CA2300383A1 - Large scale production of human or animal proteins using plant bioreactors - Google Patents

Abstract

The present invention relates to an expression vector for the large scale production of a human or animal protein, which comprises a DNA construct consisting of operatively linked DNA coding for a plant promoter, a transcription terminator and the human or animal protein to be expressed. Such human or animal proteins may be selected from the group consisting of human protein C (HPC), factor VIII, growth hormone, erythropoietin, interleukin 1 to 7, colony stimulating factors, relaxins, polypeptide hormones, cytokines, growth factors and coagulation factors. The present invention also relates to the plant bioreactor and to the method for the large scale production of human or animal proteins.

Description

LARGE SCALE PRODUCTION OF HUMAN OR ANIMAL PROTEINS
US ING PUTT BIORBACTORs (a) Field of the Invention The invention relates to a vector for large scale production of human or animal proteins using plant cells, a method of large scale production of human or animal proteins using plant cells, to a method of large scale production of human protein C using plant cells and to plant bioreactors for the expression of human or animal proteins using plant cells.
tb) Description of Prior Art Several foreign proteins have been expressed in plants for diverse purposes: pest resistance, viral and fungal resistance, environmental stress tolerance, herbicide resistance or tolerance, food quality and processing, experimental studies, and for the expres sion of specialty chemicals.
Among the proteins which were expressed in plant cells, there is the following:
A human neuropeptide which was linked to a frag-ment of the plant 2S albumin gene. The peptide accumulated in the seeds of Arabidopsis and of oilseed rape at levels up to 200 nmol and 50 nmol respectively per gram of seeds;
~ Chicken ovalbumin in alfalfa with levels of ovalbumin up to 0.01 of total soluble proteins:
~ The prepro-human serum albumin gene was trans fected in potatoes. The prosequence was not cleaved before secretion from the plant cells but the human signal sequence was recognized by the plant endoplasmic reticulum; and ~ Antibodies were produced in tobacco plants.
Using plant transgenics for the mass production of foreign proteins has several advantages, such as the low cost of growing the plants, the possibility of growing the transgenic plants on a very large scale, the transformation procedures which are well estab-lished, and the possibility of using specific plant parts as a sink for engineered proteins. However, in the case of human or animal proteins requiring post-translational modifications, the lack of knowledge concerning the post-translational modifications in plants represents a potential problem.
Agrobacterium-mediated gene transfer is the most widely used technique for plant transformation. Agro-bacterium tumefaciens is a saprophytic soil bacterium which is also a pathogen of many dicotyledonous plants causing the formation of crown galls. Pathogenic Agro-bacterium has a megaplasmid of approximately 200 kilo base pairs called the Ti plasmid or tumor-inducing plasmid. The T-DNA (or transferred DNA) within the megaplasmid is delimited by two 25 base-pair (bp) sequences called the right and left borders. Virulence genes are located outside the T-DNA. These contain the information for the excision of the T-DNA at the T-DNA
borders, and its transfer to the plant cells.
Agrobacterium-mediated gene transfer takes advantage of this system to transfer foreign DNA.
The binary vector strategy exploited here uses E. coli-Agrobacterium shuttle vector. This vector con-tains the T-DNA borders flanking the foreign DNA. This vector is introduced into Agrobacterium which moves the T-DNA in trans to the plant cells. The binary vector strategy uses disarmed Ti plasmids.
There is a demand for many human or animal pro-teins which have therapeutical applications. These proteins are sometimes difficult to produce in large quantities.

Since the use of human protein C (HPC) concen-trate as a therapy appears promising (Dreyfus, M. et al., 1991, N. Engl. J. Med., 325:1565-1568), several isolation and production systems have been studied.
The purification of HPC from human plasma con-stitutes a challenge since HPC is present in the plasma at concentrations of approximately 4 ~g/ml and con-taminants from similar vitamin K-dependent plasma pro-teins may be difficult to remove. In addition, there is always the possibility of infectious agent contami-nation. Nevertheless, Velander et al. (1991, In pro-tein C and related anticoagulants, Bruley, D.F. and W.N. Drohan (eds), Portfolio, The Woodlands, Texas, p.ll-27) designed a protocol to purify HPC from plasma.
The starting material is either cryopoor plasma or reconstituted Cohn IV-1 paste which is filtered and adsorbed on an anion-exchange chromatography column.
The eluate containing HPC is treated with solvent and detergent to inactivate viruses and it is adsorbed on a protein C immunoaffinity column. The eluate is again adsorbed on anion-exchange chromatography and HPC
finally undergoes diafiltration before formulation.
The amount of purified HPC is small and may not be use-ful for industrial applications.
Synthesis of biologically active recombinant protein C by bacteria or yeast is precluded because those organisms are unable to perform some of the critical post-translational modifications. Production of vitamin K-dependent plasma proteins by most mammalian cells resulted in partially processed pro-teins and low transcription levels (Anson et al., 1985, Nature, 315:683-685; Busby et al., 1985, Nature, 316:271-273; Grinnell et al., 1987, BiolTechno.Iogy, 5:1189-1192; de la Salle et al., 1985, Nature, 316:268-270). However, improved cell lines have been described *rB

r" _ transgenic animals. Velander et al. (1991, Ann. N.Y.
Acad. Sci., 665:391-403) demonstrated that encineered mice could produce biologically active HPC in their milk at concentrations of up to 3 ~cg/ml. Velander et al. (1992, Proc. Natl. Acad. Sci. USA, 89:12003-12007) , also reported that transgenic swine were capable of high-level expression of HPC. A concentration of 1 g per litre of milk was detected from the best animal.
The use of animals for the production of human protein t U ~. i 5 i i W . a c o i 1 ci it i c a a c ~. v ~. i W :: ~:. ~ y ~ W . v.:
~ ~. s. ,., ,,:.; ,.; .;
with the purification of the human transgenic protein away from related animal proteins.
International Application published under No. WO
97/04122 on February 6, 1997 discloses an expression o vector comprising DNA coding for a plant promoter, a transcription terminator and a protein to be expressed, such as hormones and growth factors.
International Application-published under No. WO
91/02066 on February 21, 1991 discloses a DNA construct zo for the expression of human serum albumin (HSA) in plants.
It would be highly desirable to be provided with a vector for the large scale production of human or animal proteins using plant cells.
zs It would be highly desirable to be provided with L.; ..~-~~r.t-~,r ~r"-~ tho l ~,-r_ro crV~~l c r,rnr7osnt; nn of lpoman nr ...._..__~_,.~_ _,.~ ___ __- __ ,.._ __ animal proteins using plant cells.
It would be highly desirable to be provided with a method of large scale production of human or animal 3o proteins using plant cells.
It would be highly desirable to be provided with a method of large scale production of human protein C
using plant cells.
~i,r~rJDED sHEET

r. _ »a . . , -~ ,.
r SUMMARY OF THE INVENTION
One aim of the present invention is to provide a vector for the large scale production of human or ani s mal proteins using plant cells.
Another aim of the present invention is to pro-vide a method of large scale production of human or animal proteins using plant cells.
Another aim of the present invention is to pro-1U VlliC a 1J1V1tCGt.tV1 ivW .iic iui~c .~s.~..y:. ~i:.:~.:: .......~
human or animal proteins using plant cells.
F,ivi~fVGED SHEET .

Another aim of the present invention is to pro-vide a bioreactor for the large scale production of human or animal proteins using plant cells.
Another aim of the present invention is to pro s vide a method of large scale production of human pro tein C using plant cells.
In accordance with one embodiment of the present invention, there is provided an expression vector for the large scale production of a human or animal protein, which comprises a DNA construct consisting of operatively linked DNA coding for a plant promoter, a transcription terminator and said human or animal pro-tein to be expressed. More specifically, the expres-sion vector is referred to as pCP2.
In accordance with another embodiment of the present invention, there is provided an expression vec-tor for the large scale production of a human or animal protein, which comprises a DNA construct consisting of operatively linked DNA coding for a plant promoter, a mRNA stabilizer, a transcription terminator, and said human or animal protein to be expressed. More specifi-cally, the expression vector is referred to as pLG3.
In accordance with the present invention there is provided a plant bioreactor for the large scale pro duction of a human or animal protein, which comprises dicotyledonous plants transformed with a DNA construct consisting of operatively linked DNA coding for a plant promoter, a transcription terminator and said human or animal protein to be expressed flanked by T-DNA borders and a suitable selectable marker for plant transforma-tion.
In accordance with the present invention there is provided a plant bioreactor for the large scale pro-duction of a human or animal protein, which comprises dicotyledonous plants transformed with a DNA construct consisting of operatively linked DNA coding for a plant promoter, a mRNA stabilizer, a transcription terminator and said human or animal protein to be expressed flanked by T-DNA borders and a suitable selectable marker for plant transformation.
In accordance with the present invention there is provided a method of large scale production of human or animal proteins, which comprises the steps of:
a) inserting a suitable recombinant expression vec for in plant cells using Agrobacterium transfor mation, said expression vector comprising operatively linked DNA coding for a plant pro moter, a transcription terminator and said human or animal protein to be expressed flanked by T
DNA borders and a suitable selectable marker for plant transformation; and b) recovering said expressed human or animal pro-tein of step a) from said culture medium.
In accordance with the present invention there is provided a method of large scale production of human or animal proteins, which comprises the steps of:
a) inserting a suitable recombinant expression vec-for in plant cells using Agrobacterium transfor-mation, said expression vector comprising operatively linked DNA coding for a plant pro moter, a mRNA stabilizer, a transcription termi nator and said human or animal protein to be expressed flanked by T-DNA borders and a suit able selectable marker for plant transformation;
and b) recovering said expressed human or animal pro-tein of step a) from said culture medium.
In accordance with the present invention there is provided a method of large scale production of human protein C, which comprises the steps of:

-a) inserting a suitable recombinant expression vec-tor in plant cells using Agrobacterium transfor-mation, said expression vector comprising operatively linked DNA coding for a plant pro-s moter, a transcription terminator and said human protein C to be expressed flanked by T-DNA bor-ders and a suitable selectable marker for plant transformation; and b) recovering said expressed human protein C of step a) from said culture medium.
In accordance with the present invention there is provided a method of large scale production of human protein C, which comprises the steps of:
a) inserting a suitable recombinant expression vec for in plant cells using Agrobacterium transfor mation, said expression vector comprising operatively linked DNA coding for a plant pro moter, a mRNA stabilizer, a transcription termi nator, and said human protein C to be expressed flanked by T-DNA borders and a suitable select-able marker for plant transformation; and b) recovering said expressed human protein C of step a) from said culture medium.
For the purpose of the present invention the following terms are defined below.
The term "human or animal proteins" is intended to mean any human or animal protein which include with-out limitation, human protein C (HPC), factor VIII, growth hormone, erythropoietin, interleukin 1 to 7, colony stimulating factors, relaxins, polypeptide hor-mones, cytokines, growth factors and coagulation fac-tors.
The term "plants" is intended to mean any dicotyledonous plants, which include without limitation tobacco, tomato, potato, crucifers.

WO 99!09187 PCT/CA97/00590 - g -The term "operatively linked" is intended to mean that the elements are physically joined on the same piece of DNA to produce a unit with a specific purpose.
The term "T-DNA borders" is intended to mean the 25 base pair Agrobacterium-derived sequences that delimit the fragment of DNA that will be transferred to the plant cell with the help of Agrobacterium proteins.
The expression "a suitable selectable marker for plant transformation" is intended to mean any gene cod ing for a function that will allow the identification of transformed plant cells, such as kanamycin resis tance.
BRILF DESCRIPTION OF THE DRAWINGS
Fig. 1 illustrates the construction of plasmid pCP2;
Fig. 2 illustrates the construction of plasmid pLG3;
Fig. 3 is a graph of first screen for HPC pro-duction of samples 1 to 27;
Fig. 4 is a graph of first screen for HPC pro-duction of samples 28 to 54;
Fig. 5 is a graph of first screen for HPC pro-duction of samples 55 to 81;
Fig. 6 is a graph of first screen for HPC pro-duction of samples 82 to 104;
Fig. 7 illustrates a Western immunoblot of reduced samples from ion-exchange #7 using a rabbit anti-HPC serum.
DETAILED DESCRIPTION OF '~1HE INVENTION
The present invention was designed to express human or animal proteins, namely human protein C (HPC), in plants. This novel approach has several advantages g over mammalian transgenics: the low cost of growing the plants, the ability to produce the protein on a very large scale, and the elimination of contamination by related animal proteins during purification of the expressed human or animal protein.
The preferred bioreactors in accordance with the present invention include most dicotyledonous plants, particularly the solonacae, which include (but are not limited to) tobacco, potato and tomato, and the cruci-fern as transformed with the vector of the present invention.
The preferred plants in accordance with the pre-sent invention include without limitation tobacco, tomato, potato and crucifers.
Many other human or animal proteins, beside human protein C, may be prepared in accordance with the present invention. The preferred human or animal pro-teins to be expressed in accordance with the present invention include, but are not limited to, anti-coagu-lation proteins, such as human protein C, factor VIII, or tissue plasminogen activator, and other proteins of pharmaceutical or veterinary interest.
Human protein C (HPC) is a vitamin K-dependent plasma glycoprotein which is a key element of the anti coagulation cascade. It is synthesized by the liver cells as a single peptide but is modified into a het-erodimer linked by a disulfide bond before its secre-tion to the bloodstream. Some individuals are par-tially or completely HPC deficient, a situation that increases the likelihood of an early thrombotic event which may be lethal. Purified HPC injection has been used as an experimental treatment for homozygous defi-cient patients who are not producing HPC, but is also a promising drug for several other complications such as septic shock, thrombolytic therapy, and hip replace-ment. The annual demand in the U.S.A. for HPC repre-sents about 96 kg. At the moment, HPC is purified from human plasma. Several researchers are experimenting with the synthesis of HPC in the milk of transgenic animals. In accordance with the present invention, the production of HPC is intended to serve as an example of the human or animal proteins which can be produced at a very large scale using plants as bioreactors.
The method of the present invention involved engineering tobacco plants using Agrobacterium-mediated gene transfer associated with the binary vector strategy of Hoekema et a1. (1983, Nature, 303:179-180).
Agrobacterium-mediated gene transfer takes advantage of the gene transfer system provided by the bacterium.
The binary vector strategy consists of using A. tvmefa-ciens with a Ti plasmid, inserting an accessory plasmid called the binary vector into the bacterium, and allow-ing T-DNA transfer. The accessory plasmid contains T-DNA border sequences with the desired genes and regu-lating elements located between them.
Engineering tobacco plants to produce human or animal proteins, or for example HPC, was achieved via the use of either of the two plasmids with different elements for the regulation of the expression of the introduced cDNA. The first plasmid included the con-stitutive caulif lower mosaic virus ( CaMV ) 3 5S promoter to drive gene expression, the second included a dimer of the 35S constitutive promoter with an alfalfa mosaic virus (AMV) leader sequence to enhance stability of the transcript. Duplicating the promoter has previously been found to enhance transcription, while the leader sequence enhanced translation.
Expression of the T-DNA was verified by ana lyzing HPC and neomycin phosphotransferase II (NPTII) synthesis by enzyme linked immunosorbent assay (ELISA) and inheritance of the T-DNA insert was observed by germinating R1 seeds on antibiotic-containing medium or by ELISA to HPC on Rl seedlings.
Purification protocols were created and pre y liminary experiments are described. Biological activ ity of various protein fractions was measured.
Tobacco plants engineered with the human protein C (HPC) cDNA and plant promoters expressed HPC. This was demonstrated by ELISA and Western assays using a combination of antibodies to human protein C. No similar protein was found in non-transformed plants.
The protein had the expected molecular weight of the uncleaved form of HPC.
Changes in coagulation times were observed in several experiments when tobacco extracts were tested for clotting activity.
Plant transformation and selection Several cell types or tissues can be used but cells must be totipotent, that is, able to regenerate mature plants. Plant cells or tissues are co-culti vated with Agrobacterium for a few days to allow T-DNA
transfer. After co-cultivation, plant cells and tis sues are grown on media with antibiotic which sup presses bacterial growth. Engineered plant cells sur-vive because an antibiotic resistance marker gene is transferred with the foreign DNA. This system allows elimination of non-transformed plant cells. Plantlets are regenerated for analysis using plant tissue culture media with various plant growth regulator levels.
HPC structure and post-translational modifications HPC is a complex vitamin K-dependent plasma gly coprotein. The HPC mRNA codes for a single peptide including a signal peptide and a propeptide sequence.
After cleavage of the single chain by removal of the KR
dipeptide (Lys-Arg), HPC is secreted to the bloodstream as a two-chain glycoprotein with a molecular weight of 62,000 Da. The light chain (21,000 Da) and the heavy chain (41,000 Da) remain attached by a disulfide bond.
However, before secretion, HPC must undergo several post-translational modifications.
Determination of SPC's biological activity A cDNA clone coding for HPC was inserted down stream of the CaMV 35S promoter and of a dimer of the CaMV 35S promoter. Tobacco plants were transformed using Agrobacterium and a binary vector strategy.
Kanamycin resistant plants were regenerated. T-DNA
integration was tested to insure that plants were stably transformed. Rl seedlings were also analyzed.
A second round of transformation was performed in order to increase the level of HPC expression. Partial pro-tein purification (using ion-exchange chromatography), dialysis and ultrafiltration were followed by various analyses (SDS-PAGE, Western immunoblot) in order to assess protein purity and activity. Clotting assays were performed in order to determine whether the plant-produced HPC was biologically active.
1. Recombinant D~tA manipulations The binary vector pBI121 and a culture of Agro-bacterium tumefaciens strain LBA4404 were purchased from Clontech. The plasmid pBI524 and pLPC were pro-vided by Dr. Bill Crosby from Agriculture Canada and by Dr. Jeff Turner (Department of Animal Science, McGill University). Restriction and modifying enzymes were purchased from New England Biolabs and the DNA marker (IKB DNA ladder) from BRL. Standard recombinant DNA
manipulations were used during the construction of the binary vectors and all plasmid manipulations were performed using E. coli strain DHSa if not specified.
Finally, the concentration of agarose for gel electrophoresis was 0.8~ unless mentioned, and gels were run in 0.7X TBE buffer (89mM Tris, 89mM boric acid, 2mM EDTA, pH 8.3). Specific DNA fragments to be recovered after enzymatic digestion were electropho-resed in TAE buffer (40mM Tris-acetate, 2mM Na2EDTA, pH
8.5) and the GenecleanT"t II kit (Bio 101) was used to purify and extract the DNA from the agarose gel.
1.1 Non-radioactive hybridisation Plasmid DNA was digested with Xbal and EcoRI, electrophoresed in l~ agarose gels (TBE), and trans ferred onto HybondTM N+ nylon membrane (Amersham) by alkaline transfer with 0.4 M NaOH according to the mem brane manufacturer's instructions. The Bcll fragment containing the HPC cDNA from pLPC was digested, elec trophoresed in an agarose gel in TAE, and isolated by GenecleanTM II. The Bcll fragment was used as the probe. Probe labeling, hybridization, and detection was carried out using non-radioactive DIG'S"" DNA Labeling and Detection Kit (Boehringer Mannheim) according to the manufacturer's instructions.
1.2 Genomic DNA isolation Plant genomic DNA from transgenic tobacco was isolated using the CTAB (hexadecyltrimethylammonium bromide) method. Five grams of leaf material (previ ously frozen at -70°C) were homogenized in 15 ml of prewarmed CTAB buffer (100 mM Tris-Cl pH 8.0, 1.4 M
NaCl, 20 mM EDTA, 0.2$ 13-mercaptoethanol, 2$ CTAB) in a WaringT"' blender and incubated for 30 minutes at 60°C.
The solution was mixed every 5 minutes. An equal vol-ume of chloroform-isoamyl alcohol (24:1) was added to the tube, mixed well and centrifuged 3000 g for 15 minutes at 4°C. The aqueous phase was collected and chloroform-extracted once more. The aqueous phase was transferred to a new tube and precooled (-20°C) isopro-panol was added 2/3 volume of the aqueous solution.
The mixture was inverted a few times and incubated at -20°C for at least 60 minutes. DNA was precipitated by centrifugation 10,000 g for 10 minutes at 4°C. Super-natant was removed and the DNA pellet washed with 10 mM
ammonium acetate in 76$ ethanol. DNA was vacuum-dried and resuspended in TE buffer (10 mM Tris-C1 and 1 mM
EDTA, pH 8.0).
1.3 Southern hybridization Genomic DNA (20 fig) was digested with Sau3AI and DNA fragments were separated by agarose gel electro phoresis (l~ agarose) in TBE buffer. DNA was trans ferred to HybondT'" N+ nylon membrane (Amersham) using alkaline transfer following the manufacturer's instruc tions. The DNA probe was the cDNA of HPC (Bc~I frag ment of pLPC) labeled with oc32P-dCTP (ICN) using the T7 QuickprimeT"~ kit of Pharmacia. The nylon membrane was incubated in 30 ml of prehybridization buffer (250 mM
NaHP04 pH 7.2, 2.5 mM EDTA, 7~ sodium dodecyl sulfate (SDS), 1$ blocking reagent (Boehringer Mannheim), 50~
deionized formamide) for 24 hours at 42°C. The prehy-bridization buffer was replaced by 15 ml of hybridiza-tion buffer (prehybridization buffer with 10~ dextran sulfate) and incubated with the probe and the membrane overnight at 42°C. The membrane was washed three times 15 minutes at 42°C with 2X SSC (20X SSC was 3 M NaCl and 0.3 M Na3citrate, pH 7.0) and 0.1$ SDS, followed by 0.5X SSC and 0.1$ SDS, and then O.1X SSC and 0.1~ SDS.
The final wash was with O.1X SSC and 0.1~ SDS for 30 minutes at 52°C. The X-ray film (KodakTM X-OMAT AR) was exposed for approximately 7 days.
1.4 Transfer of plasmids into Agrobacterium Three plasmids (pBI121, pCP2, and pLG3) were purified from E. coli using an alkaline lysis DNA
minipreparation. Vector DNA was transferred to A.
tumefaciens strain LBA4404 using a freeze and thaw method.

2. Plant transformation 2.1 Plant material propagation Seeds of Nicotiana tabacum cv. Xanthi were ster ilized for 15 minutes in a 10~ bleach solution with a drop of detergent (TweenT~-20). Seeds were washed at least five times with sterile distilled water and allowed to germinate and grow on artificial medium composed of basal MS salts, B5 vitamins, 3~ sucrose, and 0.6~ agar (Anachemia) at pH 5.7-5.8. Seed were grown under a 16 hour photoperiod with a light inten-sity of 50 ~E and a temperature of 24°C.
2.2 Preparation of Agrobacterium inoculum A. tumefaciens was grown in Luria Broth (LB) (l~
tryptone, 0.5~ yeast extract, 85 mM NaCl, pH 7.0) medium supplemented with 50 ~g/ml kanamycin and ~g/ml of streptomycin for 18 hours or until the optical density at 595 nm reached 0.5 to 1Ø Cells 20 were spun down at 3,000 g for 5 minutes and the pellet was resuspended to its initial volume with MS-104 medium (MS basal salts, B5 vitamins, 3$ sucrose, 1.0 ~g/ml benzylaminopurine (BAP), 0.1 ~g/ml naph taleneacetic acid (NAA), pH 5.7-5.8, and 0.8$ agar) 25 without agar.
2.3 Leaf disc transformation Leaf squares of about 64 mm2 were dissected us ing a sharp scalpel, immersed in the inoculum for 15-30 minutes and plated onto MS-104 medium for 2 days under a 16 hour photoperiod, under low light intensity (20 ~E) at 24°C for cocultivation. Leaf discs were washed alternately three times with sterile distilled water for 1 minute and sterile distilled water supplemented with 500 ~g/ml of carbenicillin for 5 minutes. Leaf discs were transferred to MS-104 medium with 500 ug/ml of carbenicillin for another 2 days under the same environmental conditions.
2.4 Selection and regeneration Leaf discs were washed as above, plated on MS-104 medium with 500 ~tg/ml of carbenicillin and 100 ~g/ml of kanamycin, and grown using the above envi-ronmental conditions until calluses appeared. The light intensity was increased to 50 ~E and the explants were allowed to grow until development of well-formed shoots. Shoots were excised and transferred onto MS-rooting medium (MS-104 but with 0.6~ agar and no plant growth regulators) with 500 ~rg/ml of carbenicillin and 100 ~g/ml of kanamycin. Surviving plantlets with well-formed roots were removed from the artificial medium, dipped alternately in a 0.06 50WP BenlateTM solution and in a rooting powder (Stim-root #1) containing indole-3 butyric acid, and transplanted into pasteurized PromixT"~ soil mixture. Plantlets were cov-ered with a transparent cover which was gradually lifted during the following 7 days.
3. T-DNA expression analysis 3.1 NPTII immunoassay A double antibody sandwich enzyme linked immu-nosorbent assay (DAS-ELISA) was used to analyze T-DNA
expression. The DAS-ELISA for NPTII detection was based on the Nagel et a1. (1992, Plant Mol. Biol. Rep., 10:263-272) procedure. Approximately 100 mg of leaf material was homogenized in 300 ~l of PBS-TP (137 mM
NaCl, 43 mM Na2HP04, 27 mM KC1, 14 mM KH2P04, 0.05 Tween'~-20, 2$ polyvinylpyrrolidone (PVP), pH 7.4).
Debris was removed by centrifugation (10,000 g for 2 minutes) and the concentration of soluble proteins was determined for every sample using the Bradford method (Bradford, N.M. 1976, Anal. Bio. Chem., 72:248-254).
Samples were diluted to 400 ~g/ml in PBS-TP.

Microtiter plates (Falcon) were coated with 200 ~1 of rabbit anti-NPTII (5 Prime -~ 3 Prime Inc.) diluted 1:500 in carbonate buffer (35 mM NaHC03, 15 mM Na2C03, pH 9.6). The antibody was incubated for 2 hours at 37°
C. Wells were washed five times with PBS-T (PBS-TP
without PVP) by alternately filling the wells with a multichannel pipette and emptying the plates in the sink. Wells were blocked with a solution of PBS-T
containing 2$ BSA for 30 minutes at room temperature (RT). Wells were washed five more times. Leaf samples (200 ~l) were added and incubated for 2 hours at RT.
Wells were washed five times and 200 ~1 of biotinylated NPTII antibody (5 Prime -~ 3 Prime Inc.), diluted 1:500 in PBS-TPO (PBS-TP with 0.2$ BSA), was added and incu-bated at RT for 1 hour. Wells were washed five times and 200 ~1 of P-nitrophenyl phosphate (PNP) diluted to 1 mg/ml in substrate buffer (9.7$ diethanolamine, pH
9.8) was incubated for approximately 40 minutes.
Absorbance was measured by a microtiter plate reader (Bio-RadT" 450) with a 405 nm filter.
3.2 HPC immuaoassay Plant samples were homogenized as for the NPTII
ELISA. A polyclonal rabbit anti-HPC (Sigma) was diluted 1:2000 in carbonate buffer and used to coat the wells of microtiter plates (Falcon) for 2 hours at 37°
C. Wells were washed five times with PBS-T and blocked with a solution of PBS-T and 2$ BSA for 30 minutes at RT. Wells were washed five more times and 200 ~1 of leaf extract was added and incubated overnight at 4°C.
Wells were washed five times and 200 ~1 of a polyclonal goat anti-HPC (Biopool), diluted 1:2000 in PBS-TPO, was added and incubated at RT for 2 hours. Wells were washed five times and 200 ~1 of a swine anti-goat IgG
(Cedarlane), diluted 1:3000 in PBS-TPO, was added and incubated at RT for 1 hour. Wells were washed five more times and 200 ~1 of 1 mg/ml PNP dissolved in sub-strate buffer was added. Color development was allowed to proceed in the dark for at least 1 hour and color intensity was measured using a microtiter plate reader with a 405 nm filter. A standard curve was also made using purified HPC (American Diagnostica) diluted in PBS-T.
3.3 Germination of R1 seeds on kanamycin-containing medium Seeds produced by RO plants were collected and germinated on artificial medium ( see described in sec-tion 2.1) containing 100 ~.g/ml of kanamycin to assay for antibiotic resistance among the R1 generation and segregation of the transferred gene.
3.4 Double transformation Rl seeds from S-2B transformed tobacco plant were grown in vitro and transformed using the pCP2 vec for inserted in Agrobacterium tumefaciens (as described in section 2. above).
A total of three S-2B controls (transformed once) and 104 potentially double transformant plants were analyzed for their HPC production. DAS-ELISA was used to determine HPC concentration (using PBS-T as a blank) while the Bradford method was used to measure the soluble protein concentration (the latter analysis was made with the Bio-RadT~ protein assay kit using ddH20 as a blank).
HPC production [~] - DAS-ELISA (HPC) x 100 Bradford (soluble proteins) 4. Rngineering tobacco for the expression of pro-tein C
The tobacco genome was modified in order to syn-thesize HPC. This involved the construction of two plasmids which contained a T-DNA and the HPC cDNA, the transfer of HPC cDNA into tobacco and the analysis of T-DNA expression among regenerated plants.
4.1 Binary vectors Two binary vectors for the expression of plant HPC were constructed in order to avoid using a single construct which could have errors acquired during DNA
manipulations. Both pCP2 (Fig. 1) and pLG3 (Fig. 2) constructs contained the right and left T-DNA border sequences and the selectable marker gene NPTII, which provides resistance to kanamycin and allows quick selection of engineered plants.
Plasmid pCP2 is a derivative of pBI 121. Its T
DNA is delimited by two T-DNA border sequences (triangles, Fig. 1) which flank the selectable marker gene neomycin phosphotransferase II (NPTII) preceded by the nopaline synthase promoter (P) and terminated with the nopaline synthase terminator (T). In addition, the cDNA of HPC ( cDNA HPC ) was c loned in between the CaMV
35S promoter (35S) and a nopaline synthase terminator (T). Approximate sizes of the elements between the border sequences are indicated and key restriction sites are indicated above the diagram.
Plasmid pLG3 is a derivative of pBI 121. Its T
DNA is delimited by two T-DNA border sequences (triangles, Fig. 2)) which flank the selectable marker gene neomycin phosphotransferase II (NPTII) preceded by the nopaiine synthase promoter (P) and terminated with the nopaline synthase terminator (T). In addition, the cDNA of HPC (cDNA HPC) was cloned downstream of a double CaMV 35S promoter (35S) and an AMV leader sequence and upstream of a nopaline synthase terminator (T). Approximate sizes of the elements between the border sequences are indicated and key restriction sites are indicated above the diagram.

The NPTII gene is regulated by the nopaline synthase promoter and the NOS-T.
pCP2 construct A 1420 by BcII fragment, which contained the cDNA of HPC, was cut out from pLPC and cloned into the BamHI site of vector pBI524. pBI524 is a derivative of pUC9 with, (5' to 3'), a dimer of the CaMV 35S pro-moter, an alfalfa mosaic virus (AMV) leader sequence, a polylinker (NcoI, XbaI, BamHI), and a NOS-T. The new construct was called pCPl. The cDNA sequence was pre-ferred to the genomic sequence because it is easier to manipulate smaller DNA sequences, and it is not known if plant cells will correctly splice out human introns.
In order to verify the orientation of the cloned HPC cDNA, a BglII restriction digestion was performed on plasmid DNA isolated from recovered E. coli colo-nies. BglII was expected to cleave at the 3' end of the double CaMV 35S promoter and 210 by away from the 5' end of HPC cDNA thus generating two DNA fragments of approximately 300 by and 4800 by if the cDNA was well oriented, that is the ATG codon from the HPC cDNA was immediately downstream of the AMV leader sequence. Two bands of the correct size were observed.
An XbaI-EcoRI cassette was isolated from pCPl and ligated in place of the XbaI-EcoRI cassette from pBIl2l. Therefore, the GUS gene with the NOS-T was replaced by the cDNA of HPC with its accompanying NOS-T
forming the plasmid pCP2. The HPC cDNA is under the control of the constitutive CaMV 35S promoter from pBI121 which is known to highly express foreign pro-teins.
pLG3 construct Vector pBI524 contained an undesirable ATG which is part of the NcoI restriction site. The ATG was deleted by cleaving pBI524 with NcoI, removing single stranded sticky ends with mung bean nuclease, and ligating the modified vector which was named pLGl. The removal of the Ncol restriction site was verified with a double digestion with Ncol and ScaI. Two bands were observed when pBI524 was digested with the two restric-tion enzymes while only one band appeared for pLGl indicating that the NcoI site was missing.
The HPC cDNA BclI fragment was cloned into the BamHI site of pLGl to create pLG2. Then, a HindIII
EcoRI cassette from pLG2 was cloned in place of the HindIII-EcoRI from pBIl2l. Therefore, the CaMV 35S
promoter along with the GUS gene and a NOS-T were replaced by a dimer of the CaMV 35S promoter with an AMV leader sequence, the cDNA of HPC, and a NOS-T.
Doubling the CaMV 35S promoter is known to markedly increase transcription, while the leader sequence enhances translation of the expressed protein.
Transfer of Plasmids pCP2 and pLG3 into A. tu~uefaciens Plasmids pCP2 and pLG3 were transferred into A.
tumefaciens LBA4404 using a freeze/thaw method. In order to verify whether the plasmids were successfully transferred, a non-radioactive Southern hybridization was attempted. The cDNA of HPC was observed to hybrid-ize to plasmid DNA isolated from A. tumefaciens and E.
coli transfected with pCP2 while there was no hybridi-zation with the control plasmid pBI121.
4.2 $ngineering tobacco and selection of transfor-wants Leaf discs were inoculated with four Agrobacte-rium inocula:
A) Seventy-five leaf discs were inoculated with LBA4404 without any binary vector. Half of the leaf discs were grown on kanamycin-containing medium in order to verify the efficacy of kanamycin selection while the other half were grown on medium without anti-biotic in order to recover negative control plants (untransformed tobacco).
B) One hundred leaf discs were inoculated with LBA4404 with the binary vector pBI121 as a control to monitor the transformation procedure's efficiency.
C) Two hundred leaf discs were inoculated with LBA4404 with the binary vector pCP2 for the expression of HPC.
D) Two hundred leaf discs were inoculated with LBA4404 with the binary vector pLG3 for the expression of HPC.
Kanamycin repressed regeneration of leaf discs inoculated with Agrobacterium without binary vector when cultivated on kanamycin-containing medium. Thus, the antibiotic selection was efficient in removing untransformed plants. Thirty-six plants were kanamycin resistant and survived transplantation to the greenhouse following inoculation with the binary vector pBI121. This indicates that the T-DNA transfer took place. A total of 230 kanamycin-resistant plants were regenerated: 118 engineered with pCP2 and 112 engineered with pLG3.
4.3 Analysis of HPC and NPTII expression Screening Rp plants for HPC expression A DAS-ELISA procedure was preferred to other ELISA
methods because this type of assay is less susceptible to non-specif is binding of antibodies. Two polyclonal antibodies for HPC were selected for the sandwich com-plex with the HPC antigen because polyclonal antibodies recognize several epitopes of HPC. The ELISA for HPC
was used to screen ail recovered plants engineered with the pCP2 and pLG3 binary vectors, and to find which RO
plants were potentially highly expressing HPC. A
dilution of 1:2,000 was used for coating the rabbit anti-HPC antibody and 1:1,000 for the goat anti-HPC
antibody. The dilution of swine anti-goat IgG was 1:3,000 as recommended by the manufacturer. Finally, the concentration of soluble protein in sap extracts was adjusted to 1 mg/ml. Negative control plants had ELISA values, after the background was removed, (PBS-TP
buffer in place of sap extract) ranging from -0.003 to 0.125 with~an average of 0.061 (n - 38). Results for engineered plants varied from 0.025 to 0.513. All plants with ELISA values above 0.375 were selected for the optimization of the ELISA conditions and for the accurate quantification of HPC expression.
Optimisation of HPC DA8-BLISA
Prior to the final HPC quantification, different ELISA conditions were tested in order to optimize the assay. Dilutions of the rabbit anti-HPC antibody and of the swine anti-goat IGg were kept as before while a 1:2,000 dilution of the goat anti-HPC antibody was used since reducing the quantity of antibody was found to only slow down color development. Eight concentrations of sap extract were tested: 1,000, 500, 250, 125, 63, 32, 16, and 8 ~g/ml and the quantity of HPC present was determined using purified HPC. Overall, the percentage of HPC increased with the dilution of the sap extract (Table 1). Sap was extracted from six plants which had ELISA readings greater than 0.375 during HPC screening.
Numbers represent the percentage of plant-produced HPC
among total soluble proteins. ELISA readings were blanked against sap of non-transformed plants and the percentage of HPC determined against a purified HPC
standard curve. A protein concentration of S ~g/ml was selected because a lower concentration of soluble protein would be too close to the lower limit of detection and a higher concentration would lead to underestimation of the amount of HPC present.

Table 1 Determination of optimal soluble protein concentration in sap extractions for the immunodetection of 8PC
Nghnlleaf82F S84B S82E 84F SS1A SS1B

extract 1000 0.0002 0.0000 0.0001 0.0000 0.0001 0.0001 500 0.0004 0.0002 0.0002 0.0001 0.0001 0.0003 250 0.0006 0.0003 0.0003 0.0001 0.0002 0.0004 125 0.0013 0.0005 0.0006 0.0001 0.0005 0.0007 63 0.0022 0.0011 0.0014 0.0002 0.0012 0.0016 32 0.0034 0.0017 0.0028 0.0003 0.0023 0.0032 18 0.0057 0.0028 0.0050 0.0005 0.0048 0.0054 8 0.0079 0.0031 0.0072 0.0004 0.0068 0.0089 Plant identification starting with "S" were engineen3d with the plasmid pCP2 while those starting with "SS" were engineered with pLG3.
Quantification of protein C expression To confirm results from the first ELISA screen-ing and to obtain a better estimate of the amount of protein C in transformed plants, an ELISA assay using the above antibodies and sample dilutions was per-formed. Sap was extracted from each Rp plant selected during the screening, and triplicates of the samples were incubated with the antibodies. Twelve non-trans-formed plants were used as negative controls to remove background. due to plant proteins, and ELISA values were plotted against a purified HPC standard curve. Some R~
plants expressed HPC at almost 0.03$ of their proteins, others failed to produce significant amounts (Table 2).
The five best plants, SSN, SSR, S2H, SS2D, and SSFF, were selected for a final quantification of HPC and verified for the expression of the NPTII marker gene.
The percentage of HPC relative to plant proteins was determined as well as standard deviation.

Table Z
Quantification of HPC among ap plaats which were potentially highly expressing protein C
PLANT % 8PC STD DRV PLANT % 8PC STD DBV

S5N 0.028 0.001 S5F 0.009 0.001 S5R 0.025 0.001 S1B 0.009 0.001 S2B 0.025 0.001 SS7G 0.008 0.002 SS2D 0.023 0.001 S1F 0.008 0.001 SSFF 0.020 0.000 S5W 0.008 0.001 SS7R 0.019 0.001 S5U 0.008 0.001 SSlA 0.019 0.001 SSSG 0.007 0.003 S6B 0.018 0.002 S1I 0.007 0.001 S5B 0.018 0.001 SS1F 0.005 0.002 S8C 0.017 0.003 SSCC 0.004 0.002 SS1B 0.017 0.005 SS4B 0.004 0.000 S7L 0.017 0.002 S4F 0.003 0.001 SS5J 0.017 0.001 S5H 0.002 0.002 SS3F 0.016 0.000 S7F 0.001 0.001 S5L 0.015 0.000 S5X 0.000 0.000 S2F 0.015 0.006 S70 -0.000 0.002 S5G 0.014 0.001 SSSB -0.002 0.001 SSBM 0.014 0.003 S7K -0.003 0.001 SS6B 0.014 0.002 S6H -0.003 0.002 SSSQ 0.013 0.000 S5K -0.005 0.001 SS2E 0.012 0.000 S5I -0.005 0.000 SS6P 0.011 0.002 Plant identfications starting with "S" v~re engineered with the plasmid pCP2 while those starting with "SS" were engineered with pLG3.
Readings represent the average of three replicates. "96 HPC" is the percent among total soluble tobacco proteins.
$xpression of the marker gene NPTII and HPC among the bast five plants The same HPC ELISA procedure as above was used to confirm HPC quantification of the best five HPC-pro ducing tobacco plants. In addition, NPTII ELISA was used to verify the expression of the marker gene.

After removing background (PBS-TP buffer) from ELISA
readings, all five plants had positive NPTII readings while all four negative control plants had negative NPTII readings (Table 3). Moreover; HPC percentage among soluble proteins was slightly negative for con-trol plants because they consistently had ELISA read-ings lower than the PBS-TP buffer used to blank read-ings. Some engineered plants produced 0.02 to 0.03 of HPC. S2B, SSN, SSR, SSFF were transformed with pCP2, SS2D was transformed with pLG3 and C1B, C2A, C3A, C3D
were not transformed with a binary vector.
Table 3 Percentage of 8PC and detection of NPTII from selected Rp plants NPTII NPTII % HPCj' HPC

RSADINGSa STD DBV. STD DEV.

S2B 0.371 0.015 0.033 0.001 S5N 0.048 0.014 0.024 0.001 85R 0.040 0.010 0.023 0.001 SS2D 0.155 0.030 0.021 0.001 S5Fg 0.134 0.050 0.020 0.002 C1B -0.079 0.017 -0.008 0.001 C2A -0.101 0.018 -0.007 0.001 C3A -0.152 0.007 -0.010 0.001 C3D -0.084 0.002 -0.004 0.003 a: NPTII EIISA readings were blanked against PBS-TP buffer reading.
b: °~6 HPC is the percentage of HPC among tobacco soluble proteins.
Numbers represent the mean of three replicates.
Expression of HPC in R1 families The ELISA assay for HPC was used to verify the synthesis of HPC among the progenies of two of the best five HPC-producing tobacco plants. Seeds of S2B, SSN, and C3A were collected and germinated in soil. When R1 plants had approximately four leaves, an ELISA was used to quantify HPC levels (Table 4). Twenty seedlings were tested per mother plant and ratios of seedlings synthesizing HPC versus those not synthesizing HPC were statistically analyzed. Mother plants were S2B and SSN, which were transformed with pCP2. The ratio of S5N progenies expressing HPC to those not expressing HPC was in agreement with the segregation of one dominant gene (Table 4). All S2B progeny expressed HPC, suggesting that two or more T-DNAs were present in the plant genome of the S2B mother plant.
Table 4 Inheritance of HPC expression in the R1 generation % HPC S28a SSNb C3A

Detected 0.02 11 0 0 0.01 9 14 0 0.00 0 6 10 Numbers represent the number of R1 plants from which 0.02, 0.01, or 0.0096 of soluble HPC protein was detected by ELISA.
a: Chi-square analysis at p = 0.05 for the segregation of iwo or more T-DNA
(15:1,..:1).
b: Chi-square analysis at p = 0.05 for the segregation of one T-DNA (3:1).
Double transformation The S2B tobacco plants used here for a second transformation had previously been screened for their ability to survive on kanamycin medium. It was there-fore not possible to use kanamycin selection to iden-tify plants that were transformed a second time, since the same plasmid was used for both transformations.
Plants capable of producing larger amounts of HPC (as a result of a second transformation event) were identi-fied by comparing the putative double transformants to three single transformed plants (F1 generation of S-2B
plant) for the level of HPC using a DAS-ELISA.
First screen Figs. 3 to 6 show the relative HPC content of various second transformants; of 104 plants, eight produced significantly higher amounts of HPC: A 9-4 (#37). B 4-5 (#80), B 5-1 (#81), B 5-2 (#82), B 8-4 (#94), B 9-2 (#97), B 9-4 (#99) and B 10-1 (#100). To avoid eliminating plants that were high producers but were not among the eight best identified above, the 17 best plants (DAS-ELISA units / Bradford units higher than 7) were transferred to the greenhouse for further analysis (indicated by arrows on Figs. 3 to 6).
Second and third screens Following two additional quantitative evalu-ations, three plants showed consistently higher HPC
content:' plants B 2-2, B 5-1 and B 8-1. During the second and the third evaluations, their average HPC
production were respectively 42~ and 23~ higher than the three controls.
The plasmid pCP2 contained the marker gene NPTII
and the cDNA of HPC under the control of the CaMV 35S
promoter. Plasmid pLG3 contained NPTII and the cDNA of HPC was controlled by a dimer of CaMV 35S promoter with an AMV leader sequence. Growing non-transformed tobacco plants on kanamycin-containing medium indicated that the antibiotic selection was efficient while engineering leaf discs with pBI121 showed that the T-DNA was transferred properly to plant cells. The best HPC-producing tobacco plants were shown to express NPTII and HPC at levels representing 0.02 to 0.03 of their soluble proteins. The expression of HPC in R1 plants was transmitted with a 3:1 ratio for S5N progeny while all S2B progeny expressed HPC, suggesting that the mother plant had two or more T-DNA inserts in its genome.
Some plants, transformed a second time, produced more HPC than the original transformants, with the best "twice-transformed" plant producing 43~ more than the original mother plant.

5. Partial purification of plant-produced protein C
A series of experiments were designed to par tially purify HPC in order to eventually characterize the protein and assay for its activity. All manipula tions were performed either on ice or in a cold room at 4°C .
Anion-exchange chromatography HPC expressed in tobacco plants was partially purified using an affinity chromatography purification protocol. Approximately IO g of tobacco leaves with positive HPC ELISA readings were homogenized in a WaringT"~ blender with 30 ml of extraction buffer ( 20 mM
Tris-C1 pH 7.4, 150 mM NaCl, 4 mM EDTA pH 7.4, 5 mM
benzamidine-HCl). Most debris was removed by filtra-tion through MiraclothT"~ and by centrifugation at 15,000 g for 10 minutes. A 16 X 200 mm column (Pharmacia) was filled with 10 ml of Fast Flow Q'~'~ Sepharose (FFQ) anion exchanger (Pharmacia). FFQ resin was washed with 30 ml of equilibration buffer ( 20 mM Tris-C1 pH 7 .4, 150 mM
NaCl, 2 mM EDTA pH 7.4, 2 mM benzamidine-HC1). The sample was applied at the surface of the resin followed by another 30 ml of equilibration buffer. HPC was eluted by injecting 30 ml of elution buffer (20 mM
Tris-C1 pH 7.4, 150 mM NaCl, 10 mM CaCl2, 2 mM
benzamidine-HCl). A high salt elution was then applied ( 20 mM Tris-C1 pH 7. 4, 400 mM NaCl, 2 mM benzamidine-HC1) and the resin was cleaned with a solution of 2 M
sodium acetate.
Sephadex desalting Salts were removed from sap extracts immediately after centrifugation, using PD-10'''M columns (Pharmacia) according to the manufacturer's instructions.

Complete debris removal After centrifugation of the sap extract at 15,000 g, the solution was centrifuged for 1 hour at 100,000 g.

6. Partial purification of plant-produced HPC using ion-exchange chromatography Following tests to optimize the pH of buffers (as described above in section 5.), three preliminary tests were made on the automated liquid chromatography system (BioRad) in order to set larger scale working conditions.
During experiments, the relative amount of total protein was monitored using the system's UV monitor, whereas the relative amount of HPC was determined on eluted samples using DAS-ELISA. All experiments were conducted under the following conditions .
Samples: Approximately 5.0 g of HPC+ tobacco leaves were homogenized with 15.0 ml of extraction buffer and removed from debris using Miracloth T"~ and 10 minutes of centrifugation at 15,000 g.
Buf f ers Extraction buffer: 20 mM Tris-C1, 150 mM NaCl, 4 mM
EDTA, 5 mM benzamidine-HC1, pH 8.
Equilibration buffer:20 mM Tris-C1, 150 mM NaCl, 2 mM
EDTA, 2 mM benzamidine-HC1, pH 5 or 8.
Elution buffer: 20 mM Tris-C1, 150 mM NaCl, 2 mM
benzamidine-HC1, pH 8.
Ca2+ elution buffer: 20 mM Tris-C1, 150 mM NaCl, 10 mM
CaCl2, 2 mM benzamidine-HC1, pH
8.
Ion-exchange conditions Column: Sepharose'"d Q (Sigma), 5 ml Buffers: Extraction and equilibration at pH 8 Elution at pH 8 mM Ca2+ elution at pH 8 10 mM Ca2+ salt elution with 200 and 400 mM NaCl at pH 8 7. Analysis of plant-produced 8PC
7.1 SDS-PAG$ aad Western immunoblot Human Protein C is a 62,000 Da protein made of 10 two subunits (41,000 and 21,000 Da) linked by a disul fide bridge. The addition of mercaptoethanol to the sample before loading onto the gel has the effect of destroying that disulfide bridge. Gel electrophoresis and Western blots were used to characterize the protein produced in tobacco plants.
Western immunoblot Immunodetection was conducted by blocking 30 minutes with PBS and 5$ powdered milk. Rabbit anti-HPC
antibody (1:7,000) was added to the blocking solution.
After an overnight incubation, the nitrocellulose was washed three times for 10 minutes each with PBS + 0.5~
Triton's followed by a 10 minutes wash in TBS. The nitrocellulose was then transferred to TBS + 5~ pow-dered milk supplemented with a 1:14,000 dilution of goat anti-rabbit IgG conjugate (Bio-Rad) for 60 min-utes. The nitrocellulose was washed four times in TBS
for 15 minutes each. Colorimetric detection of HPC was carried out using alkaline phosphatase activity.
Using these parameters, the samples used were:
S 25 ng of standard HPC (Sigma);
T+ HPC+ tobacco plant (S-2B);
T- HPC- tobacco plant; and T++ HPC+ tobacco plant (second transformant).
This immunoblot confirmed the presence of HPC in transformed tobacco plants ( Fig. 7 ) . Both transformed plants displayed a major protein reacting with anti-HPC

serum (lanes T+ and T++: 61,300 MW) whereas no band was detected in the control (non-transformed) tobacco plant (lane T-). Lane S contained 25 ng of Sigma HPC. Two major proteins were detected corresponding to the heavy and light chains (40,100 and 23,700 Da, respectively).
From these results, it was concluded that: 1-HPC was produced in transformed plants; 2- HPC was not completely processed and possibly not cleaved into the heavy and light chain.
7.2 HPC location in tobacco plant HPC concentration was measured in different tis-sues of a twice-transformed tobacco plant: 1- roots;
2- stem; 3- primary vein of the leaf ; 4- leaf without the primary vein. An equal amount of each part of the plant (5.0 g) was homogenized with 15.0 ml of extrac tion buffer (20 mM Tris, 10 mM benzamidine-HC1, pH
7.4), filtered and centrifuged 10 minutes at 15,000 g.
These four extracts were analyzed for HPC content using DAS-ELISA.
HPC content of different tissues of tobacco plants was measured using DAS-ELISA. Leaves showed a higher concentration than other parts tested (Table 5).
HPC concentration was lowest in the roots.

Table 5 HPC content of various tissues measures using Plant tissue Amount of HPC
(Ng HPC / g fresh tissue) Leaves 0.388 Stem 0.274 Veins 0.235 Roots 0.173 7.3 Biological activity using delay in coagulation time The Acticlot'~'M assay kit (from American Diagnos-tica) was used. HPC+ and HPC- sera, dilution buffer and solutions were prepared according to the manufac-turer's instructions. Tubes, ActiclotT"~ activator and CaCl2 stock solution were prewarmed to working tempera-ture (37°C).
Prior to testing, HPC samples were prepared as follows: 50 ~,1 undiluted sample, 50 ~tl HPC deficient plasma, 400 ~,1 American Diagnostica's dilution buffer.
A 50 ~.~,1 volume of this prepared sample was mixed with an equal amount of HPC deficient plasma and incubated for 2 minutes. A volume of 50 ~,.~,1 of Acticlot activator was mixed with the sample solution and incubated five more minutes. Finally, 50 ~1 of calcium chloride stock solution was added and clotting time was monitored by the tilt-tube technique.
Changes in coagulation times were observed in several experiments when tobacco extracts were tested for clotting activity. Some of the clotting assays indicated that biological activity was present.

Coagulation times of a tobacco extract using American Diagnostica's Acticlot assay kit HPC+ HPC chromato chromato plasma #9 #8 HPC neg.

tobacco control Assay 1 60 sec. 44 sec. 52 sec. 107 sec.

Assay 2 - - 145 sec. 130 sec.

Assay 3 - - 40 sec_ 60 sec.

Assay 4 - - 140 sec. 130 sec.

The present invention will be more readily un derstood by referring to the following examples which are given to illustrate the invention rather than to limit its scope.
Expression vector pCP2 construction for the production of human protein C
A 1420 by BclI fragment, which contained the cDNA of HPC, was cut out from pLPC and cloned into the BamHI site of vector pBI524. pBI524 is a derivative of pUC9 with, (5' to 3'). a dimer of the CaMV 35S pro-moter, an alfalfa mosaic virus (AMV) leader sequence, a polylinker (NcoI, XbaI, BamHI), and a NOS-T. The new construct was called pCPl. In order to verify the ori-entation of the cloned HPC cDNA, a BglII restriction digestion was performed on plasmid DNA isolated from recovered E. coli colonies. BglII is expected to cleave at the 3' end of the double CaMV 35S promoter and 210 by away from the 5' end of HPC cDNA thus gen-erating two DNA fragments of approximately 300 by and 4800 by if the cDNA was well oriented, that is the ATG
codon from the HPC cDNA was immediately downstream of the AMV leader sequence.

An XbaI-EcoRI cassette was isolated from pCPl and ligated in place of the Xbal-EcoRI cassette from pBI121. Therefore, the GUS gene with the NOS-T was replaced by the cDNA of HPC with its accompanying~NOS-T
forming the plasmid pCP2. The HPC cDNA is under the control of the constitutive CaMV 35S promoter from pBI121.
Plasmid pCP2 was transferred into A. tumefaciens LBA4404 using the freeze/thaw method. In order to verify whether the plasmids were successfully trans ferred, a non-radioactive Southern hybridization was performed. The cDNA of HPC was observed to hybridize to plasmid DNA isolated from A. tumefaciens and E.
coli .
Seeds of Nicotiana tabacum cv. Xanthi were ster-ilized for 15 minutes in a 10$ bleach solution with a drop of detergent (TweenTM-20). Seeds were washed at least five times with sterile distilled water and allowed to germinate and grow on artificial medium composed of the basal MS salts, B5 vitamins, 3$
sucrose, and 0.6~ agar (Anachemia) at pH 5.7-5.8. Seed were grown under a 16 hour photoperiod with a light intensity of 50 ~E and a temperature of 24°C.
A. tumefaciens was grown in Luria Broth (LB) (1$
tryptone, 0.5$ yeast extract, 85 mM NaCl, pH 7.0) medium supplemented with 50 ~g/ml kanamycin and 25 ~g/ml of streptomycin for 18 hours or until the optical density at 595 nm reached 0.5 to 1Ø Cells were spun down at 3,000 g for 5 minutes and the pellet was resuspended to its initial volume with MS-104 medium (MS basal salts, B5 vitamins, 3$ sucrose, 1.0 ~g/ml benzylaminopurine (BAP), 0.1 ~g/ml naph-taleneacetic acid (NAA), pH 5.7-5.8, and 0.8~ agar) without agar.

Leaf squares of about 64 mm2 were dissected using a sharp scalpel, immersed in the inoculum for 15-30 minutes and plated onto MS-104 medium for 2 days under a 16 hour photoperiod, under low light intensity (20 ~E) at 24°C for cocultivation. Leaf discs were washed alternately three times with sterile distilled Water for 1 minute and sterile distilled water supple-mented with 500 ~g/ml of carbenicillin for 5 minutes.
Leaf discs were transferred to MS-104 medium with 500 ~g/ml of carbenicillin for another 2 days under the same environmental conditions.
Leaf discs were washed as above, plated on MS-104 medium with 500 ~g/ml of carbenicillin and 100 ~g/ml of kanamycin, and grown using the above envi-ronmental conditions until calluses appeared. The light intensity was increased to 50 ~E and the explants were allowed to grow until development of well-formed shoots. Shoots were excised and transferred onto MS-rooting medium (MS-104 but with 0.6$ agar and no plant growth regulators) with 500 ~g/ml of carbenicillin and 100 ~g/ml of kanamycin. Surviving plantlets with well-formed roots were removed from the artificial medium, dipped alternately in a 0.06$ 50WP Benlate solution and in a rooting powder (Stim-root #1) containing indole-3 butyric acid, and transplanted into pasteurized Promix soil mixture. Plantlets were covered with a transpar-ent cover which was gradually lifted during the follow-ing 7 days.
LE II
Expression vector phG3 construction for the production of human protein C
Vector pBI524 contained an undesirable ATG which is part of the NcoI restriction site. The ATG was deleted by cleaving pBI524 with NcoI, removing single stranded sticky ends with mung bean nuclease, and ligating the modified vector which was named pLGl. The removal of the NcoI restriction site was verified with a double digestion with NcoI and ScaI. Two bands were observed when pBI524 was digested with the two restric-tion enzymes while only one band appeared for pLGl indicating that the NcoI site was missing.
The HPC cDNA BclI fragment was cloned into the BamHI site of pLGl to create pLG2. Then, a HindIII
EcoRI cassette from pLG2 was cloned in place of the HindIII-EcoRI from pBIl2l. Therefore, the CaMV 35S
promoter along with the GUS gene and a NOS-T were replaced by a dimer of the CaMV 35S promoter with an AMV leader sequence, the cDNA of HPC, and a NOS-T.
Plasmid pLG3 was transferred into A, tumefaciens LBA4404 using the freeze/thaw method. In order to verify whether the plasmids were successfully trans-ferred, a non-radioactive Southern hybridization was performed. The cDNA of HPC was observed to hybridize to plasmid DNA isolated from A. tumefaciens and E.
coli .
Seeds of Nicotiana tabacum cv. Xanthi were ster-ilized for 15 minutes in a 10$ bleach solution with a drop of detergent (TweenTM-20). Seeds were washed at least five times with sterile distilled water and allowed to germinate and grow on artificial medium composed of the basal MS salts, B5 vitamins, 3~
sucrose, and 0.6$ agar (Anachemia) at pH 5.7-5.8. Seed were grown under a 16 hour photoperiod with a light intensity of 50 ~E and a temperature of 24°C.
A. tumefaciens was grown in Luria Broth (LB) (1~
tryptone, 0.5$ yeast extract, 85 mM NaCl, pH 7.0) medium supplemented with 50 ~g/ml kanamycin and 25 ~g/ml of streptomycin for 18 hours or until the optical density at 595 nm reached 0.5 to 1Ø Cells were spun down at 3,000 g for 5 minutes and the pellet was resuspended to its initial volume with MS-104 medium (MS basal salts, B5 vitamins, 3~ sucrose, 1.0 ~g/ml benzylaminopurine (BAP), 0.1 ~g/ml naph-taleneacetic acid (NAA), pH 5.7-5.8, and 0.8~ agar) without agar.
Leaf squares of about 64 mm2 were dissected using a sharp scalpel, immersed in the inoculum for 15-30 minutes and plated onto MS-104 medium for 2 days under a 16 hour photoperiod, under low light intensity (20 ~E) at 24°C for cocultivation. Leaf discs were washed alternately three times with sterile distilled water for 1 minute and sterile distilled water supple-mented with 500 ~g/ml of carbenicillin for 5 minutes.
Leaf discs were transferred to MS-104 medium with 500 ~g/ml of carbenicillin for another 2 days under the same environmental conditions.
Leaf discs were washed as above, plated on MS
104 medium with 500 ~g/ml of carbenicillin and 100 ~g/ml of kanamycin, and grown using the above envi ronmental conditions until calluses appeared. The light intensity was increased to 50 NE and the explants were allowed to grow until development of well-formed shoots. Shoots were excised and transferred onto MS-rooting medium (MS-104 but with 0.6$ agar and no plant growth regulators) with 500 ~g/ml of carbenicillin and 100 ug/ml of kanamycin. Surviving plantlets with well-formed roots were removed from the artificial medium, dipped alternately in a 0.06$ 50WP Benlate solution and in a rooting powder (Stim-root #1) containing indole-3 butyric acid, and transplanted into pasteurized Promix soil mixture. Plantlets were covered with a transpar-ent cover which was gradually lifted during the follow-ing 7 days.

BXAMPLE III
i~xpression vector construction for the production of chicken nuclear oncoprotein p53.
Proceeding as for Example I, but using a cDNA
sequence coding for chicken nuclear oncoprotein p53 instead of the cDNA of the human protein C gene.
Vector pBI524 contained an undesirable ATG which is part of the NcoI restriction site. The ATG was deleted by cleaving pBI524 with NcoI, removing single stranded sticky ends with mung bean nuclease, and ligating the modified vector which was named pLGl. The removal of the NcoI restriction site was verified with a double digestion with NcoI and ScaI. Two bands were observed when pBI524 was digested with the two restric-tion enzymes while only one band appeared for pLGl indicating that the NcoI site was missing.
The chicken cDNA for nuclear oncoprotein p53 is excised using EcoRI restriction digestion. T4 DNA
polymerase is used to fill in the 3' recessed end and therefore eliminate the EcoRI site. BamHI linkers (5'-CGGATCCG-3'.) are added by ligation with T4 DNA ligase.
The BamHI ends are then digested with BamHI and ligated into the BamHI site of pLGl to create pLG53. Then, a HindIII-EcoRI cassette from pLG2 is cloned in place of the HindIII-EcoRI from pBI121. Therefore, the CaMV 35S
promoter along with the GUS gene and a NOS-T were replaced by a dimer of the CaMV 35S promoter with an AMV leader sequence, the cDNA of HPC, and a NOS-T.
Plasmids pLG53 is transferred into A. tumefaciens LBA4404 using the freeze/thaw method.
Seeds of Nicotiana tabacum cv. Xanthi are ster-ilized for 15 minutes in a 10~ bleach solution with a drop of detergent (TweenTM-20). Seeds are washed at least five times with sterile distilled water and allowed to germinate and grow on artificial medium composed of the basal MS salts, 85 vitamins, 3~ sucr-ose, and 0.6$ agar (Anachemia) at pH 5.7-5.8. Seed are grown under a 16 hour photoperiod with a light inten-sity of 50 ~E and a temperature of 24°C.
A. tumefaciens is grown in Luria Broth (LB) (1$
tryptone, 0.5$ yeast extract, 85 mM NaCl, pH 7.0) medium supplemented with 50 ~g/ml kanamycin and 25 ~.g/ml of streptomycin for 18 hours or until the optical density at 595 nm reaches 0.5 to 1Ø Cells are spun down at 3, 000 g for 5 minutes and the pellet is resuspended to its initial volume with MS-104 medium (MS basal salts, B5 vitamins, 3~ sucrose, 1.0 ~g/ml benzylaminopurine (BAP), 0.1 ~g/ml naphtaleneacetic acid (NAA), pH 5.7-5.8, and 0.8~ agar) without agar.
Leaf squares of about 64 mm2 are dissected using a sharp scalpel, immersed in the inoculum for 15-30 minutes and plated onto MS-104 medium for 2 days under a 16 hour photoperiod, under low light intensity (20 ~E) at 24°C for cocultivation. Leaf discs are washed alternately three times with sterile distilled water for 1 minute and sterile distilled water supplemented with 500 ~.g/ml of carbenicillin for 5 minutes. Leaf discs are transferred to MS-104 medium with 500 ~g/ml of carbenicillin for another 2 days under the same environmental conditions.
Leaf discs are washed as above, plated on MS-104 medium with 500 ~g/ml of carbenicillin and 100 ~g/ml of kanamycin, and grown using the above environmental con-ditions until calluses appeared. The light intensity is increased to 50 ~E and the explants allowed to grow until development of well-formed shoots. Shoots are excised and transferred onto MS-rooting medium (MS-104 but with 0.6~ agar and no plant growth regulators) with 500 ~g/ml of carbenicillin and 100 ~g/ml of kanamycin.
Surviving plantlets with well-formed roots are removed from the artificial medium, dipped alternately in a 0.06 50WP Benlate solution and in a rooting powder (Stim-root #1) containing indole-3 butyric acid, and transplanted into pasteurized Promix soil mixture.
Plantlets are covered with a transparent cover which is gradually lifted during the following 7 days.
While the invention has been described in con-nection with specific embodiments thereof, it will be understood that it is capable of further modifications and this application is intended to cover any varia-tions, uses, or adaptations of the invention following, in general, the principles of the invention and including such departures from the present disclosure as come within known or customary practice within the art to which the invention pertains and as may be ap-plied to the essential features hereinbefore set forth, and as follows in the scope of the appended claims.

Claims (11)

I CLAIM:

1. An expression vector pCP2 for the large scale production of a human or animal protein, which comprises a DNA construct consisting of operatively linked DNA coding for a plant promoter (35S), a mRNA
stabilizer (AMV), a transcription terminator (NOS) and said human or animal protein to be expressed, wherein the human or animal protein is selected from the group consisting of colony stimulating factors, relaxins, polypeptide hormones, growth factors and coagulation factors.

2. An expression vector pCP2 for the large scale production of a human or animal protein, which comprises a DNA construct consisting of operatively linked DNA coding for a plant promoter (35S), a mRNA
stabilizer (AMV), a transcription terminator (NOS) and said human or animal protein to be expressed, wherein said human or animal protein is selected from the group of proteins consisting of human protein C (HPC), factor VIII, growth hormone, erythropoietin, interleukin 1, interleukin 2, interleukin 3, interleukin 4, interleukin 5, interleukin 6, and interleukin 7.

3. An expression vector pLG3 for the large scale production of a human or animal protein, which comprises a DNA construct consisting of operatively linked DNA coding for a plant promoter (double 35S), a mRNA stabilizer (AMV), a transcription terminator (NOS), and said human or animal protein to be expressed, wherein the human or animal protein is selected from the group consisting of colony stimulating factors, relaxins, polypeptide hormones, cytokines, growth factors and coagulation factors.

4. An expression vector pLG3 for the large scale production of a human or animal protein, which comprises a DNA construct consisting of operatively linked DNA coding for a plant promoter (double 35S), a mRNA stabilizer (AMV), a transcription terminator (NOS), and said human or animal protein to be expressed, wherein said human or animal protein is selected from the group of proteins consisting of human protein C (HPC), factor VIII, growth hormone, erythro-poietin, interleukin 1, interleukin 2, interleukin 3, interleukin 4, interleukin 5, interleukin 6, and interleukin 7.

5. A plant bioreactor for the large scale production of a human or animal protein, which comprises dicotyledonous plants transformed with the expression vector of claim 2 which further includes a suitable selectable marker for plant transformation.

6. A plant bioreactor for the large scale production of a human or animal protein, which comprises dicotyledonous plants transformed with the expression vector of claim 4 which further includes a suitable selectable marker for plant transformation.

7. The plant bioreactor of claim 5 or 6, wherein said suitable selectable marker is kanamycin resistance.

8. A method of large scale production of human protein C, which comprises the steps of:
a) inserting the recombinant expression pCP2 vector of claim 2 into a plant cell using Agrobacterium transformation, said expression vector comprising operatively linked DNA coding for a plant promoter, a transcription terminator and said human protein C to be expressed flanked by T-DNA
borders and a suitable selectable marker for plant transformation, whereby the plant cell becomes a transformed plant cell; and b) extracting said expressed human protein C from said transformed plant cell.

9. The method of claim 8, wherein said suitable selectable marker is kanamycin resistance.

10. A method of large scale production of human protein C, which comprises the steps of:
a) inserting the recombinant expression vector pLG3 of claim 4 into a plant cell using Agrobacterium transformation, said expression vector comprising operatively linked DNA coding for a plant promoter, a mRNA stabilizer, a transcription terminator, and said human protein C to be expressed flanked by T-DNA borders and a suitable selectable marker for plant transformation, whereby the plant cell becomes a transformed plant cell; and b) extracting said expressed human protein C from said transformed plant cell.

11. The method of claim 10, wherein said suitable selectable marker is kanamycin resistance.