Abstract
The present invention relates to the use of transgenic plants for the expression of vitamin B12 (cobalamin) binding proteins. Plant cells are transformed with nucleotide sequences adapted for expression and secretion of vitamin B12 binding proteins. The present invention also relates to the use of recombinant vitamin B12 binding proteins from plants in analytical tests and treatment of vitamin B12 deficiency. Also disclosed is a method for purification of recombinant vitamin B12 binding proteins.
Claims
1. A composition, the composition comprising intrinsic factor and a plant specific carbohydrate chain on the intrinsic factor, wherein all of the intrinsic factor is in the apo form.
Description
BRIEF DESCRIPTION OF THE DRAWINGS
(1) The invention will now be described with reference to the following examples, which should not be construed as in any way limiting the invention.
(2) The examples refer to the figures in which:
(3) FIG. 1 shows the nucleotide sequence encoding extensin signal peptide sequence fused to the mature human intrinsic factor encoding region and part of the 3′-untranslated region (SEQ ID NO:2). This nucleotide sequence is a fusion of a nucleotide sequence (position 7-119) adopted from GenBank accession no. AF104327 which encodes an Arabidopsis thaliana extensin-like signal peptide with the amino acid sequence MASSSIALFLALNLLFFTTISA (SEQ ID NO:1). The methionine start codon (ATG) of this signal peptide sequence is underlined. The nucleotide sequence encoding mature human intrinsic factor (position 120-1316) is shown in bold letters and adopted from GenBank accession no X76562. This sequence is followed by a translational stop codon (TAA) which is underlined and a nucleotide sequence from the 3′-untranslated region of the intrinsic factor mRNA. The underlined restriction sites XbaI (position 1-6) and XmaI (position 1425-1430) was introduced to facilitate cloning in the plant transformation vector.
(4) FIG. 2 shows the amino acid sequence of mature human intrinsic factor which was encoded by the nucleotide sequence shown in bold letters in FIG. 1 at position 120-1316 (SEQ ID NO:4).
(5) FIG. 3 shows a nucleotide sequence (SEQ ID NO:5) at position 7-129 from Phaseolus vulgaris CH5B-chitinase, GenBank accession no. S43926 and it encodes a signal peptide with the amino acid sequence MKKNRMMIMICSVGVVWMLLVGGSYG (SEQ ID NO:6). This nucleotide sequence was fused to the nucleotide sequence shown in bold letters in FIG. 1 that encodes the mature human intrinsic factor. The XbaI restriction site used for cloning is underlined at position 1-6 and the translational start codon ATG is underlined at position 52-54.
(6) FIG. 4 shows a nucleotide sequence (SEQ ID NO:7) at position 7-144 from Nicotiana tabacum glucan beta-1,3-glucanase gene, GenBank accession no. M60402 and it encodes a signal peptide with the amino acid sequence MSTSHKHNTPQMAAITLLGLLLVASSIDIAGA (SEQ ID NO:8). This nucleotide sequence was fused to the nucleotide sequence shown in bold letters in FIG. 1 that encodes the mature human intrinsic factor. The XbaI restriction site is underlined (position 1-6) and the translational start codon ATG is underlined at position 49-51.
(7) FIG. 5 shows the nucleotide sequence (SEQ ID NO:9) encoding the extensin signal peptide sequence fused to the mature human transcobalamin encoding region. This nucleotide sequence is a fusion of a nucleotide sequence (position 7-119) adopted from GenBank accession no. AF104327 which encodes an Arabidopsis thaliana extensin-like signal peptide with the amino acid sequence MASSSIALFLALNLLFFTTISA (first 22 amino acids of SEQ ID NO:3). The methionine start codon (ATG) of this signal peptide sequence is underlined. The nucleotide sequence encoding mature human transcobalamin (position 120-1346) is shown in bold letters and adopted from GenBank accession no 000355. This sequence is followed by a translational stop codon (TAG) at position 1347-1349 which is underlined. The underlined restriction sites XbaI (position 1-6) and XmaI (position 1350-1355) was introduced to facilitate cloning in the plant transformation vector.
(8) FIG. 6 shows the amino acid sequence (SEQ ID NO:11) of mature human transcobalamin which was encoded by the nucleotide sequence shown in bold letters in FIG. 5 at position 120-1346.
(9) FIG. 7 shows the native intrinsic factor (o) present in human gastric juice and recombinant human intrinsic factor extracted from transgenic plants of Arabidopsis thaliana (g) which were compared concerning their ability to bind cobalamin or cobinamid. Equal amounts of the respective proteins were added to a mixture containing a fixed amount of cobalt (57Co) labelled cobalamin mixed with increasing amounts of non-radioactive cobalamin or cobinamide (X axis). Free and bound ligand were separated and the amount of 57Co attached to the protein was measured in a gamma counter. The 57Co fraction bound relative to the amount of 57Co bound in the absence of unlabeled cobalamin or cobinamide was calculated (Y axis). The figure shows that recombinant IF behaves as does native IF with specificity for binding cobalamin but not cobinamid.
(10) FIG. 8 shows how polyclonal antibodies against human gastric intrinsic factor and alkaline phosphatase-conjugated immunoglobulins were used to visualize intrinsic factor on a Western blot containing proteins separated by SDS-PAGE. Transgenic Arabidopsis thaliana plants expressing recombinant human intrinsic factor were harvested and homogenized in 0.2 M phosphate buffer before centrifugation and analysis of the supernatant. For comparison transgenic yeast (Pichia pastoris) containing an insert of human intrinsic factor was used for expression of recombinant human intrinsic factor. Secreted proteins form the fermentation media were precipitated with ammonium sulfate (80% w/v) before dialysis against 20 mM Tris pH 8.0 and analysis. Purified human gastric intrinsic factor was also used for comparison. An aliquot of each sample was treated with the enzyme PNGaseF to remove carbohydrate from the proteins. Lane: “Plant IF” contains the proteins from the plant extract; “Plant IF+PNGaseF” contains the proteins from the plant extract treated with PNGaseF; “Yeast IF” contains yeast protein; “Yeast IF+PNGaseF contains yeast protein treated with PNGaseF; “Human gastric IF” contains purified human IF and “Human gastric IF+PNGaseF” contains purified human IF treated with PNGaseF. The arrow marks the 45 kDa band with deglycosylated IF from the three samples treated with PNGaseF.
(11) The blot shows that the glycosylated form of IF from the three samples have significant differences in molecular weight but the deglycosylated samples contain mature IF with the same molecular weight.
(12) FIG. 9 compares the glycosylation and molecular weight of intrisicic factor prepared from various sources. PAS staining was used to visualize glycoproteins. Recombinant human intrinsic factor (rhIF) from transgenic plants (Arabidopsis thaliana) and transgenic yeast (Pichia pastoris) were isolated and purified by affinity chromatrography on a column with cobalamin. Purified human gastric IF was used for comparison. Bovine PAS-3 is heavily glycosylated and function as a control for PAS staining. Bovine serum albumin (BSA) has no glycosylation and function as a “negative” control for PAS staining. The samples were run in SDS-PAGE in duplicate gels. The second gel was stained with coomassie brilliant blue.
(13) The gels show that recombinant plant IF is glycosylated since an approximately 50 kDa rhIF band from transgenic plants is stained with PAS.
(14) FIG. 10 shows the purification method for isolating rhIF and a blot of the protein fragments used for N-terminal sequencing. Purified recombinant human intrinsic factor (rhIF) was run by SDS-PAGE, blotted onto a PVDF-membrane and stained with coomassie brilliant blue. The mature rhIF and the partially cleaved rhIF fragments (proteases present in Arabidopsis thaliana cleave rhIF) were analyzed by amino acid sequencing. N-terminal sequencing of the three fragments marked with arrows showed that the approximately 50 kDa fragment has the same N-terminus as mature human gastric intrinsic factor. The lower two fragments (30 and 20 kDa) are a result of a proteolytic split in front of amino acid residue 285 of mature hIF.
(15) FIG. 11 shows the expression of transcobalamin in transgenic plants. Polyclonal antibodies against human transcobalamin (TC) and alkaline phosphatase-conjugated immunoglobulins were used to visualize TC on a Western blot containing proteins separated by SDS-PAGE. Transgenic Arabidopsis thaliana plants (no 9-11) expressing recombinant human TC were harvested and homogenized in 0.2 M phosphate buffer before centrifugation and analysis of the supernatant.
(16) The blot shows a single protein with the expected molecular size of 45 kDa in each of the three plants
(17) FIG. 12 shows the ability of plant recombinant human transcobalamin (o) and plant recombinant human intrinsic factor (g) to bind cobalamin or cobinamid. Both recombinant proteins were extracted form transgenic Arabidopsis thaliana plants. Equal amounts of the respective proteins were added to a mixture containing a fixed amount of cobalt (57Co) labeled cobalamin mixed with increasing amounts of nonradioactive cobalamin or cobinamide (X axis). Free and bound ligand was separated and the amount of 57Co attached to the protein was measured in a gamma counter. The 57Co fraction bound relative to the amount of 57Co bound in the absence of unlabelled cobalamin or cobinamide was calculated (Y axis). The figure shows that recombinant human intrinsic factor and recombinant human transcobalamin both bind cobalamin but not cobinamid.
(18) FIG. 13 shows the binding of intrinsic factor to the human intestinal receptor protein cubilin. The binding between intrinsic factor and its receptor cubilin was analysed employing a BIAcore 2000 instrument (Biacore AB, Uppsala, Sweden). An increase in Resp. Diff. (Y axis) indicates binding between intrinsic factor and cubilin. At time approximately 100 a solution with intrinsic factor is added to the immobilized cubilin. Intrinsic factor was isolated from transgenic plants, human gastric juice and hog stomach. Preparations of vitamin B 12 saturated and vitamin B12 unsaturated intrinsic factor were used. At time approximately 600 intrinsic factor is removed from the solution and the dissociation between cubilin and intrinsic factor is followed. The results indicate that recombinant intrinsic factor-like native intrinsic factor-binds to cubilin only when saturated with vitamin B12 and that the binding characteristics for recombinant and native intrinsic factor are alike.
DETAILED DESCRIPTION OF THE INVENTION
Example 1
(19) As shown in FIG. 1, the extensin signal peptide encoding nucleotide sequence from Arabidopsis thaliana was fused to the nucleotides encoding mature human intrinsic factor (FIG. 2). This construct was inserted in the plant transformation vector CRC179.
(20) Construction of the CRC-179 Vector
(21) The vector pPZP 211 (Hajdukiewicz, P; Svab, Z; & Maliga, P. 1994 Plant Mol. Biol. 25, 989-994) was digested with EcoRI and KpnI and a pAnos sequence was released from pGPTV KAN (Becker, D; Kemper, E; Schell, J & Masterson, R. 1992 Plant Mol Biol 20, 1195-1197) by the same set of enzymes and cloned into the pPZP 211 vector. The resulting vector was digested with PstI and KpnI and blunt-ended. A blunt-ended EcoRI and HindIII fragment containing the 35S CaMV promoter from the vector described in Jefferson, R A; Kavanagh, T A & Bevan, M W. 1987 EMBO J. 6, 3901-3907 was cloned into the blunt-ended vector. This vector was named CRC-179.
(22) The bacteria Agrobacterium tumefaciens was transformed with this recombinant vector.
(23) Culture of Agrobacterium tumefaciens
(24) The Agrobacterium tumefaciens strain used was GV3101 (pMP90) (Koncz and Schell, 1986) carrying the binary plasmid CRC-179 with an insert encoding a CBP cloned into the XbaI-XmaI sites. The insert sequences are shown in the FIGS. 1,3, 4, & 5. The bacteria were grown to stationary phase in 200 ml liquid culture at 28-30 C, 250 rpm in sterilized LB media (10 g tryptone, 5 g yeast extract, 5 g NaCl per liter water) carrying added rifampicin (100 mg/ml), streptomycin (100 mg/ml) and gentamycin (50 mg/ml). Cultures were started from a 1:200 dilution of a smaller overnight culture and grown for 16-18 hours. Bacteria cells were harvested by centrifugation for 10 min at 5500 g at room temperature and then resuspended in 400 ml inoculation medium (10 mM MgCl2, 5% w/v sucrose and 0.05% v/v Silwet L-77 (Lehle Seeds, Round Rock, Tex., USA)).
(25) The recombinant A. tumefaciens bacteria were used to transform Arabidopsis thaliana plants.
(26) Transformation of Arabidopsis Plants
(27) The Arabidopsis plants were transformed by the floral dip method (Clough and Bent, 1998).
(28) Plant Growth
(29) Arabidopsis plants (ecotype Col-0) were grown to flowering stage in growth chamber, 20 C day/18 C night with LiCl lighting for 18 h/day, humidity 70%. Between 20 and 25 plants were planted per 64 cm2 pot in moistened soil mixture consisting of: 40 kg soil orange and 40 kg soil green (Stenrgel Mosebrug A/S Kjellerup, DK), 25 liter 4-8 mm Fibroklinker (Optiroc, Randers, DK), 12 liter Vermiculite (Skamol, DK) and 300 g Osmocote plus NPK 15-5-11, 3-4 months (Scott's, UK).
(30) To obtain more floral buds per plant, inflorescences were clipped after most plants had formed primary bolts, relieving apical dominance and encouraging synchronized emergence of multiple secondary bolts. Plants were dipped when most secondary inflorescences were about 7-13 cm tall (7-9 days after clipping).
(31) The transgenic Agrobacterium suspension was added to a 400 ml beaker and plants were inverted into the suspension such that all above-ground tissues minus the rosette were submerged. The plants were removed after 10-15 sec of gentle agitation and placed in horizontal position in a sealed plastic bag for 24 hours at room temperature. After 24 hours the plants were moved to the growth camber and the plastic bags were removed. Plants were grown 3-4 weeks until siliques were brown and dry. Seeds were harvested and allowed to dry at room temperature for 7 days.
(32) Selection of Transformants
(33) Seed were surface sterilized by a treatment with 0.5% sodium hypochlorite containing 0.05% v/v Tween 20 for 7 min, then with 70% ethanol for 2 min, followed by three rinses with sterile water.
(34) To select for transformed plants the sterilized seeds were plated on kanamycin selection plates at a density of approximately 2000 seeds per 144 cm2 and grown for 8-10 days at 21 C under light for 16 hours per day. Selection plates contained 1×MS medium (Duchefa, Haarlem, NL #M 0222), 1% w/v sucrose, 0.9% w/v agar noble (Difco, Detroit, USA), 50 mg/ml kanamycin, pH 5.7. After selection the transformed plants were transferred to growth chambers (see Plant growth).
(35) Seeds from these infected plants were planted and recombinant plants were identified by western-blotting analysis. Seeds from recombinant plants were used to grow new recombinant plants called IF-plants.
(36) One kilogram of three week old IF-plants was homogenization with 2 liters of phosphate buffer and clarified by centrifugation. This extract contained 100 mg recombinant human IF with CBC. This IF had specificity for cobalamin binding whereas the analog cobinamid was not bound by the IF as tested by the method described by (15). FIG. 7 shows that the rhIF from plants and native human gastric IF have the same specificity for binding Cbl, but not cobinamid.
(37) The transgenic plants were analyzed as described by (16) and shown to contain no cobalamin. This shows that the plant rhIF was at the apo-form. Analysis of the protein from these transgenic plants showed that they express a protein of approximately 50 kDa as recognized on a Western-blot with antibodies against human IF (see FIG. 8,10). Amino acid sequencing of the N-terminal region of this 50 kDa protein showed the same sequence as that present in mature natural IF (FIG. 2). Therefore, the post-translational cleavage of the extensin signal peptide from the fusion protein results in the secretion of a recombinant IF with the correct N-terminus. The size of approximately 50 kDa indicates that the protein was full-length. A mature full-length IF is 399 amino acids with a calculated molecular weight of 43412 Da. The recombinant IF contains some carbohydrates as shown by PAS-staining of blotted recombinant protein separated by SDS-PAGE (see FIG. 9). Deglycosylation of rhiF from plants results in a protein with an approximately similar observed and calculated molecular weight (FIG. 8). The molecular size of natural human IF was approximately 5 kDa larger than the recombinant plant IF as estimated from the western-blot. Some difference in the carbohydrate composition is expected between natural human IF and recombinant plant IF since carbohydrate composition is tissue specific and to some extent unique to each individual. Differences in the molecular weight between natural and recombinant IF may therefore be a result of different carbohydrate composition. This was shown by removal of the carbohydrates from human gastric IF, human IF expressed in yeast and human IF expressed in plants. The deglycosylated form of these three IF proteins to have the same molecular weight (see FIG. 8).
(38) On a Western-blot containing purified rhiF from transgenic Arabidopsis thaliana, two minor bands of approximately 30 and 20 kDa were observed in addition to the 50 kDa band with mature rhiF (FIG. 10). N-terminal sequencing of these two bands showed, that the 30 kDa band contained the N-terminal sequence of mature human IF. The 20 kDa band contained an N-terminus located at glycine-285 of mature IF. The calculated molecular weight of the fragment containing amino acid residues 1-284 was 30612 Da and the fragment containing amino acid residues 285-399 had a calculated molecular weight of 12817 Da. The inconsistency between the observed 20 kDa and the calculated 12817 Da most likely was a result of glycosylation of one or more of the four potential N-linked glycosylation sites of the fragment containing amino acid residue 285-399. The PAS-staining also recognized the 20 kDa band showing that some glycosylation was present on the fragment. No PAS-staining was observed for the 30 kDa fragment although one potential N-linked glycosylation site was present on this fragment. This is consistent with that no difference was found between the observed and calculated molecular weight for the fragment containing amino acid residues 1-284.
(39) FIG. 13 shows that rHIF from plants in complex with vitamin B12 binds to the human intestinal receptor protein cubilin. The apo-form of rHIF does not bind to the cubulin receptor. For comparison binding of human gastric IF and hog IF were also shown to bind to the cubulin receptor only when in holo-form. These results show, that recombinant human intrinsic factor from plants behaves as native gastric intrinsic factor with respect to receptor binding.
Example 2
(40) Arabidopsis thaliana was transformed with a vector construct which contained a nucleotide sequence encoding the signal peptide from the Phaseolus vulgaris chitinase CH5B fused to the nucleotide sequence encoding mature human intrinsic factor (FIG. 3). This construct was used to generate transgenic Arabidopsis thaliana plants. These plants were shown to contain CBC at the same level as most of the extensin-IF plants showing that the choice of signal peptide for intrinsic factor expression is not restricted to one sequence.
Example 3
(41) Arabidopsis thaliana was transformed with a vector construct which contained a nucleotide sequence encoding the signal peptide from the Nicotiana tabacum glucan beta-1, 3-glucanase fused to the nucleotide sequence for mature human intrinsic factor (FIG. 4). This construct was used to generate transgenic Arabidopsis thaliana plants. These plants were shown to contain CBC at the same level as most of the extensin-IF plants showing that the choice of signal peptide for intrinsic factor expression is not restricted to one sequence.
Conclusions from IF-Plants in Examples 1-3
(42) As far as we have tested the recombinant human IF from plants it behaves as natural human gastric IF concerning its mature N-terminus, recognition by anti-IF antibodies, binding of cobalamin, lack of cobinamide binding, binding to the intestinal receptor, and presence of carbohydrates. In contrast to gastric juice where IF is present together with another CBP, haptocorrin, and to some extent cobalamin from the food, transgenic IF plants contain no cobalamin or other CBP than IF. The glycosylation of IF from transgenic plants was different from human gastric IF. Another difference between IF from human gastric mucosa and IF from transgenic plants is that IF from plants is at the apo-form whereas IF from human beings is a mixture of the apo- and holo-form.
Example 4. TC-Plants
(43) Extracts from one kilogram of transgenic Arabidopsis thaliana (TC-plants) transformed with an extensin-transcobalamin construct (FIG. 5) contained 20 mg of recombinant human TC with CBC. Western blot analysis showed a single band of approximately 45 kDa which reacted with antibodies against human TC (FIG. 11). The calculated molecular weight of mature TC (FIG. 6) with 409 amino acid residues is 45536 Da, showing that the observed and calculated molecular weights are similar. These results show that a plant expression system is able to produce recombinant human TC of the expected size and with CBC.
(44) The transgenic plants were analyzed as described by (16) and shown to contain no cobalamin. This shows that the recombinant human TC was at the apo-form. As for rhIF, rhTC obtained from plants binds to Cbl but not cobinamid (FIG. 12). N-terminal amino acid sequencing showed that the extensin signal peptide was removed from the secreted rhTC generating the normal mature N-terminal found in native human transcobalamin.
Example 5. Purification of the Recombinant Intrinsic Factor from Plants
(45) 1 kg of the crude plant material was chopped and homogenized in 2 L of 0.2 M Sodium Phosphate buffer, pH 7.5. The homogenate was centrifuged at 4000 rpm for 10 min and filtered through Watman paper on a Buchner funnel. The filtrate can be stored frozen at that stage. Intrinsic factor was adsorbed from the solution on an affinity matrix according to a previously described method (NexX, E., 1975 Biochim. Biophys. Acta 379, 189-192). After elution from the column intrinsic factor (saturated with cobalamin=holo-form) was subjected to gel filtration, dialyzed against water and lyophilized. Preparation of cobalamin unsaturated intrinsic factor (=apo-form) required an additional step: dialysis against 5 M guanidine HCl for two days followed by dialysis against 0.2 M Sodium Phosphate buffer, pH 7.5.
REFERENCES
(46) 1. Allen, R. H. (1975) Prog. Hematol. 9, 57-84. 2. Rothenberg, S. P., and Quadros, E. V. (1995) Bailliere S Clin. Haematol. 8, 499-514. 3. Nicolas, J. P., and Gueant, J. L. (1995) Bailliere S Clin. Haematol. 8, 515-531. 4. Next, E. (1998) in Vitamin B12 and B12-Proteins (Krautler, B., Angoni, D., and Golding, B. T., eds) pp. 461-475, Wiley-VCH, Weinheim, Germany. 5. Hall, C. A., and Begley, J. A. (1977) Am. J. Hum. Genet. 29, 619-626. 6. Katz, M., Mehlman, C. S., and Allen, R. H. (1974) J. Clin. Invest. 53, 1274-1283. 7. Li, N., Rosenblatt, D. S., Kamen, B. A., Seetharam, S., and Seetharam, B. (1994) Hum, Mol, Genet. 3, 1835-1840. 8. VanKapel, J., Loef, B. G., Lindemans, J., and Abels, J. (1981) Biochim, Biophys. Acta 676, 307-313. 9. Quadros, E. V., Rothenberg, S. P., Pan, Y. C., and Stein, S. (1986) J. Biol. Chem. 261, 15455-15460. 10. Gordon, M., Chokshi, H., and Alpers, D. H. (1992) Biochim. Biophys. Acta 1132, 276-283. 11. Quadros, E. V., Sai, P., and Rothenberg, S. P. (1993) Blood 81, 1239-1245. 12. Fedosov, S. N., Berglund, L., Next, E., and Petersen, T. E. (1999) J. Biol. Chem. 274, 26015-26020. 13. Wen, J., Kinnear, M. B., Richardson, M. A., Willetts, N. S., Russell-Jones, G. J., Gordon, M. M., and Alpers, D. H. (2000) Biochim. Biophys. Acta 1490, 43-53. 14. Next, E. and Andersen, J. (1977) Scand. J. Clin. Lab. Invest. 37, 723-728. 15. Stupperich, E. And Next, E. (1991) Eur. J. Biochem. 199, 299-303. 16. Next, E. And Gimsing, P. (1981) Scand. J. Clin. Lab. Invest. 41, 465-468. 17. Clough, S. J., and Bent, A. F. (1998) Plant J. 16, 735-743. 18. Koncz, C., and Schell, J. (1986) Mol. Gen. Genet. 204, 383-396. 19. Nexo et al. 2000, Clinical Chemistry 46 (10): 1643-1649 20 Next, 1975, Biochim. Biophys. Acta, 379: 189-192.
