AND METHOD FOR ITS EXPRESSION

BACKGROUND OF THE INVENTION

Field of the Invention This invention relates to synthetic genes and their expression products. Specifically, this invention relates to a synthetic protease gene and its expression product. Description of the Related Art The presence of protease protein in purified virion preparation was shown only by immunological techniques. The HIV protease sequence together with the gag and pol sequence or fusion proteins have been expressed from viral DNA in bacteria. Examples of such disclosures include: 1. Henderson, et al., 1988, "Human Retroviruseε, Cancer and AIDS: Approaches to Prevention and Therapy", D. Boloαnesi Ed. Published by Alan R. Liεs Inc., New York, NY. pp.135-147; 2. Debouck, et al., 1987, P.N.A.S.. 84 _. :8903-8906, and 3. Mous, et al., 1988, J. Virol, 62:1433-1436.

The primary sequences of the HIV protease has been determined by protein analysis and by the nucleotide sequence of the proviral DNA. It was thus determined that the protease is a 99 amino acid long protein encoded by a 297bp long stretch of the HIV provirus. All previous experiments on the protease gene and on its expression were carried out by utilizing nucleotide sequences cloned out from the cDNA of the provirus. The inventors' work using synthetic DNA proves that the nucleotide sequence of the proviruε DNA and also the deduced aminoacid sequence are correct.

The complete nucleotide sequence of the HIV-1

SUBSTITUTESHEET

proviral DNA was published by Ratner et al., 1985, Nature, 313:277-284. The sequence coding for the protease in the pol open reading frame of HIV was determined by previous analysis and corresponds to nucleotide 1609 to 1906 The N terminus and the C terminal a ino-acids are proline and phenylalanine respectively. This sequence coding for the HIV-I 99 a inoacid protease is 297bp long as follows.

10 20 30 40 50 CCTCAGATCA CTCTTTGGCA ACGACCCCTC GTCACAATAA AGATAGGGGG GGAGTCTAGT GAGAAACCGT TGCTGGGGAG CAGTGTTATT TCTATCCCCC

60 70 80 90 100

GCAACTAAAG GAAGCTCTAT TAGATACAGG AGCAGATGAT ACAGTATTAG CGTTGATTTC CTTCGAGATA ATCTATGTCC TCGTCTACTA TGTCATAATC 110 120 130 140 150

AAGAAATGAG TTTGCCAGGA AGATGGAAAC CAAAAATGAT AGGGGGAATT TTCTTTACTC AAACGGTCCT TCTACCTTTG GTTTTTACTA TCCCCCTTAA

160 170 180 190 200

GGAGGTTTTA TCAAAGTAAG ACAGTATGAT CAGATACTCA TAGAAATCTG CCTCCAAAAT AGTTTCATTC TGTCATACTA GTCTATGAGT ATCTTTAGAC 210 220 230 240 250

TGGACATAAA GCTATAGGTA CAGTATTAGT AGGACCTACA CCTGTCAACA ACCTGTATTT CGATATCCAT GTCATAATCA TCCTGGATGT GGACAGTTGT 260 270 280 290 TAATTGGAAG AAATCTGTTG ACTCAGATTG GTTGCACTTT AAATTTT ATTAACCTTC TTTAGACAAC TGAGTCTAAC CAACGTGAAA TTTAAAA

The industry is lacking a synthetic DNA sequence that encodes a specific enzyme or protease which is essential for the completion replication) of an infective human immunodeficiency virus (HIV)-. This DNA sequence is desirable to express this protease by recombinant methodology in bacteria and or in

TITUTESHEET

eukaryotic cells, and to produce enough protease for biochemical and physical characterization in order to design and produce potent inhibitors of this enzyme, and thereby to block the production of infective HIV particles.

BRIEF DESCRIPTION OF THE INVENTION

The invention is a gene for encoding a protease of human immunodeficiency virus. The gene consists essentially of a synthetic nucleotide sequence for a protease essential to infectivity of human immunodeficiency virus.

The protease is desirably a protease of HIV-1 or HIV-2 that is essential for the infectivity of these viruses. The preferred embodiment of this inventions is a synthetic gene and the coding sequence for expression of the HIV-1 protease is represented above by the top rows of nucleotide sequence. BRIEF DESCRIPTION OF THE DRAWING Figure 1 presents the expressed HIV protease as analyzed in Western blot.

Figure 2 illustrateε a strategy for the synthesis of the HIV-1 protease gene. The 3' overhangs are in lower case. The complementary strands (not shown) were provided with 3 ' overhangs to match the coding strands.

Figure 3 illustrates the induction of the gene at various periods of time.

Figure 4 illustrates the activity of the expressed protease using a synthetic peptide asa substrate. DESCRIPTION OF THE PREFERRED EMBODIMENT

The invention is a synthetic DNA sequence for

SUBSTITUTESHEET

encoding a specific enzyme or protease. The protease is essential for the infectivity of the human immunodeficiency virus (HIV) . The invented gene iε desirable for the expression of the protease by recombinant methodology in bacteria and or in eukaryotic cells and the production of a commercially desirable amount of the protease for biochemical and physical characterization. This characterization is necessary for the design and production of potent inhibitors of this enzyme. The invention also includes synthesis and expression of the protease gene of other retroviruseε εuch aε HIV-2, the human leukemia viruses such as HTLV I, II, and other human and animal RNA containing viruses causing leukemia sarcoma and other malignencies.

The nucleotide sequence for the preferred embodiment of this invention was obtained from a published paper by Ratner, et al., supra. The sequence in the pol open reading frame coding for the protease of HIV-1 corresponds to nucleotide 1609 to 1906. The N-terminal and the C-terminal amino-acids are proline and phenylalanine reεpectively. This sequence coding for the 99 aminoacid protease is 297bp long as shown above. Minor substitutions of one or more bases in this and other genes useful in this invention can produce a variant gene capable of expresεing the desired protease.

This sequence was synthesized as five fragments using the DNA synthesizer. Complementary strands corresponding to these five fragments were also synthesized. The 3' overhangs of four bases were provided for appropriate sequences to efficiently ligate each of the five fragments and to provide the correct coding sequence of the protease gene.

SUBSTITUTE SHEET

Nucleotide ATG were added to the fragment corresponding to the 5' end of the gene and TAA at the 3' end.

A procaryotic expression vector was used to clone and then to express the synthetic sequence coding for the protease. The expression can be in prokaryotes (bacteria) or in other appropriate expression systems. Recombinant clones screened by colony hybridization using a labelled fragment (62bp) spanning the internal region of the protease gene. Positive clones were further analyzed for the size of the insert. Clones which answered positive were induced for expression and analyzed in Western blots to determine the protein product using specific antibodies. Figure l gives an example. Of the clones screened so far, 3 clones have been identified to express a product of 11.5kd, reacting against specific antibodies as illustrated in Figure 1.

Conditions for the induction of a protease gene were studied in E. coli and optimized. The inventors have shown that the gene product has specific protease activity, as it is capable of cleaving both synthetic and natural substrates. The enzyme has been purified by specific column chro atographic techniques, including affinity chromotography. The method of this invention can produce enough active protease to study the structure of the protease, its mechaniεms of action, with a goal of devising specific inhibitors to this enzyme, of a therapeutic application for the treatment of the diseaεeε, such as AIDS, cauεed by the viruses. Other embodiments of this invention can utilize a gene to express another protease εuch as the following gene for the HIV-2 protease. CCTCAATTCTCTCTTTGGAAAAGACCAGTAGTCACAGCATACATTGAGGGTCAGCCA

SUBSTITUTESHEET

GTAGAAGTCTTGTTAGACACAGGGGCTGACGACTCAATAGTAGCAGGAATAGAGTTA GGGAACAATTATAGCCCAAAAATAGTAGGGGGAATAGGGGGATTCATAAATACCAAG GAATATAAAAATGTAGAAATAGAAGTTCTAAATAAAAAGGTACGGGCCACCATAATG ACAGGCGACACCCAATCAACATTTTTGGCAGAAATATTCTGACAGCCTTAGGCATGT CATTAAATCTAC

Figure 1 demonstrates the expression of the HIV protease in E. coli. Cells transformed with the synthetic sequence of HIV protease in an appropriate expression vector were induced and the bacterial lysate was electrophoresed in SDS-PAGE. After transfer of proteins into a nitrocellulose membrane, immunoblotting procedure was performed using the specific antibody to the HIV protease. Detection of Ag-Ab complex was made using I ¹²⁵ protein A. The autoradiograph lane A represents E. coli transformed with the plasmid, and laneε B and C E. coli transformed with the plasmid bearing synthetic DNA encoding the HIV protease. On the left are protein molecular weight markers in kilodalton. The 11.5 kd band is the protease. The synthetic DNA of the invention also obviates any need to manipulate (infectious) viral material and overcomes limitations in the quantities which can be obtained by other means. EXAMPLES The following materials and methods were used to perform the examples. PLASMID. BACTERIAL STRAINS. AND CHEMICALS:

Plasmid PKK233-2, a procaryotic expression vector was purchased from Pharmacia. PKK233-2 was used to transform in a laq-q host, E. coli cell JM105" or RB791. The cells were selected in M9 minimal media containing lug/ml thiamine, prior to using them for transformation. All chemicals utilized in the synthesis of oligonucleotides were from Applied

SUBSTITUTESHEET

Bioεyste s Inc. T4 polynucleotide kinase, DNA ligase, and Klenow fragment of E. coli DNA polymerase I were obtained from New England Biolabs. Reεtriction endonucleases, PMSF and IPTG were from Boehringer Mannheim, Bethesda Research Laboratories and Promega respectively. DNA SYNTHESIS. PLASMID CONSTRUCTION AND SCREENING:

DNA fragments were synthesized using a ABI DNA synthesizer (model 381A) . All synthetic fragments were purified by electrophoreεiε in a 12% polyacrylamide/8M urea sequencing gel. DNA was visualized by UV-shadowing and full-length fragments were eluted from the gel as known in the art. The full-length fragments were checked for their purity using standard techniques.

Appropriate complementary fragments were mixed in equimolar concentrations, annealed, kinased and ligated as described elsewhere. The efficiency of ligation was monitored by polyacrylamide gel lectrophoresis. The linearized plasmid and the protease gene in appropriate concentrations were ligated and used for transformation of E. coli. JM105. Recombinant clones were screened by colony hybridization using a 62 bp fragment labelled by kinasing. Small εcale iεolation of plaεmid DNA from the recombinant clones was performed by the boiling method and the size of the insertε waε viεualized by autoradiography after labelling the 3 • recessed terminal using the Klenow fragment of E. coli DNA polymerase.

ANTIBODIES TO THE HIV PROTEASE

The polyclonal antibodies were raised in rabbits against (i) a complete synthetic sequence of 1 to 99 aminoacids of the HIV-1 proteaεe and (ii) a

SUBSTITUTESHEET

tridecapeptide corresponding to the C-terminus of the protease.

ANALYSIS OF THE EXPRESSED PROTEINS

E. coli cells bearing the appropriate plaε id construct were grown to log phase, induced, and lysed by sonication. Total cell extracts were analysed by NaDodS0 ₄ /PAGE and subjected to immunoblot analysis. ASSAY FOR THE ACTIVITY OF THE EXPRESSED PROTEASE:

Oligopeptides were synthesized in a Peptide Synthesizer (Applied Biosystemε Model 43OA) , according to the method previouεly published (Copeland and Oroszlan, 1981) . The cleavage products were analysed by RP-HPLC on a uBondapak ^c 18 column (Waters Associateε) . Peak fractionε were analysed for amino- acid composition using a Pico-Tag a ino acid analyser (Waters Associates) . EXAMPLE 1

This example represents the preferred embodiment. RESULTS:

SYNTHESIS OF THE FULL-LENGTH PROTEASE GENE:

The nucleotide sequence of the protease gene was taken from Ratner et al. The sequence in the pol open reading frame for the protease gene startε at nucleotide 1609 and endε at 1906, for coding 99 aminoacidε. This sequence and its complement were synthesized as five individual fragments of approximately 60 bases as shown in Figure 2. The 3• overhangs of 4 bases (shown in lower case) were provided for the fragments to selectively ligate the appropriate fragments to form the correct coding sequence. Translational initiation codon ATG and termination codon TAA were provided at the appropriate endε of the protease gene. A sequence was added to

UBSTITUTE SHEET

provide a protruεion at the 5 ¹ end of the gene, having a coheεive end compatible to the reεtriction enzyme εite Ncol. The 5' protruεtion at the 3' end of the gene waε added to provide a Hind3 compatible end. The complementary strands (not shown) were provided with 3 ' overhangs to match the coding strandε.

EXPRESSION OF THE SYNTHETIC HIV-1 PROTEASE GENE IN E. COLI

Three clones (PR-C, PR-H, and PR-J) bearing the correct coding sequence of 297bp in the expression vector PKK233-2 were analyzed for expression to select conditions for the optimal induction of the gene. Figure 3 shows examples of Western blot analysis of the gene product. Figure 3 illustrates expresεion of the synthetic protease gene in E. coli. Clone PR-C bearing the coding sequence to the protease was induced for expression. The proteins (75ug of bacterial extract) were electrophoresed in a NaDodS0 ₄ /PAGE tranεferred to nitrocelluloεe and subjected to immunoblot analysiε uεing a mixture of the two protease specific rabbit polyclonal antibodies raised against (i) a complete synthetic sequence of 1-99 amino acids of the HIV-1 protease and (ii) a tridecapeptide corresponding to the C terminuε of the proteaεe. Figure 3 shows the induction of the gene with 0.4mM IPTG at various periods of time. Figure 3B shows the induction for 30 minutes. With increaεing concentrationε of inducer IPTG. 1-5 repreεent mM concentration of IPTG at 0.28, 0.56, 1.12, 2.24 and 4.48 reεpectively. Figure :3C shows the analysis after 60 minutes of induction with ImM IPTG and lysing the cells in various buffers. Bl denotes lysis of cells in 50mM Tris-HCl at pH 7.0, 150mM NaCl, ImM EDTA, ImM PMST, ImM DTT and 0.5 percent

SUBSTITUTE SHEET

NP-40. B2 iε the same as Bl, but without NaCl and EDTA. B3 is in 50mM potassium phosphate at pH 6.0, ImM PMSF and ImM DTT. B4 is the same as B3 with a pH of 6.5. Poεitionε of protein molecular weight markerε are inducated on the left in kilodaltonε.

E. coli cells bearing plasmid PR-C were grown in Luria broth to an optical density of 0.4 A600nm, and then induced at various periods of time for expression from the trc pro otor by adding IPTG (isopropyl-beta-D- thiog-alactopyranoside) at a concentration of 0.4mM as seen in Figure 3A. The cloned gene expressed a single, unfuεed protein band of 11.5kd. Expreεεion was maximal after 30 minutes of induction. This level decreased to about 25 percent at 60 minutes. There waε no detectable expreεεion after 120 minuteε of induction and at 0 minuteε. This pattern of induction was similar in the other clones (PR-H and PR-J) that were analyzed (not shown) .

The results of the induction for 30 minutes with varying concentrations of inducer are shown in

Figure 3B. Induction with IPTG in the range of ImM to 4mM resulted in maximum amount of expression. Similar data were obtained on clones PR-H and PR-J (not shown) . In order to select the conditions that efficiently solubilize the protease for enzymatic analysis, different buffer systemε were uεed for the lysis of cells (clone PR-C) after optimal induction with ImM IPTG. It was obεerved that sonication in a buffer system of 50mM Tris-cl at pH 7.5, ImM DTT, ImM PMSF and 0.5% nonidet P-40 released 50 to 70 percent of the protease in the soluble fraction (Figure 3C) . This was estimated by Western blot analysiε aliquόtε of soluble extract and inεoluble pellet for the content of the expressed product.

DEMONSTRATION OF SPECIFIC PROTEOLYTIC ACTIVITY

Figure 4 illustrates the activity of the expreεεed proteaεe uεing a synthetic peptide as a substrate. Protease assays were carried out with 22.5ug of bacterial lysate at 37°C obtained from clone PR-C, induced (A,B,C), uninduced (D) , and control cells bearing just the plasmid PKK233-2 (data not shown) . The nonapeptide was used as a substrate in reaction buffer (0.25 M potassium phosphate), pH 7.0, 0.5 percent (v/v) NP 40, 5 percent (v/v) glycerol, 5 inM

Dithiotreit and 2 M NaCl. Aliquots of 25 ul each were taken at 0 hours (A) , 1 hour (B) 3 hours (C) and 6 hours (D) analyzed by RP-HPLC. S denotes the substrate and PI and P2, cleavage products 1 and 2 respectively. To asεeεs the activity of the cloned HIV-1

Protease a synthetic nonapeptide corresponding to the HIV-1 pl7-p24 cleavage εite (Henderson, et al. 1988) was used as a substrate (4E) . The substrate in reaction buffer was mixed with aliquots of various cell extracts (see description of Figure 4 above) and incubated at 37°C. Equal eliquots of incubation mixture were taken at various time points and analyzed by RP- HPLC. The substrate in the 0 hour sample eluted as a single peak as shown in Figure 4A. After incubation for 1 hour, two newly appearing peaks, products labelled PI and P2, can be seen, correlating with a significant decrease of the substrate peak. Subsequent amino acid analysis of the recovered peaks demonstrated that product 1 and product 2 corresponded to the expected cleavage products as shown in Table 1 proving a Tyr-Pro bond cleavage, which is the determined natural cleavage site. Extended incubation for 3 hours showed a further decrease of the substrate peak and substantial increase in the peak height of product 1,

indicating progresεion of the hydrolyεis of the Tyr- Pro bond. However, the peak of product 1 seems to be smaller as expected since the absorbance of the tetrapeptide Pro-Ile-Val-Glu-NH ₂ iε substantially smaller than that of the pentapeptide having a free

COOH-terminal tyrosine. An increase of product 1 and 2 after 3 hours of incubation showed a corresponding decrease of the substrate peak.

No cleavage products have been detected in reactions uεing extractε from uninduced cellε, clone PR-C (Figure 4D) and of control cellε (control plasmid PKK233-2; data not shown). There was no decrease in the substrate peak even after 6 hours of incubation (Figure 4D) indicating that the nonapeptide iε resiεtent to degradation by bacterial proteases. This makes this substrate especially useful for assaying viral protease activities in crude extracts, facilitating purification and isolation of the proteaεe. The amino acid compoεition data for the substrate and its cleavage products are shown in Table 1. The amountε of observed amino acids correspond clearly to the expected amounts demonstrating that the cleavage occurs at the expected cleavage εite of the εynthetic peptide correεponding to the pl7-p24 εite of the gag precurεor.

Table 1. Amino add composition of the substrate and the cleavage products

:

I π

*The observed amounts of Val and lie were found lower than expected in product 1 due to a frequently observ Inefficient hydrolysis of the Ile-Val bond.