pichia pastoris cell transformed with a vector comprising a nucleic acid encoding coffee bean .alpha.-galactosidase enzyme, and recovering .alpha.-galactosidase from the culture.

11. The method of Claim 10, wherein said nucleic acid encoding coffee bean .alpha.-galactosidase enzyme is contained in the vector deposited under ATTC No. 75556.

12. The method of Claim 10, wherein said nucleic acid encoding coffee bean .alpha.-galactosidase enzyme is contained in Figure 1.

13. The recombinant coffee bean .alpha.-galactosidase produced by the method of Claim 10.

14. The recombinant coffee bean .alpha.-galactosidase produced by the method of Claim 11.

15. The recombinant coffee bean .alpha.-galactosidase produced by the method of Claim 12.