Method for Obtaining a Modified Heparinase Gene

This invention is generally in the area of heparinases and is specifically directed to the gene encoding heparinase I, expressed in Flavobacterium heparinum.

The United States government has rights in this invention by virtue of grant number 25810 from the National Institutes of Health.

Heparin is an anticoagulant that activates serine protease inhibitors (serpins), which play a key role in the blood clotting cascade, as described by Damus et al., Nature 246:355-357 (1973). According to Lindahl et al., Trends Biochem. Sci. 11:221-225 (1986), heparin is the most acidic natural polymer known to date. It consists of a major 1,4-linked disaccharide repeating unit of D-uronic acid 1,4-.beta.-D-glucosamine, and has an average of four negative charges (three sulfate groups and one carboxylate group) per monosaccharide unit. Heparin is both polydisperse, having an average molecular weight between 3,000 and 45,000 daltons, and heterogenous due to partial epimerization of D-glucuronic acid to L-iduronic acid and incomplete N- and O- sulfation, as reported by Kusche et al., Proc. Natl. Acad. Sci., 77:6551-6555 (1980) and Comper, Polymer Monograph 7, 1981.

In addition, proteoglycans like heparin have a wide range of biological influences, including in blood chemistry, growth factor interaction and wound healing, interaction with basic structural proteins in the extracellular matrix and in cellular mediated immune responses. The basic nature of protein/peptide--heparin/complex carbohydrate interaction is important. Although heparin seems fairly heterogenous, it is now quite clear that different heparin fractions exhibit distinct and unique properties indicating some compositional and possibly structural specificity for its biological role, as reviewed by Cardin, A. D. and H. J. R. Weintraub, Arteriosclerosis 9, 21-32 (1989).

Heparinase, also referred to as heparin lyase, is the only known enzyme capable of degrading heparin that has been extensively characterized. It has been designated EC 4.2.2.7 by the Enzyme Commission. According to Galliher, et al., Eur. J. Appl. Microbiol. 15:252 (1982), the enzyme is a polysaccharide lyase found in the periplasmic space of Flavobacterium heparinum, a Gram-negative soil isolate. F. heparinum utilizes heparin as its sole source of carbon and nitrogen, as described by Hoving and Linker, J. Biol. Chem. 245:6170 (1970). Heparinase is the initial enzyme of heparin catabolism. Although constitutively expressed in low amounts, Galliher, et al., App. Environ. Microbiol. 41:360 (1981), have discovered that enzyme expression is induced by heparin and reversibly repressed by sulfate in the medium. Lindhardt, et al., Appl. Biochem. Biotechnol. 9:41 (1984), have shown that heparinase is inhibited by other polyanionic polysaccharides.

Heparinase has been purified by standard chromatographic techniques and its enzymatic properties characterized extensively, as described by scientists including Yang, et al., J. Biol. Chem. 260:1849 (1985). The enzyme is a 44,000 dalton monomeric protein with a pI of approximately 9.

Heparinase acts as an eliminase, leaving behind an unsaturated double bond at the non-reducing end group. This double bond is exploited in an assay for heparinase activity by the absorbance of the unsaturated product at 232 nm. The enzyme is marginally tolerant to salts and is very specific for heparin, having a k.sub.d of 30 nM. Heparinase has an activation energy of 4.5 kcal/mol, a km of 8.times.10-6 and a Vmax of 4.times.10-7 M/min.

Heparin is often used in surgery to prevent blood clotting and to increase the compatibility of extracorporeal devices such as heart-lung and kidney dialysis machines. The enzymatic degradation of heparin by heparinase is sufficient to eliminate the anticoagulation properties of heparin in surgery. As described by Langer, et al. in Biomaterials: Inter-facial Phenomenon and Applications, Adv. in Chem. Symposium Series, Chap. 13, pp. 493-509 (1982), this property has led to the use of heparinase as an immobilized bioreactor in conjunction with heart-lung or kidney dialysis machines to deheparinize blood. Commercial application of the heparinase bioreactor is pending clinical trials.

A principal problem in the use of the heparinase bioreactor is the availability of sufficient amounts of pure heparinase to be immobilized onto a surface. This is primarily because the amount of heparinase constitutively expressed in F. heparinum is very low. Inducing expression of heparinase in F. heparinum with heparin is very expensive due to the amounts of heparin needed and the size of the fermentation to produce reasonable amounts of heparinase for any practical applications.

Cloning and expression of the heparinase gene is important in several ways. First, the only enzyme cloned and characterized to date which acts to depolymerise proteoglycans is heparinase. Second, heparin is the only anticoagulant commonly used in surgery so deheparinizing blood is an important medical problem. Moreover, heparinase catalyzed degradation of heparin into lower molecular weight heparin molecules can be used to yield products with specific anticoagulant activity, as discussed by Rosenfeld and Danishefsky, Biochem. J. 237:639-646 (1986).

Designing recombinant heparinases with altered activitie(s) would be interesting academically, as well as commercially. For example, heparinase can be used to deheparinize blood because the enzyme cleaves right at the AT-III binding oligomer. On the other hand, by further understanding the mechanism of the enzyme binding and depolymerizing heparin, recombinant heparinases with altered specificity could be designed, i.e. an AT-III binding heparin fragment not cleaved by the recombinant enzyme. This would be a very useful way of generating an AT-III binding heparin oligosaccharide, which currently is not available in large amounts, for use as an anticoagulant. Producing heparinases which could help and or improve in the enzyme purification or immobilization would also be quite valuable. For example, a tag (a particular peptide sequence) could be added at a region which does not alter the activity of the enzyme but makes the immobilization chemistry very efficient. This would help in improving enzyme loading onto the immobilization matrix.

It is therefore an object of the present invention to provide the gene encoding heparinase and a system for expression to facilitate the production of large amounts of heparinase.

It is another object of the present invention to provide methods and means for modifying the gene to produce recombinant heparinases having altered specificity and other desirable properties.

It is another object of the present invention to provide pure heparinase for use in the area of cytokine-proteoglycan interactions, as a tool or diagnostic as exemplified by fibroblast growth factor--heparin interactions.

The cloning of the heparinase gene from Flavobacterium Heparinum using the polymerase chain reaction is described. Two degenerate oligonucleotides, based on amino acid sequence derived from tryptic peptides of purified heparinase were used in the PCR with Flavobacterium genomic DNA as the template to generate a 600 base pairs probe. This probe was used to screen a pUC 18 Flavobacterium genomic library. The Open Reading Frame (ORF) corresponded to 1152 base pairs encoding a precursor protein of MW 43,800 daltons. Eleven different tryptic peptides (approximately 48% of the total amino acids) mapped into the ORF. The amino acid sequence reveals a 20-residue leader peptide.

Heparinase can be expressed from the gene. Additionally, the gene can be modified to produce heparinase with altered enzymatic activity, specificity, or binding properties. The sequence can also be used as a probe in the isolation of genes encoding other related enzymes.

FIG. 1 is a schematic representation of the PCR products Yl:C and D:C which are 600 and 160 basepairs, respectively. The 600 basepair PCR product was used as a template with D and C as primers to generate the 160 basepair D:C product.

FIG. 2 is the restriction map of the genomic DNA pUC 18 plasmid, pRS.HEP51, having an insert containing the heparinase gene. The plasmid is 5631 bases long and has approximately 2300 bases of insert. The heparinase gene is in the KpnI--KpnI fragment.

FIG. 3 is a KpnI--KpnI fragment map showing the heparinase gene structure with the different tryptic peptides mapping into the open reading frame. Six different peptides mapped into the heparinase gene translation region.

The gene encoding heparinase in F. heparinum has been cloned, wherein the coding and non-coding sequences are set forth in Sequence ID No. 1. The nucleotide and amino acid sequences are shown below:

The following sequence encodes a leader peptide, shown at positions 1 through 172 in Sequence ID No. 1: ##STR1##

The following sequence encodes the heparinase, shown at positions 173 through 1327 of Sequence ID No. 1: ##STR2##

The following is the amino acid sequence of heparinase, set forth in Sequence ID No. 2: ##STR3##

Because preliminary cloning attempts by others utilizing 1) antibody screening, 2) screening for functionally active heparinase in E. coli and 3) screening for the heparinase gene using probes derived from protein sequences regenerated by cyanogen bromine (CNBr) chemical digest were unsuccessful, the polymerase chain reaction was used to clone the heparinase gene. The reverse phase purified heparinase was reduced, alkylated and digested with trypsin to obtain approximately 60 peptide peaks which were separated and collected by reverse phase HPLC monitored at 210 nm and at 277 nm (for tyrosine and tryptophan), as described below.

Tryptic Digest and Protein Sequence Analyses

Heparinase was purified as described by Dietrich et al., J. Biol. Chem. 248:6408 (1973), Otatani et al., Carbohyd. Res. 88:291 (1981), and Yang et al., J. Biol. Chem. 260:1849 (1985), which are incorporated by reference herein. A final purification step was carried out by High Performance Liquid Chromatography (HPLC) using a reverse phase column that exploits the hydrophobic residues of the protein. A nanomole (approximately 45 .mu.g) of the purified enzyme was denatured in 50 .mu.l of an 8M urea, 0.4M ammonium carbonate solution, reduced with 5 mM dithiothreitol (DTT) at 50.degree. C., cooled to room temperature, and alkylated with 10 mM iodoacetamide for 15 minutes in the dark. The total reaction volume was 200 .mu.l. To this reaction mixture, 1/25th w/w of trypsin was added and digestion carried out at 37.degree. C. for 24 hour. The reaction was terminated by heating the sample at 65.degree. C. for 2 minutes. The digest was separated by reverse phase HPLC using a gradient of 0 to 80% acetonitrile. The tryptic peptides were monitored at 210 and 277 nm.

The tryptic peaks were collected in Eppendorff tubes. Based on the homogeneity of the peptide peak, eight different peaks were sequenced using an Applied Biosystems sequencer, model 477, with an on-line model 120 PTH amino acid analyzer located in the Biopolymers lab, Center for Cancer Research, MIT. The sequences are set forth in Table I below. The designation (K,R) is used in Table I to indicate that trypsin cuts at either lysine or arginine residues. The asterisks in Table I represent amino acids that could not be determined. The peptide designated td Lx is the longest peptide sequenced having 38 residues. Native undigested heparinase was also sequenced to determined the N-terminus amino acids.

Three sets of primers were designed and synthesized, as shown in Table II. Primers were synthesized with an Applied Biosystems sequencer, model 477, with an on-line model 120 PTH amino acid analyzer located in the Biopolymers lab, Center for Cancer Research, MIT. These primer sets were used in the PCR amplification system for cloning the heparinase gene. The symbol "I" represents the nucleotide inosine. The amino acids of each peptide, depicted in boldface type, represent the residues chosen for the primer design. Two different sets of primers were constructed for tryptic peptide 33 to reduce the degree of inosine substitution at the 3' end of the primer.

Of the six RHPLC peaks initially sequenced (Table I), three were chosen for primer design. Three sets of primers were designed (Table II). The PCR product of the combination the primers td43 and td33 was about 150 base pairs in length. The combination of td4 and td33 primers were about 600 base pairs. Primer td43 was 5' to primer td33 and primer td4 was 5' to td43 primer. Using the PCR product of td4 and td33 as a template and td43 and td4 as primers the predicted 150 base pair product was obtained confirming that td43 was between td4 and td33.

The 600 basepair product shown in FIG. 1 represents about 51% of the approximated total 1170 base pairs for the heparinase gene, assuming 43,000 dalton for heparinase and a 110 dalton average amino acid with a molecular weight corresponding to about 390 amino acids times three which is 1170 bases.

The 600 base pair probe was chosen for screening a pUC 18 library by high stringency colony hybridization. Two positive clones were identified which were carried through for three rounds of colony purification.

Genomic DNA, RNA, and Plasmid Library

The F. heparinum genomic DNA was isolated by the A.S.A.P..TM. kit (Boehringer Mannheim, Indianapolis, Ind.) with the following modifications. The DNA was desalted over a Sephadex.TM. G-50 column (Nick column, Pharmacia, Piscataway, N.J.) and concentrated using a Centricon.TM. P-30 (Amicon Division, Beverly, Mass.) to a final volume of 100 l. From 1.times.10.sup.9 cells, 105-115 g of DNA typically were obtained. Total cellular mRNA was isolated using the guanidine thiocyanate procedure set forth in the Promega technical information publication TB 087 12/89, Promega Corp. Madison, Wis. 53711. A pUC 18 plasmid was obtained from Dr. A. J. Sinskey, of the Department of Biology at the Massachusetts Institute of Technology. The library was constructed using the F. heparinum genomic DNA. The genomic DNA was sonicated and modified by adding EcoRI linkers and then ligated to the pUC 18 vector. DH5a was transformed with the pUC 18 genomic library.

Amplification of the PCR Product

Amplification of the heparinase tryptic digest primers was carried out in a 25 l reaction volume containing 50 mM KCl, 10 mM Tris HCl (pH 8.3), 1.5 mM MgCl.sub.2 and 0.01% gelatin plus the four deoxyribose nucleotide triphosphates(dNTPs) at 200M, using 0.5M primer and 3 l of the genomic DNA as the template, 2.5 units of the Taq polymerase (Cetus Corp., Emeryville, Calif.) and 25 l of mineral oil. The samples were placed on an automated heating block (DNA thermal cycler, Perkin Elmer Corp., Norwalk, Conn.) programmed for step cycles of temperatures 92.degree. C. (2 minutes), 50.degree. C. (1 minute) and 72.degree. C. (3 minutes). This cycle was repeated 35 times. The final cycle had a 72.degree. C. 10 minute extension. The PCR products were analysed on a 0.8% agarose gel containing 0.6 .mu.g/ml ethidium bromide. The control reaction was provided by the Cetus kit.

Screening of the Flavobacterium heparinum pUC 18 genomic library

The pUC 18 library was titered to give approximately 1500 colonies to be tested by the probe generated by PCR. Each plate had approximately 100 colonies which were grown directly on nitrocellulose, to an appropriate small size, and then duplicated to be grown further overnight.

The PCR probe was labelled using the Random Hexanucleotide.TM. kit (RHN) (IBI Biochemicals Ltd.) which is described briefly as follows. One microgram DNA from the PCR product run was isolated from a low melt agarose gel, denatured by boiling at 95.degree. C. for 10 minutes, and then chilled on ice. To the denatured DNA were added 10 mM dNTPs (dATP, dGTP, dCTP, dTTP), random hexanucleotides in the reaction buffer, and 50 .mu.Ci of .sup.32 PdCTP(3000 Ci/mmole). The reaction was carried with Klenow for 30 minutes at 37.degree. C. and terminated using 0.2M EDTA. Following the labelling reaction, the labelled probe was purified from the free nucleotide by using a Sephadex G-50 column (Nick Column, Pharmacia, Piscataway, N.J.)

The colonies were screened with the labelled probe using standard colony hybridization procedures as described by Maniatis et al., Molecular Cloning: A Laboratory Manual, Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y., incorporated herein by reference.

Two positive clones were isolated and the plasmids tested for their ability to generate the 600 basepair PCR product. Both of the clones tested positive and were further characterized by restriction mapping. Clone pRS Hep 51 is a 2.3 kb insert in pUC 18 (shown in FIG. 2) with a Kpn--Kpn fragment of about 1.6 kb. This fragment was a positive template for generating a 600 basepair PCR product. The KpnI--KpnI fragment of pRS 51 was subcloned into M13 and sequenced.

DNA Sequencing

DNA sequencing was performed using phage M13 and employing the dideoxyadenosine 5'-alpha-.sup.35 S-triphosphate and Sequenase (US Biochemical Corp, Cleveland, Ohio) as described by the manufacturer. The sequence data was obtained using successive nested deletions in M13 using T4 DNA polymerase as per Cyclone I Biosystems (International Biotechnologies Inc., New Haven, Conn.) or sequenced using synthetic oligonucleotide primers.

The sequence reveals a single, continuous open reading frame (ORF) of 1152 basepairs corresponding to 384 amino acids and a leader sequence of about 21 amino acids. The PCR product spans from 566 to 1216 bases from the start site and corresponds to about 57% of the total gene.

Initially six different tryptic peptides mapped into the ORF. Subsequently, five other peptides were sequenced for structural studies and all of them mapped into the ORF, for a total of about 48% of the total 367 amino acids. There are three cysteines in all, one associated with the signal peptide. The signal peptide is typical of prokaryotic sequences, having a charged N-terminal region, a core hydrophobic region and a cleavage region with a standard Ala.xxx.Ala site for cleavage.

Two different expression systems were selected for the expression of heparinase in E. coli: the Omp A expression system and the pKK hyper-expression system. The plasmid designs for both expression systems are shown in Table III.

Omp A expression system

The Omp A expression system secretes the protein of interest into the periplasmic space, as directed by the Omp A signal sequence, described by Ghrayeb, et al., EMBO J. 3:2437 (1984), incorporated herein by reference. This system was chosen since heparinase is naturally expressed into the periplasmic space of F. heparinum. The plasmid is under the control of the lac repressor and is induced by the addition of IPTG (isopropyl-.beta.-D thiogalactoside) to the medium. The plasmid was inserted in the pIN-III omp A-3 vector.

The heparinase insert was generated by PCR utilizing the N terminal and the C terminal sequences of heparinase with two appropriate restriction sites suitable for cloning into the EcoRI-BamHI sites. Two primers were constructed as shown in Table II. The insert was amplified by 5 cycles of PCR and ligated to the Omp A pIN vector with the E. coli periplasmic leader sequence. DH5.alpha. was transformed and expression was induced with 1 mM IPTG for 3-5 hours.

As shown in Table III, the construct of the Omp A expression system results in two extra amino acids at the amino terminal of the heparinase gene, Gly and Ile. The heparinase sequence begins with a Gln.

The pKK expression system

The pKK expression system is used for over-expression of proteins in accordance with the methods of Brosius and Holy, Proc. Natl. Acad. Sci., 81:6929 (1984) and Jaffe et al., Biochem. 27:1869 (1988), incorporated by reference herein. This system contains a strong tac promoter which, in appropriate hosts, is regulated by the lac repressor and induced by the addition of IPTG, as in the Omp A system. The plasmid pKK223-3 has a pUC 8 multiple cloning site and a strong rrnB ribosomal terminator immediately following the tac promotor. The ribosomal binding site of the plasmid was utilized by cloning the heparinase gene into a SmaI site, which is about 12 bases from the start codon ATG. Like the Omp A construction, the heparinase insert is obtained by PCR with SmaI and HindIII restriction sites at the N and the C terminals of the protein. As shown in Table III, the native heparinase leader sequence was used for over-production into the periplasm.

Periplasmic proteins of E. coli were isolated by osmotic shock. Briefly, 1.5 ml of cells were centrifuged after induction and washed with 10 mM Tris pH 7.5. The cells were then suspended in 20% sucrose in 10 mM Tris pH 7.5 and 5 Al of 0.5M EDTA. After a five minute incubation on ice, the cells were centrifuged and osmotically shocked by adding approximately 150 .mu.l water. The periplasmic extract was used to determine enzyme activity. Heparinase activity was determined by monitoring the wavelength at 232 nm and by the Azure A methods of Bernstein et al., Methods of Immunology 137:515 (1988), incorporated herein by reference.

The periplasmic extracts were analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) using the method of Laemmli, Nature 227:690 (1974) and stained using Coumassie blue. In addition, a Western blot assay was performed to confirm the presence of heparinase using a heparinase monoclonal antibody. Heparinase was electrophoretically transferred from the SDS-PAGE gel onto nitrocellulose using the method of Gershoni and Palade, Analytical Biochem. 131:1 (1983), and then incubated with the monoclonal antibody. This antibody was stained using a secondary antibody conjugated to horseradish peroxidase.

RNA Dot Blot Assay

The total cellular RNA was immobilized onto a Zeta probe.TM. membrane (Biorad, Richmond, Calif.) by alkaline RNA denaturation and fixation, and probed using the 600 base PCR product, used in screening for the heparinase gene. The hybridization was carried out with dot blot apparatus in accordance with the method of Thomas, Proc. Natl. Acad Sci. 77:5201 (1980). The RNA signal under different growth conditions has been investigated by Galliher, et al., Eur. J. Appl. Microbiol. (1982). It was established by those studies that heparinase at the protein level is optimally expressed under low sulphur conditions, which removes the requirement of heparin for induction. Heparinase mRNA signal under low sulphur growth conditions was therefore studied with and without heparin induction.

Both the OmpA and the pKK systems expressed heparinase. The OmpA system did not efficiently transport heparinase to the periplasm. For reasons not known, a large fraction of recombinant heparinase was retained in the cytoplasmic region along with the Omp A signal sequence. At lower temperatures (25.degree.-30.degree. C.) of growth, there was some secretion into the periplasmic space.

The pKK overproduction system produced heparinase only in the periplasmic space. The pKK system used the native F. heparinum heparinase leader sequence in which there was no problem with the transport of the recombinant protein with a foreign leader sequence. The pKK system expressed heparinase without any aberrant processing, although the expression was again optimal at lower temperatures. The presence of heparinase in the periplasm was confirmed by western blotting and by comparing in situ tryptic digest of the recombinant heparinase with that of the native heparinase, in terms of the peak profiles and some peaks which were isolated and sequenced.

A positive signal was obtained for the isolated F. heparinum mRNA using the 600 basepair probe derived from the PCR which has been used for isolating the heparinase gene, confirming that the gene isolated was a F. heparinum gene cloned in E. coli.

The expressed heparinase appeared to have at least some heparinase activity.

The sequence can be modified to alter specific enzymatic activity or binding specificity or affinity by substitution of one or more amino acids, using site directed mutagenesis or substitution of oligomers into the sequence encoding the heparinase. Methods and materials to accomplish this are known to those skilled in the art. The modified gene is then expressed and the product routinely screened for the altered activity.

Although described with reference to two specific expression systems, other expression systems are well known and commercially available. The heparinase gene can be expressed in these systems, using similar vectors and signal peptides or leader sequences.

Modifications and variations of the present invention will be obvious to those skilled in the art. Such modifications and variations are intended to come within the scope of the following claims.