Recombinant factor VIII binding peptides

Abstract

Peptides that have domains that bind to recombinant factor VIII (rFVIII) are disclosed. A method of rFVIII binding assay using the peptides deduced from a combinatorial library in a filtration plate process is described. A method of using peptides having these available binding domains in an affinity chromatography process to purify factor VIII is also taught.

Description

BACKGROUND OF THE INVENTION

1. Field

This invention is concerned generally with identifying protein-ligand interactions, and specifically with peptide ligands which bind rFVIII and which may be used in a method for the affinity purification of rFVIII.

2. Prior Art

Human Factor VIII, (antihemophilic factor) is a human plasma protein consisting of 2 polypeptides (light chain molecular weight of 80,000 and heavy chain molecular weight variable from 90,000 to 220,000). It is an essential cofactor in the coagulation pathway; required for the conversion of Factor X into its active form (Factor Xa). Factor VIII circulates in plasma as a non-covalent complex with von Willebrand Factor. Blood concentrations of Factor VIII below 20% of normal cause a bleeding disorder designated hemophilia A. Factor VIII blood levels less than 1% of normal result in a severe bleeding disorder, with spontaneous joint bleeding being the most common symptom. Recombinant DNA technology has allowed construction of plasmids that direct the expression of fusion products of Factor VIII protein in transfected mammalian cells. Factor VIII can be isolated from either a plasma derived source (cryoprecipitate) or from a genetically engineered recombinant source. The term Factor VIII is not meant to be a limitation but refers to a functional protein for treating bleeding disorders.

Several methods have been described for purification of Factor VIII from plasma sources (1,2,3). Several purification schemes utilize antibody affinity columns (4,5). To date, the most successful purifications of Factor VIII from plasma or from recombinant sources has been accomplished by using murine monoclonal antibodies specific to either Factor VIII or von Willibrand Factor.

Although monoclonal antibodies have been used successfully to obtain a relatively pure Factor VIII preparation, monoclonal antibodies can be present in the Factor VIII effluent because of leaching from the support matrix. This raises the possibility of antigenicity when the final preparation is introduced into animal systems, since murine monoclonal antibodies have been shown to be antigenic. A second disadvantage of the use of monoclonal antibodies is the requirement of cell culture facilities for producing the antibodies and the concomitant cost of purification and attachment onto a support matrix. Finally, the stability of the antibody binding site may not be amenable to the rigorous conditions necessary to sanitize the column.

Affinity chromatography is one of the most efficient techniques for purifying a protein from a complex mixture. With potential advantages including high stability, efficiency, selectivity, and low price, peptides have been investigated as affinity ligands in the pharmaceutical industry. A recent approach for identifying such ligands is to screen peptides from combinatorial peptide libraries (6,7,8,9).

Using the ‘divide-couple-recombine’ approach (10,11,12), millions of unique peptides of a defined length may be synthesized on resin beads. Each bead contains a unique peptide sequence. These library beads and their corresponding peptide sequences are then exposed to a target protein. Among these millions of peptide sequences, the target protein may bind to several unique bead-sequences. Those beads and their corresponding sequences must be detected, isolated, and identified. Several detection systems, including calorimetric two-step methods (7,12,13) as well as direct fluorescence detection methods (14,15,16) have been used.

Peptides disclosed in the U.S. patent application Ser. No. 08/595,718, incorporated herein by reference, also were found to bind Factor VIII. See also Necina et al. (23)

SUMMARY OF THE INVENTION

We have now discovered a group of peptides characterized by their ability to bind rFVIII. The sequences of the more preferred peptides having available rFVIII binding domains are Asn-Ala-Ile-Phe-Gln-Trp (SEQ ID NO:11), Ala-Phe-Val-Arg-Ser-Leu (SEQ ID NO:10), Gln-Arg-Leu-Ile-Gln-Phe (SEQ ID NO:12), Phe-Arg-Pro-His-Trp-Ala (SEQ ID NO:4), Arg-Pro-Arg-Trp (SEQ ID NO:8), and others of those presented in Table 1. A

TABLE 1
rFVIII Binding Peptide Sequences
Sequences discovered from screening of hexameric peptide library using
14 C-rFVIII with confirmed binding of rFVIII.
Seq ID
1	Lys	Pro	Asn	Pro	Leu	Ala
2	Arg	Asn	Pro	Pro	Asn	Asn
3	Tyr	Val	Gln	Gly	Leu	Trp
4	Phe	Arg	Pro	His	Trp	Ala
5	Leu	Asn	Trp	Lys	Tyr	Gly
6	His	Tyr	Trp	Phe	Tyr	Lys
7	Ile	Arg	Phe	Tyr	Ser	Glu
8	Arg	Pro	Arg	Trp
9	Phe	Ala	Leu	Pro	Gly	Arg
10	Ala	Phe	Val	Arg	Ser	Leu
11	Asn	Ala	Ile	Phe	Gln	Trp
12	Gln	Arg	Leu	Ile	Gln	Phe
13	Lys	Ala	Gln	Glu	Thr	Trp
14	Glu	Pro	Arg	Val	Ile	Gly
15	Val	Tyr	Gly	Val	Gly	Gly
16	Trp	Arg	Arg	His	Arg	Tyr
17	Phe	Tyr	Arg	Phe	Trp	Asn
18	Trp	Leu	Trp	Ser	His	Asn
19	Phe	His	Phe	Gly	Leu	Gln
20	Trp	His	His	His	Arg	Gly
21	His	Phe	Gln	Ile	Phe	Gly
22	Phe	Val	Phe	Leu	Val	Arg

method of rFVIII binding assay using the peptides deduced from a combinatorial library in a filtration plate process is described. We also describe a method of using the peptides in an affinity chromatography process to purify rFVIII. The method comprises passing a rFVIII containing solution over a substrate which has bound thereupon peptides disclosed herein, and then eluting the rFVIII.

As used herein, an available rFVIII binding domain means a peptide sequence that is sterically available to bind with rFVIII in the surrounding solution and which adopts a conformation that ligates rFVIII with moderate to strong avidity under controlled conditions of pH, ionic strength, and solvent composition. The affinity of binding may be increased or decreased by altering the amino acids adjacent to the above listed sequences. The avidity may be modified further by altering the above mentioned conditions of solvent composition and temperature.

The peptides were isolated and identified using a modified version of a radiological screening technique (17,18,19). We also describe a method of rFVIII binding assay using the identified peptides in a filtration plate format, wherein the method comprises incubating a 14 C-rFVIII containing solution over a substrate that has bound thereupon peptides disclosed herein, and then washing the substrates with appropriate buffer conditions. Finally, a method of rFVIII binding assay using the identified peptides in a column format is also described, wherein the method comprises passing a rFVIII containing solution over a substrate that has bound thereupon peptides disclosed herein, and then eluting the rFVIII with appropriate buffer conditions. Ultimately, this method leads to an affinity chromatography process to purify rFVIII by adopting appropriate elution conditions.

SPECIFIC EMBODIMENTS

Materials

Purified rFVIII (at 500 IU/vial and DOGENATE rFVIII Kogenate at 591 IU/vial of rFVIII in 5-10 mg/mL hSA), ultrafiltered tissue culture fluid containing expressed FVIII (UF-TCF), and tissue culture fluid containing expressed FVIII which had been contacted with a DEAE-Sepharose column resulting in a DEAE concentrate (DEAE eluate) were manufactured by the Bayer Corporation (Berkeley, Calif.). Human serum albumin (hSA) and plasma protein feed (PPF) were also from the Bayer Corporation (Clayton, N.C.). SUPERBLOCK® blocking buffer, Blocker® blot to nonfat powdered milk solution (5% w/v) and BLOCKER casein in Tris buffered saline (TBS) were from Pierce (Rockford, Ill.). Fmoc amino acids were from Novabiochem (San Diego, Calif.). Radiolabeled 14 C-formaldehyde was from NEN Life Science Products (Boston, Mass.). A-arose was from BioRad (Hercules, Calif.). All other chemicals were reagent grade or better.

General Methods

Peptides and a 6-mer combinatorial library were synthesized on TOYOPEARL AF amino 650 M (TosoHaas, Montgomeryville, Pa.) using standard Fmoc chemistry as described (7). Peptides were synthesized robotically with a Gilson AMS 422 (Middleton, Wis.). Peptide densities achieved with the above scheme were typically at 0.1 mmole/g resin. To control the peptide density, a mixture of 1:10 of Fmoc-L-Alanine to tBoc-L-Alanine was used as described (22).

Peptides were re-synthesized directly onto the TOYOPEARL AF amino 650 M resin, at a final substitution density of 0.1 mmole/g resin.

Analvtical Methods

Immunoassay on rFVIII using a sandwich capture ELISA for the light chain and activity assay using a chromogenic substrate (COATEST Factor VIII kit from diaPharma, Franklin, Ohio) were performed on a BIOMEK® automated liquid handling system from Beckman Instruments (Brea, Calif.). The monoclonal antibody for rFVIII light chain (known as C7F7), the ELISA assay standard (lot A70K019), and the control for the activity assay (lot A70A001 @ 140 IU/mL) were from the Bayer Corporation (Berkeley, Calif.). The standard for the activity assay was at 11.2 IU/mL (Bayer, Clayton, N.C.), calibrated against the MEGA 1 standard, which is a working standard used by American manufacturers for measuring factor VIII activity (21).

Peptide densities were determined by quantitative amino acid analysis performed at Commonwealth Biotechnologies, Inc., Richmond, Va., using a Hewlett Packard AMINO QUANT Chemistry system.

Discovery of Binding Peptides To rFVIII

An assay similar to the one described by Mondorf et al. (19) was used to deduce peptides that bind rFVIII. A 6-mer combinatorial peptide library was synthesized directly onto a TosoHaas chromatography resin, TOYOPEARL AF amino 650M, at a final substitution density of 0.1 mmole/g resin. The library was synthesized using Fmoc chemistry with 18 of the 20 natural amino acids (excepting cysteine and methionine).

Sequence identification utilized a radiological detection approach. Radiolabeling of rFVIII was performed at 5° C. by reductive methylation utilizing sodium cyanoborohydride and 14 C-formaldehyde as described (20). The labeling resulted in radioactivity yields ranging from 2 to 57×10 14 dpm/mole rFVIII.

Aliquots of library beads were placed in a reaction column (BioRad, Hercules, Calif.). To prevent nonspecific interactions between peptide and 14 C-rFVIII, the library beads were first incubated with a blocking solution containing SUPERBLOCK blocking buffer, BLOCKER blot to nonfat powdered milk solution (5% w/v) or VLOCKER casein in TBS, followed by PPF in equilibration buffer (20 mM imidazole, 10 mM CaCl2, 0.105 M NaCl, pH 6.9) for 1 hour each, on a rotating plate at ambient temperature. Radiolabeled 14 C-rFVIII at a final concentration of 60 nM in a mixture of PPF, blocking solution and UF-TCF or DEAE eluate was then added to the preblocked library beads and incubated for 2 hours on the rotating plate at 5° C.

After completion of blocking and binding, the beads were rinsed in the reaction column with equilibration buffer, until radioactivity count went down to background level. After rinsing, the beads were transferred to a container by adding low-melt agarose solution (1% by weight), and the slurry poured onto 8×″8″ sheets of GENBOND film (FMC, Rockland, Me.). To minimize the loss of beads, the reaction column and the transferring container were rinsed a second time with low melt agarose gel, and this gel was also plated out. The agarose gels were air-dried and exposed to photographic films (HYPERFILM βmax autoradiography film, Amersham Life Science, Arlington Heights, Ill.). After 5 days exposure at room temperature, the films were developed. The background level was extremely low. Re-exposure of the gels for additional 15 days resulted in confirmation of the signals. Careful alignment of the films and the agarose gels allowed for identification of positive beads which were isolated. From the signals, beads were excised, sonicated under heating to free the beads from agarose residues, and sequenced by Edman degradation. The sequence analysis was performed at the Texas A&M University, College Station, Tex., using a Hewlett Packard G1005A Protein Sequencer.

Table 2 lists the screening experiments performed using the above described protocol with variations in the type of blocking agent used (SUPERBLOCK blocking buffer, BLOCKER blot to nonfat powdered milk solution (5% w/v) or BLOCKER casein in TBS), concentration of PPF in the further blocking step, type of feed stream used (UF-TCF or DEAE eluate), amount of library beads used in each experiment, operating temperature, additional washes with 2 M NaCl, pH 6, pH 8, and 1 M CaCl 2 , and the number of beads selected from each experiment. A total of 62 beads were isolated from the experiments listed in Table 2. Among these, 30 were sequenced by Edman degradation, resynthesized on TOYOPEARL AF amino 650 M resin, and their capacity for binding of rFVIII studied in the confirmatory binding assays.

Twenty-two of the peptide sequences are listed in Table 1. Several consistencies are noted in these sequences. There are three types of residues found in all the sequences: 1) positively-charged residues (Arg, His, Lys); 2) aromatic residues (Tyr, Phe, Trp); and, 3) hydrophobic residues (Ile, Leu, Val, Pro).

Binding Confirmation

Confirmatory binding assays were performed in a filtration plate format using individual peptide sequences synthesized directly onto the TOYOPEARL AF amino 650 M resin, at a final substitution density of 0.1 mmole/g resin. A slurry of each individual peptide sequence was prepared using deionized water. A 100 μL volume of beads (corresponding to 5 mg) was loaded into each well of the filtration plate (MULTISCREEN Assay System, Millipore, Bedford, Mass.), in triplicate for each peptide sequence. Beads were washed with 20% ethanol and incubated overnight with equilibration buffer (20 mM imidazole, 10 mM

TABLE 2
Primary screening experiments leading to discovery of the binding
peptides to rFVIII.
			Peptide
		Feed Stream	Library		N#
	Blocker	(FS)	used	Comments	beads
Exp	SUPER-	DEAE eluate,	6 mg	T = amb	3
18a	BLOCKER	SB, 14 C-rFVII		(for blocker)
	(SB)			T = 5 C
Exp	blotto,	UF-TCF,	100 mg	(for FS)	6
21	casein,	casein,
Exp	1:5 PPF	14 C-rFVIII	100 mg		9
22
Exp	casein,	DEAE eluate,	100 mg		1
23	1:10 PPF	SB, 14 C-rFVII
Exp			100 mg	T = amb	7
24
Exp		UF-TCF,	200 mg	extremes	32
25		14 C-rFVIII		added:	not se-
				2M NaCl, pH6,	quenced
Exp		DEAE eluate,	200 mg	pH8, 1M CaCl 2	4
26		14 C-rFVIII		T = 5 C

CaCl 2 , 0.105 M NaCl, pH 6.9). A 100 μL aliquot of 14 C-rFVIII diluted in neat PPF to a final concentration of 20-30 nM was loaded into each well and incubated for 1 hour at 5° C. Filtrate (unbound 14 C-rFVIII) was collected into a microtiter plate and transferred to liquid scintillation vials for dpm counting. Beads were washed sequentially with: 100 μL of equilibration buffer for 15 minutes, 200 μL of 0.5 M CaCl 2 (in equilibration buffer), 200 μL of 1.0 M CaCl 2 (in equilibration buffer), and 200 uL of 2% by volume glacial acetic acid in water, for 1 hour each at 5° C. Scintillation counting on the unbound 14 C-rFVIII (plus equilibration buffer wash), salt and acid washes, as well as on the beads was then performed and the overall mass balance calculated.

Performance of the rFVIII Binding Peptides

Additional confirmatory binding assay and performance of the discovered binding peptides in terms of total rFVIII and activity recovery was performed in a column chromatographic format on a microbore HPLC. The resins were packed into 0.4×5 cm columns (Thomson, Springfield, Va.) and tested on the MICHROM HPLC system (Michrom BioResources, Auburn, Calif.). The elution program for the trials utilized the following method. The column was pre-equilibrated with equilibration buffer (as above). A 1 mL aliquot of 1:10 PPF followed by 1 mL of rFVIII (10,000 ng) in 1:10 PPF were injected and allowed sufficient time to flow through the column using equilibration buffer. The column was then washed with step elutions of 0.5 M and 1.0 M CaCl 2 in equilibration buffer. The column was exposed to equilibration buffer. Remaining protein was eluted with 2% by volume glacial acetic acid in water. Columns were kept at 5° C. during the chromatographic runs. For each 1 mL injection (equilibration buffer, 1:10 PPF and 10,000 ng rFVIII in 1:10 PPF), the chromatographic cycle was run for 45 minutes (injection, flow through, 0.5 M CaCl 2 , 1.0 M CaCl 2 , equilibration buffer, 2% acetic acid, equilibration buffer).

A Gilson FC204 fraction collector at the HPLC outlet collected individual eluates for further analysis. Each collection tube had 100 μL hSA added to it to stabilize the rFVIII. Fractions collected from the 2% acetic acid eluates were neutralized with 2 M Tris, pH 10.5 to bring the pH to neutrality. The fractions were analyzed for total protein by the integration of the absorbance at 280 nm, for rFVIII by immunoassay (sandwich capture ELISA), and for rFVIII activity by a chromogenic substrate assay.

Results

Tables 3 and 4 present the results from the confirmatory binding assays using the plate format. Beads presenting the appropriate peptide sequence on the surface were placed in the filtration plate wells and incubated sequentially with equilibration buffer, 14 C-rFVIII in PPF, equilibration buffer, 0.5 M CaCl 2 wash, 1.0 M CaCl 2 wash and 2% acetic acid wash. After each incubation, filtrate was collected in microtiter plates, and dpm counted on each

TABLE 3
Confirmation of binding for peptide sequences 1 to 8 and
16 to 18 using 14 C-rFVIII in the plate format binding assay.
Sequences 1 to 8 resulted in 14 C-rFVIII eluted in the 0.5M CaCl 2 wash
while sequences 16 to 18 in the 2% acetic acid wash. Binding results
from a non binder sequence IGF are included for comparison, as well as
the results from the control ( 14 C-rFVIII loaded into wells without beads).
100 μL of 14 C-rFVIII at 20 nM and 3.0 × 10 15 DPM/mole were
loaded into each well containing 5 mg beads.
				2%
		0.5M		acetic
Seq.		CaCl2	%	acid	%	beads	%	Total
ID	Sequence	dpm	total	dpm	total	dpm	total	dpm
1	KPNPLA	2694	85	368	12	99	3	3162
2	RNPPNN	2366	86	322	12	79	3	2767
3	YVQGLW	1502	74	379	19	140	7	2020
4	FRPHWA	2905	78	631	17	209	6	3744
5	LNWKYG	3519	86	430	10	143	3	4091
6	HYWFYK	3236	83	458	12	187	5	3881
7	IRFYSE	3184	85	415	11	165	4	3764
8	RPRW	3543	86	377	9	178	4	4098
16	WRRHRY	608	21	1302	45	986	34	2897
17	FYRFWN	498	17	2156	72	355	12	3009
18	WLWSHN	651	25	1752	67	205	8	2608
	IGF	219	80	41	15	14	5	274
con-	no beads	98	86	16	14		0	114
trol

filtrate sample. Sequences 1 to 8 (Table 3) and 9 to 15 (Table 4) resulted in high percentage 14 C-rFVIII bound and eluted in 0.5 M CaCl 2 , ranging from 48 to 95% eluted by the salt wash. Sequences 16 to 18 (Table 3) and 19 to 21 (Table 4) resulted in most of the bound rFVIII (from 45 to 72%) eluted by the 2% acetic acid wash. Finally, sequence 22 resulted in tight binding, with 65% of the bound 14 C-rFVIII still on the peptide bead after salt and acid washes. Results for a non-binding peptide sequence (IGF), as well as for the control with no beads in the inicrotiter plate wells, are also shown for comparison.

TABLE 4
Confirmation of binding for peptide sequences 9 to 15 and
19 to 22 using 14 C-rFVIII in the plate format binding assay.
Sequences 9 to 15 resulted in 14 C-rFVIII eluted in the 0.5M CaCl 2 wash
while sequences 19 to 21 in the 2% acetic acid wash. Sequence 22 is a
tight binder, not eluting the bound 14 C-rFVIII by the salt or acid washes
performed. The control ( 14 C-rFVIII loaded into wells without beads) is
also presented. 100 μL of 14 C-rFVIII at 30 nM and 5.0 × 10 15
dpm/mole were loaded into each well containing 5 mg beads.
				2%
		0.5M		acetic
Seq.		CaCl2	%	acid	%	beads	%	Total
ID	Sequence	dpm	total	dpm	total	dpm	total	dpm
9	FALPGR	3414	93	176	5	68	2	3658
10	AFVRSL	3491	93	188	5	85	2	3764
11	NAIFQW	1049	48	518	23	641	29	2208
12	QRLIQF	3191	90	271	8	95	3	3557
13	KAQETW	3884	95	135	3	77	2	4096
14	EPRVIG	1994	91	130	6	60	3	2184
15	VYGVGG	2162	93	115	5	53	2	2330
19	FHFGLQ	457	15	2148	72	382	13	2987
20	WHHHRG	485	12	2099	54	1300	33	3884
21	HFQIFG	722	21	2208	63	554	16	3484
22	FVFLVR	675	16	783	19	2658	65	4116
con-	no beads	98	75	33	25		0	131
trol

In Table 3, the 14 C-rFVIII was loaded at 20 nM (in PPF) and the radiolabeling yield was at 3.0×10 15 dpm/mole. In Table 4, the 14 C-rFVIII was loaded at 30 nM (in PPF) and the radiolabeling yield was at 5.0×10 15 dpm/mole.

Tables 5 and 6 present the results from the additional confirmatory binding assays using the column format. The chromatography method was run by injecting the samples at a flow rate of 865 μL/min (413 cm/hr). In order to provide a 10 minute residence time of sample in the column the flow was decreased to 17 μL/min (10 cm/hr) upon injection.

TABLE 5
Confirmation of binding for peptide sequences 1 to 8
and 16 to 18 in the column format binding assay.
Sequences 1 to 8 resulted in rFVIII eluted in the 0.5M
CaCl 2 wash while sequences 16 to 18 in the 2% acetic acid
wash. Binding results from a non binder sequence IGF are included
for comparison, as well as the results from the control
(rFVIII injected in the HPLC without any column attached).
						2%
		Flow	%	0.5M	%	acetic	%
Seq		through	to-	CaCl2	to-	acid	to-	Total
ID	Sequence	ng	tal	ng	tal	ng	tal	ng
1	KPNPLA	583	10	4961	88	61	1	5651
2	RNPPNN	447	7	5745	92	31	0	6245
3	YVQGLW	n.a.		n.a.		n.a.		n.a.
4	FRPHWA	225	4	5510	90	160	3	6123
5	LNWKYG	25	0	5530	97	125	2	5680
6	HYWFYK	105	2	5658	90	160	3	6320
7	IRFYSE	228	4	5836	94	77	1	6225
8	RPRW	122	2	5449	97	10	0	5623
16	WRRHRY	56	2	338	9	2550	71	3615
17	FYRFWN	127	3	1622	37	2336	54	4348
18	WLWSHN	8	0	2362	56	760	18	4200
	IGF	4719	95	201	4	10	0	4962
con-	no column	5110	98	80	2	10	0	5212
trol

Following the incubation, the flow was set at 865 μL/min (413 cm/hr) for the step clutions of 0.5 M CaCl 2 , 1.0 M CaCl 2 , and 2% acetic acid. Results are expressed in terms of the total mass of rFVIII in each fraction, as determined by the sandwich capture ELISA assay for the rFVIII light chain. Sequences 1 to 8 and 18 (Table 5) and 9 to 15 (Table 6) resulted in high percentage of the protein bound and eluted in 0.5 M CaCl 2 , ranging from 32 to 98% eluted by the salt wash. Sequences 16 to 17 (Table 5) and 19 to 22 (Table 6) resulted in 54 to 100% of the bound rFVIII eluted by the 2% acetic acid wash. A control using a non-binding peptide sequence (IGF) is also shown for comparison—in this case, 95% of the rFVIII loaded to the column was collected in the flow through fraction. Additional controls with no

TABLE 6
Reconfirmation of binding for peptide sequences 9 to 15
and 19 to 22 in the column format binding assay.
Sequences 9 to 15 resulted in rFVIII eluted in the 0.5M
CaCl 2 wash while sequences 19 to 21 in the 2% acetic
acid wash. Results for the control (rFVIII injected in the
HPLC without column) are also presented for comparison.
						2%
		Flow	%	0.5M	%	acetic	%
Seq		through	to-	CaCl2	to-	acid	to-	Total
ID	Sequence	ng	tal	ng	tal	ng	tal	ng
9	FALPGR	0	0	7154	97	181	2	7403
10	AFVRSL	0	0	11023	98	135	1	11272
11	NAIFQW	639	7	7481	78	1069	11	9641
12	QRLIQF	0	0	9223	96	188	2	9579
13	KAQETW	641	7	8018	90	87	1	8863
14	EPRVIG	5828	66	2850	32	75	1	8851
15	VYGVGG	3572	42	4611	55	121	1	8427
19	FHFGLQ	0	0	1075	25	2569	60	4283
20	WHHHRG	0	0	36	1	4719	95	4966
21	HFQIFG	0	0	1239	22	3284	60	5510
22	FVFLVR	0	0	4	0	4500	100	4504
con-	no column	8109	95	334	4	63	1	8564
trol

colummns connected to the HPLC system were also run for comparison. Theoretically, 10,000 ng of rFVIII (in 1:10 PPF) were expected to be loaded in each column. In Table 5, the mass balance on the total amount of rFVIII in the no column control showed only about 52% of theroetical value actually loaded in the system. In Table 6, the mass balance on the control resulted in 86% of the theoretical value loaded in the system.

Performance of the rFVIII Binding Peptides

Tables 7 and 8 list the sequences for which activity recovery was determined by a chromogenic substrate assay (COASTEST kit, diaPharma, Franklin, Ohio). Chromatographc conditions are the same as described above for the results from Tables 5 and 6. It can be

TABLE 7
Activity recovery for peptide sequences 1, 2, 5, 6 and 7.
The sequences which resulted in rFVIII eluted in the 0.5M
CaCl 2 showed high activity recovery, ranging from 86 to
96%. Activity recovery for the control (rFVIII injected through
the HPLC without column) is also shown for comparison.
						2%
		Flow	%	0.5M	%	acetic	%
Seq		through	to-	CaCl2	to-	acid	to-	Total
ID	Sequence	IU	tal	IU	tal	IU	tal	IU
1	KPNPLA	0.03	0	32.8	90	0.58	2	36.5
2	RNPPNN	0.4	1	51.6	95	0.5	1	54.3
5	LNWKYG	0.0	0	36.3	96	1.2	3	37.8
6	HYWFYK	0.0	0	59.4	86	0.5	1	68.7
7	IRFYSE	0.0	0	40.6	96	0.4	1	42.5
con-	no column	54.8	96	1.9	3	0.0	0	56.9
trol

TABLE 8
Activity recovery for peptide sequences 9 to 15 and 19 to 22.
The sequences which resulted in rFVIII eluted in the 0.5M
CaCl 2 showed high activity recovery, ranging from 25 to
98%. Activity recovery for the sequences that elute rFVIII in the
acid fraction is very low. Results for the control (rFVIII injected
through the HPLC without column) are also shown for comparison.
						2%
		Flow	%	0.5M	%	acetic	%
Seq		through	to-	CaCl2	to-	acid	to-	Total
ID	Sequence	IU	tal	IU	tal	IU	tal	IU
9	FALPGR	0.0	0	68.9	98	0.1	0	70.2
10	AFVRSL	0.0	0	53.1	98	0.1	0	54.3
11	NAIFQW	2.5	4	54.4	90	0.0	0	60.4
12	QRLIQF	0.0	0	45.1	97	0.08	0	46.6
13	KAQETW	0.8	2	48.3	96	0.09	0	50.3
14	EPRVIG	42.9	73	14.7	25	0.00	0	58.6
15	VYGVGG	17.3	38	27.2	60	0.08	0	45.4
19	FHFGLQ	0.5	2	9.9	51	1.4	7	19.5
20	WHHHRG	0.0	0	0.4	4	3.5	45	7.8
21	HFQIFG	0.0	0	11.4	53	4.1	19	21.2
22	FVFLVR	0.0	0	0.0	0	2.3	100	2.3
con-	no column	37.5	92	2.6	6	0.2	0	40.8
trol

observed that most of the rFVIII eluted in the salt fraction (0.5 M CaCl 2 ) showed activity recovery, while for the rFVIII eluted in acid, practically no activity is recovered. Sequences 1, 2, 5, 6, 7, 9, 10, 11, 12, and 13 resulted in high activity recoveries, ranging from 86 to 98%.

Table 9 presents a summary of the performance of the discovered peptide sequences that resulted in binding and elution of the rFVIII in the 0.5 M CaCl 2 fractions (sequences 1 to 15). Chromatographic conditions are the same as described above for the results from Tables 5 and 6. The rFVIII present in these fractions ranged from 32-98% of the total rFVIII

TABLE 9
Performance of the peptide sequences that elute rFVIII in the 0.5M CaCl 2.
Peak areas (x106) of the 0.5M CaCl 2 fractions correlated with the
percentage total rFVIII and rFVIII activity recovered. Results based on the
injection of 10,000 ng rFVIII (in 5 mg PPF) in equilibration
buffer at 5° C.
		ELISA		Activity
		results		results		Peak
Seq ID	Sequence	ng	% total	IU	% total	area
1	KPNPLA	4961	88	32.8	90	7.7
2	RNPPNN	5745	92	51.6	95	5.0
3	YVQGLW	n.a.		n.a.		n.a.
4	FRPHWA	5510	90	n.a.		81.9
5	LNWKYG	5530	97	36.3	96	3.6
6	HYWFYK	5658	90	59.4	86	3.4
7	IRFYSE	5836	94	40.6	96	4.2
8	RPRW	5449	97	n.a.		18.1
9	FALPGR	7154	97	68.9	98	4.9
10	AFVRSL	11023	98	53.1	98	22.8
11	NAIFQW	7481	78	54.4	90	0.4
12	QRLIQF	9223	96	45.1	97	13.7
13	KAQETW	8018	90	48.3	96	4.4
14	EPRVIG	2850	32	14.7	25	0.0
15	VYGVGG	4611	55	27.2	60	2.0

loaded, and the activity recovery ranged from 25 to 98%. The integrated peak areas (based on absorbance at 280 nm) for the injection of 10,000 ng rFVIII (in 1:10 PPF, with a total protein of approximately 5 mg/mL) at 5° C. are also presented. All the sequences eluted enriched rFVIII in the salt fraction.

CONCLUSION

Peptides sequences with varying degrees of specificity and affinity for factor VIII were identified from the screening of a combinatorial peptide library. A number of sequences including Asn Ala Ile Phe Gln Trp (SEQ ID NO:11) showed the ability to purify factor VIII. The affinity of these peptide sequences for factor VIII may be altered by deleting, substituting, or adding amino acids to the disclosed sequences. It is expected that the binding domains of the disclosed peptide sequences may be contained within other, longer peptides and still bind to factor VIII.

The above examples are intended to illustrate the invention, and it is thought variants will occur to those skilled in the art. Accordingly, it is intended that the scope of the invention should be limited only by the claims below.

REFERENCES

1. Tuddenham, E. G. D., et al., “The properties of factor VIII coagulant activity prepared by immunoadsorbent chromatography”, J Lab Clin Med, 93: 40 (1979).

2. Austen, D. E. G., “The chromatographic separation of factor VIII on arninohexyl sepharose”, Brit J Heamotol, 43: 669 (1979).

3. Zimmerman, et al., “Ultrapurification of factor VIII using monoclonal antibodies”, U.S. Pat. No. 4,361,509 (Nov. 30, 1982).

4. Wood, W. I., et al., Expression of active human factor VIII from recombinant DNA clones”, Nature, 312: 330-337 (1984).

5. Rotblat, F., et al., “Purification of human factor VIII:C and its characterization by western blotting using monoclonal antibodies”, Biochemistry, 24: 4294 (1985).

6. Baumbach, G. A., et al., “Protein Purification Using Affinity Ligands from Peptide Libraries”, BioPharm, May 1992, 24-35.

7. Buettner, J. A., et al., “Chemically derived peptide libaries: A new resin and methodology for lead identification”, Int J Peptide Protein Res, 47: 70-83 (1996).

8. Huang, P. Y., et al., “Affinity Purification of Proteins Using Ligands Derived from Peptide Libraries”, Biotechnol & Bioeng, 47: 288-297 (1995).

9. Huang, P. Y., et al., “Affinity Purification of von Willebrand Factor Using Ligands Derived from Peptide Libraries”, Bioorg & Med Chem, 4: 699-708 (1996).

10. Furka, A., et al., “General method for rapid synthesis of multicomponent peptide mixtures”, Int J Peptide Protein Res, 37: 487-493 (1992).

11. Lam, K. S., et al., “A new type of synthetic peptide library for identifying ligand-binding activity”, Nature, 354: 82-84 (1991).

12. Houghten, R. A., et al., “Generation and use of synthetic peptide combinatorial libraries for basic research and drug discovery”, Nature, 354: 84-86 (1991).

13. Lam, K. S., et al., “Application of a dual color detection scheme in the screening of a random combinatorial peptide library”, J Immunol Meth, 180: 219-223 (1995).

14. Meldal, M., et al., “Portion-mixing peptide libraries of quenched fluorogenic substrates for complete subsite mapping of endoprotease specificity”, Proc Nat Acad Sci USA, 91: 3314-3318 (1994).

15. Meldal, M., et al., “Direct visualization of enzyme inhibitors using a portion mixing inhibitor library containing a quenched fluorogenic peptide substrate. Part 1. Inhibitors for subtilisin Carlsberg”, J. Chem. Soc. Perkin Trans. 1: 1591-1596 (1995).

16. Needels, M. C., et al., “Generation and screening of an oligonucleotide-encoded synthetic peptide library”, Proc Nat Acad Sci USA, 90: 10700-10704 (1993).

17. Kassarjian, A., et al., “Screening of Synthetic Peptide Libraries with Radiolabeled Acceptor Molecules”, Peptide Research, 6: 129-133 (1993).

18. Turck, C. W., “Radioactive Screening of Synthetic Peptide Libraries”, Methods: A Companion to Methods in Enzymology, 6: 396-400 (1994).

19. Mondorf, K, et al., J Pept. Res. in press (1998).

20. Jentoft, N. et al., “Protein Labeling by Reductive Alkylation”, Meth Enzymol, 91: 570-579 (1983).

21. Barrowcliffe, T. W., “Recommendations for the assay of high-purity factor VIII concentrates” Thromb Haem 70: 876-877 (1993).

22. Baumbach, G. A., et al., “Peptides which bind to prothrombin and thrombin”, U.S. Pat. No. 5,831,003 (Nov. 3, 1998).

23. Necina, R., et al., “Peptide affinity chromatography of human clotting factor VIII: Screening of the vWF-binding domain”, J Chrom B 715: 191-201 (1998).

22 1 6 PRT Artificial Sequence Description of Artificial SequenceSynthetic 1Lys Pro Asn Pro Leu Ala 1 5 2 6 PRT Artificial Sequence Description of Artificial SequenceSynthetic 2Arg Asn Pro Pro Asn Asn 1 5 3 6 PRT Artificial Sequence Description of Artificial SequenceSynthetic 3Tyr Val Gln Gly Leu Trp 1 5 4 6 PRT Artificial Sequence Description of Artificial SequenceSynthetic 4Phe Arg Pro His Trp Ala 1 5 5 6 PRT Artificial Sequence Description of Artificial SequenceSynthetic 5Leu Asn Trp Lys Tyr Gly 1 5 6 6 PRT Artificial Sequence Description of Artificial SequenceSynthetic 6His Tyr Trp Phe Tyr Lys 1 5 7 6 PRT Artificial Sequence Description of Artificial SequenceSynthetic 7Ile Arg Phe Tyr Ser Glu 1 5 8 4 PRT Artificial Sequence Description of Artificial SequenceSynthetic 8Arg Pro Arg Trp 1 9 6 PRT Artificial Sequence Description of Artificial SequenceSynthetic 9Phe Ala Leu Pro Gly Arg 1 5 10 6 PRT Artificial Sequence Description of Artificial SequenceSynthetic 10Ala Phe Val Arg Ser Leu 1 5 11 6 PRT Artificial Sequence Description of Artificial SequenceSynthetic 11Asn Ala Ile Phe Gln Trp 1 5 12 6 PRT Artificial Sequence Description of Artificial SequenceSynthetic 12Gln Arg Leu Ile Gln Phe 1 5 13 6 PRT Artificial Sequence Description of Artificial SequenceSynthetic 13Lys Ala Gln Glu Thr Trp 1 5 14 6 PRT Artificial Sequence Description of Artificial SequenceSynthetic 14Glu Pro Arg Val Ile Gly 1 5 15 6 PRT Artificial Sequence Description of Artificial SequenceSynthetic 15Val Tyr Gly Val Gly Gly 1 5 16 6 PRT Artificial Sequence Description of Artificial SequenceSynthetic 16Trp Arg Arg His Arg Tyr 1 5 17 6 PRT Artificial Sequence Description of Artificial SequenceSynthetic 17Phe Tyr Arg Phe Trp Asn 1 5 18 6 PRT Artificial Sequence Description of Artificial SequenceSynthetic 18Trp Leu Trp Ser His Asn 1 5 19 6 PRT Artificial Sequence Description of Artificial SequenceSynthetic 19Phe His Phe Gly Leu Gln 1 5 20 6 PRT Artificial Sequence Description of Artificial SequenceSynthetic 20Trp His His His Arg Gly 1 5 21 6 PRT Artificial Sequence Description of Artificial SequenceSynthetic 21His Phe Gln Ile Phe Gly 1 5 22 6 PRT Artificial Sequence Description of Artificial SequenceSynthetic 22Phe Val Phe Leu Val Arg 1 5

Claims

1. A composition comprising a peptide having an available rFVIII binding domain, wherein the binding domain is selected from the group consisting of Arg Asn Pro Pro Asn Asn; Tyr Val Gln Gly Leu Trp; Phe Arg Pro His Trp Ala; His Tyr Trp Phe Tyr Lys; Ile Arg Phe Tyr Ser Glu; Asn Ala Ile Phe Gln Trp; Gln Arg Leu Ile Gln Phe; Lys Ala Gln Glu Thr Trp; and Glu Pro Arg Val Ile Gly (SEQ ID NOS: 2-4, 6, 7, and 11-14, respectively).

2. The composition of claim 1, wherein the peptide is immobilized upon a water-insoluble support.

3. The composition of claim 2, wherein the support is a chromatography medium.