Role of the B Domain for Factor VI11 and Factor V

From www.bloodjournal.org by guest on December 22, 2014. For personal use only.
Role of the B Domain for Factor VI11 and Factor V Expression
and Function
By Debra D. Pittman, Kimberly A. Marquette, and Randal J. Kaufman
Factor V and factor Vlll are homologous cofactors in the
blood coagulation cascade that have the domain structure
Al-A2-B-A3-Cl-C2, of which the B domain has extensively
diverged. In transfected COS-l monkey cells, expression of
factor Vlll is approximately 10-fold less efficient than that of
factor V, primarily because of inefficient protein secretion
and, t o a lesser extent, reduced mRNA expression. To study
the functional significance and effect of the B domain on
expression and activity, chimeric cDNAs were constructed
in which the B domains of factor V and factor Vlll were
exchanged. Expression of a factor Vlll chimera harboring the
B-domain of factor V yielded a fully functional factor Vlll
molecule that was expressed twofold more efficiently than
wild-type factor Vlll because of increased mRNA expression.
Thus, sequences within the factor Vlll B domain were not
responsible for the inefficient secretion of factor Vlll compared with factor V. Expression of a factor V chimera harbor-
ing the B domain of factor Vlll was slightly reduced compared with wild-type factor V, although thesecreted molecule had significantly reduced procoagulant activity correlating with dissociated heavy and light chains and resistance
to thrombin activation. Interestingly, the factor V chimera
containing the factor Vlll B domain wasefficiently activated
by Russell's viper venum (RW).Afactor V B domain deletion
(residues 710-1545) molecule also exhibited significantly reduced procoagulant activity caused by resistance t o thrombin cleavage and activation, although thismolecule was activatable by R W . These results show that, in contrast t o
V requires sefactor VIII, thrombin activationoffactor
quences within the B domain. In addition, thrombin activation of factor V occurs through a different mechanism than
activation by RVV.
0 1994 b y The American Society of Hematology.
B
suggesting that the B domains evolved from a single exon
and may function in similar roles in regulating the
expression
and activity of factors V and VIII.
In plasma, factor VI11 circulates as a metal-ion bridged
heterodimer consisting of a heavy chain
of 200 kD and a
light chain of 80 kD that are processed from a single-chain
precursor polypeptide within the B domain onsecretion from
theThis
heterodimer is stabilized by interaction with
anotherplasma glycoprotein,von Willebrand factor." In
contrast, plasma factor Visasingle-chainpolypeptide
of
330 kD and is not detected in association with other plasma
proteins.l4.ls Both factor V',""' and factor VIII'9-2' require
proteolyticactivation
to elicit full procoagulantactivity.
Thrombin activation of factor VI11 results in cleavage initially after residue 740 and subsequently after residues 372
and 1689(Fig1). Thrombin-activatedfactor VI11 consists
of a heterotrimer of a 50-kD AI-domain-derived polypeptide,a 43-kD A2-domain-derived polypeptide, and a 73kD-derived
light-chain
Thrombin
cleavage
of
factor V occurs first after residue 709 and then afterresidues
1018 and 1545 to yield the activated heterodimer consisting
of the 94-kD (alsoreferred to as a 105-kD polypeptide)25
heavy-chain fragment and the 74-kD light-chain
Upon thrombin activation, the B domains of both factors V
and VIII are released.
Previous studies showed that the B domain of factor VI11
is dispensible for procoagulant activity.**-'" Factor VI11 Bdomain deletion molecules extending from residue 740 to
1689 exhibitspecific activity similar to wild-type factorVIII
and are expressed more efficiently because of a more efficient
One B-domain deletion molecule, LAVIII, was active in vivo, as measured by its ability to correct
the cuticle bleeding time
on infusion into a hemophilic dog."
However, detailedcharacterizationindicated
that different
B-domain deletion mutantshave altered patternsof thrombin
activation,possibly caused by analtered context of the
thrombin cleavage
Similarly,
deletion of residues
81 1-1491 within the B domaindid notdestroy factor V
activity in vitro." In the studiesdescribed here, we evaluated
the influence of the B domain on expression and functional
LOOD COAGULATION is controlledby the regulated
activation of serine proteases in the coagulation cascade. Factor VI11 and factor V are two large glycoproteins
that function as essential cofactors for proteolytic activation
of factor X andprothrombin, respectively.Both proteins
circulate in plasma as inactive precursors that are activated
through limited proteolysis by either thrombin or activated
factor X (Xa). After activation, factor VIIIa and factor Va
assemble with their respective substrates (factor X and prothrombin) andenzymes (factor IXaandfactor
Xa) on a
negatively charged phospholipid surface in the presence of
calcium ions.',2 Both cofactors act to increase the Vmax of
substrate activation by four ordersof magnitude. Elucidation
of the primary structure of factors V',4 and VII15.6showed
that they share amino acid identity and have a conserved
domain organization of Al-A2-B-A3-Cl-C2 (Fig l). The A
domains are homologous to the
A domains of ceruloplasa copper-bindingplasmaprotein,suggestinga
role
in metal-ionbinding. The C domainsarehomologousto
phospholipid-binding proteins, suggesting a role in phospholipid interaction.x Whereas the A and C domains are 40%
identical between factors V and VIII, there is only limited
homology between the B domains.' However, bothBdomains do containa large number of asparagine-linked oligosaccharides. Within the genome, the factor VI11 and factor
VB domainsreside onunusuallylarge
singleexons,'".''
From the Genetics Institute, Cambridge, MA; and the Howard
Hughes Medical Institute, Department of Biological Chemistry, University of Michigan Medical Center, Ann Arbor, MI.
Submitted May 3, 1994; accepted August 26, 1994.
Address reprint requests to Randal J. Kaufman, PhD, Howard
Hughes Medical Institute, Department of Biological Chemistry, University of Michigan Medical Center, Ann Arbor, MI 48105.
The publicationcosts of this article were defrayed in part by page
charge payment. This article must therefore be hereby marked
"advertisement"in
accordance with 18 U.S.C. section 1734 solely to
indicate this fact.
0 1994 by The American Society of Hematology.
0006-4971/94/8412-0022$3.00/0
421 4
Blood, Vol 84, No 12 (December 15). 1994: pp 4214-4225
From www.bloodjournal.org by guest on December 22, 2014. For personal use only.
4215
ROLE OFTHE B DOMAIN IN FACTORS V AND Vlll
E.
Fig 1. Domain structure and processing offactors
V and VIII. Thestructural domeins of factor Vlll and
factor V are depicted. Both proteins have acidic regions ).(
that aresitesof
tyrosine suulfation."~"
Upon secrotion from the cell, factor Vlll is cleaved
after residue 1313 and 1648 to generate the 200-kD
heavychain and an80-kD light chain (A andB).
Thrombin (Ha) cleaves the heavy chain at residues
740 and 372 to generate the 43-kD and 50-kD polypeptides and after 1689 to generate the 73-kD light
Vlll
chain (C). Factor Vis secreted as a single-chain polypeptide (D) and is cleaved by thrombin at arginine
VlllB5
residues 709,1018, and l545 to generate the heterodimer composed of the 94-kD heavy chain and 74LA-VI11
kD light chain. Wild-type factor Vlll (VIII), the factor
Vlll B-domain deletion (LA-VIII), the factor Vlll hybrid V
harboring the B-domain of factor V (VlllB5). wildtype factor V (V), the factor V hybrid harboring the
VB8
factor Vlll Bdornain (VB8), and the factor V B-domain deletion
),,V
,(
molecules aredepicted below
v94ff4
([U] factor Vlll sequence; [
F.
activity of factors V and VI11 by analysis of B-domain deletion molecules and of chimeric molecules between factors
V and VI11 in which B-domain sequences were exchanged.
The results show a unique requirement of the factor V B
domain for activity and thrombin activation.
MATERIALS AND METHODS
Materials. Rabbit anti-factor V polyclonal antibody was purchased from Dako Corp (Carpinteria, CA). Anti-factor VIII heavy
chain monoclonal antibody (MoAb) F-8 was previously described?'
Anti-factor VI11 light chain MoAb was obtained from Hybritech
Corp (La Jolla, CA). Factor V MoAb E9 directed against the factor
V light chain'* was kindly provided by Dr K.G. Mann (University
of Vermont, Burlington, VT). Factor VIII-deficient plasma, factor
V-deficient plasma, and normal pooled human plasma were obtained
from George B. King Biomedical Inc (Overland Park, KS). Thromboplastin, soybean trypsin inhibitor, phenylmethylsulfonyl floride,
and aprotinin were purchased from Sigma Corp (St Louis, MO).
Human thrombin, human factor Xa, human factor V, and Russell's
viper venom (RVV) were obtained from Hematological Technology
(Burlington, VT). [35S]-methionine(> 1,OOO Ci/mmol) was obtained
from Amersham Corp (Arlington Heights, L).Activated partial
thromboplastin ( A m ) was obtained from Organon Teknika (Durham, NC).
Isolation of a genomic clone encoding the factor V B domain.
Previously reported human factor V cDNA sequence^'.^ differed in
the number of a 9 amino acid repeat within the B domain. We
therefore characterized whether the cDNA previously isolated4 was
representative of that encoded by the human genome. The factor V
B domain was isolated from a human genomic DNA library constructed in the Lambda Fix replacement vector (Stratagene, La Jolla,
CA) by hybridization to an [(u-~'P]-~ATP
and [(Y-~'P]-~CTP
nicktranslated 2.5-kb Pst I-Kpn I fragment from the factor V cDNA
encoding the entire B d~main.'~
The genomic B domain fragment
from phage 6 was isolated as an approximately 2-kb Sac I fragment
and subcloned for sequencing. DNA sequences were determined on
both strands using the collapsed-supercoil dideoxy chain termination
methodgs with Bluescript primers and factor V-specific synthetic
oligonucleotides. The DNA sequence of the entire B domain, spanning from the 5' to the 3' introdexon boundary, was identical to
the previously identified cDNA sequence? The deduced amino acid
From www.bloodjournal.org by guest on December 22, 2014. For personal use only.
4216
sequence differed at 4 amino acid residues from the human genomic
sequence previously published.” The sequence of the B-domain
confirmed the presence of 37 repeats of 9 amino acids and ruled out
the presence of additional repeats within the factor V gene that were
deleted through cDNA cloning.
Derivation of the factor V expression vector. The full-length
factor V cDNA was assembled from two overlapping Charon 21A
recombinants each containing portions of the factor V cDNA derived
from an oligo-dT-primed human fetal liver library.4 The recombinants designated V401 and V1 were digested with Sac 11 to remove
the factor V cDNA and flanking lambda sequences and ligated to
Sac It-linearized Bluescript KS+ cloning vector (Stratagene). The
resulting plasmids were designated 401B5-6 and VlBS-IO, respectively. The lambda sequences were removed from 410B5-6 by digestion with Hha I (nucleotide 41 in the factor V CDNA)? followed by
treatment with the Klenow fragment ofDNA polymerase I. The
plasmid DNA was then digested with Bgl I1 (nucleotide 4275 in the
factor V cDNA) and the 4234 fragment was cloned into the EcoRV
and BamHI cloning sites of Bluescript KS+, generating plasmid Blu
401. The Sal I site derived from the Bluescript polylinker provides
a unique cloning site 5’ to the initiator methionine codon.
VlBS-10 was digested with Hpa I (nucleotide 6854 in the factor
V cDNA4) removing flanking lambda sequence and 71 bp of the 3’
noncoding sequence. An EcoRI adapter (5’-AATTCCGTCGACTCTAGAG-3’) containing a Sal 1 site was ligated to the Hpa Idigested VlBV-10 resulting in the plasmid BluVI-RI. The unique
BstEII site (nucleotide 3666 in the factor V cDNA) present in both
clones was used to assemble the full-length factor V cDNA. The
3625 Sal I-BstEII fragment from Blu401 contributing the 5’ end of
the factor V cDNA and the 3188 Sal I-BstEII fragment from BluV 1RI contributing the 3’ end of the cDNA were ligated to Sal Ilinearized Bluescript KS, resulting in the plasmid containing the
full-length factor V cDNA (Bluescript-V).
A Sal I derivative of the expression vector pMT2 was constructed
into
by cloning an adapter (5‘-AATTCCGTCGAACTCTAGAG-3’)
the EcoRI site. The full-length factor V cDNA was excised from
Bluescript-V on a Sal I fragment and cloned into the Sal I-containing
derivative of the expression vector pMT2” and was designated
pMT2-V.
Plasmid mutagenesis and hybrid construction. The factor VI11
expression vector pMT2-VI11 was previously de~cribed.’~
Both Bluescript-V and pMT2V were used for construction of hybrid molecules.
Site-directed oligonucleotide-mediatedmutagenesis was performed by
the heteroduplex procedure” to sequentially introduce unique Mlu I
restriction sites at proteolytic cleavage sites at residues 740 and 1648
within factor VI11 (designated pMT2-VI11 Mlu 1-740 and pMT2-VI11
Mlu l- 1648) and atresidues 709 and 1545 within factor V (designated
pMT2-V Mlu 1-709 and Bluescript-V Mlu 1-1545). The Mlu I sites
facilitate exchange of the B-domain segments. The oligonucleotides
used for mutagenesis were the following (residue and mutagenic
oligonucleotide): factor VIII: 740, 5‘-GTAAAAACAATGCCATTGAAACGCGTAGCTTCTCCCAGAAlTC-3’, 1648, 5”CCAGTCTT
GAAACGCCATACGCGTGAAATAACTCGTACTACTC-3’;
factor V: 709,5’-GGCTGCAGCATTAGGAACGCGTTCATTCCGAAACTCATCATTG-3’.
1545,
5”CATTcCAGCATGGTAC&
GCGTAGCAACAATGGAAACAGAAG-3’. All mutations were
confirmed by DNA sequencing and extensive mapping with restriction
enzymes.
The factor V chimera harboring the B-domain of factor VIII,
designated pVB8, was constructed by a 4-way ligation of the Mlu
I and BspMI fragment from pMT2-V Mlu 1-709, the Mlu I and
BspMI fragment from Bluescript-V Mtu 1-1545, the Mlu I-EcoRV
fragment from pMT2-VIII Mlu 1-1648, and the Mlu I-EcoRV fragment from pMT2-VI11 Mlu 1-740. The factor VI11 hybrid containing
the B domain of factor V, designated pVIIIB5, was generated by a
PITTMAN, MARQUETTE, ANDKAUFMAN
4-way ligation of the Mlu I-Sal I fragment from pMT2-VIII Mlu i740, the Mlu I-BstEII fragment from pMT2-V Mlu 1-709, the Mlu
I-BsrEII fragment from Bluescript-V Mlu 1-1545, and the Mlu I-Sal
I fragment from pMT2-VI11 Mlu 1-1648. The factor V B-domain
deletion molecule was constructed by deletion of the Mllr I fragment
containing the factor VI11 B-domain from pVB8 and was designated
PV,,,, .
DNA transfection and analysis. Plasmid DNAwas transfected
into COS-l cells by the diethylaminoethyl (DEAE)-dextran procedure.” Conditioned medium was harvested 60 hours posttransfection
in the presence of 10% heat-inactivated fetal bovine serum for factor
V and factor VI11 assay. Primary translation products were analyzed
by pulse-labeling cells for 15 or 30 minutes, as indicated, with [3’S]methionine (250 pCi/mL in methionine-free medium) and preparing
cell extracts in a Nonidet-P40 lysis buffer.’* Protein secretion was
monitored by metabolically labeling cells with [”SI-methionine (250
pCilmL for 2 hours) and chase performed for 4 hours in medium
containing excess unlabeled methionine as described.” The factor
VI11 was immunoprecipitated with an anti-factor VI11 MoAb F-8‘’
coupled to CL-4B sepharose. Quantitatively similar results were
obtained with an anti-factor VI11 light chain MoAb (data not shown).
Factor V was immunoprecipitated with either MoAb E-9 specific to
the factor V light chain’’ or a rabbit polyclonal antibody coupled to
Affigel (BioRad, Richmond, CA). Theantibodies were tested before
the experiments to determine the amount of antibody required for
quantitative immunoprecipitation. In all cases, excess antibody was
used. The immunoprecipitates were washed as described.’” Immunoprecipitated proteins from the conditioned medium were resuspended
50 mmoVL Tris-HCI, pH 7.5, 150 mmol/L NaCI, 2.5 mmol/L CaC12,
and 5% glycerol (buffer A) and subjected to digestion with thrombin
( I2 U/mL for 60 minutes at 37°C)before electrophoresis on a sodium
dodecyl sulfate (SDS)-low bis-8% polyacrylamide gel.” Proteins
were visualized by autoradiography after fluorography by treatment
withEn’Hance (Dupont, Boston, MA). The band intensities were
quantitated by scanning the lanes using an LKB UltroScan XL laser
densitometer (Pharmacia LKB Biotechnology Inc, Uppsala, Sweden).
Total RNA was isolated from transfected COS-l cells at 60 hours
posttransfection using the guanidine thiocyanate method.‘“ RNA was
fractionated on a 0.8% agarose formaldehyde gel and subsequently
transferred to nitrocellulose.”’ Probes were prepared by nick translation of dihydrofolate reductase (DHFR) or chicken P-actin cDNA
fragments.” Transcripts were analyzed by hybridization at 65°C in
a solution containing 2X SSC, 0.1% SDS, 5 X Denhart‘s reagent,
and 100 pg/mL denatured salmon sperm DNA, followed by autoradiography and densitometric scanning.
Factor VI11 and factor V activity assay. Factor VI11 activity was
measured by a chromogenic assay5 or in a clotting assay using factor
VIII-deficient plasma.24 One unit of factor VI11 activity is that
amount measured in 1 mLof normal human pooled plasma. For
thrombin activation, conditioned medium was diluted into buffer A
and incubated at room temperature with 1 UlmL thrombin. At short
intervals, the sample was diluted and assayed for factor VI11 activity
in a factor VI11 clotting assay.
Factor V activity was measured in a factor V clotting assay using
factor V-deficient pla~ma.~’
Standard curves were prepared by dilution of pooled normal plasma in 20 mmol/L imidazole, pH 7.4, and
0.15 mol/L NaCl (buffer B). Total factorV was determined by
measuring the peak activity after treatment with thrombin. Factor V
conditioned medium was incubated at room temperature with 1 U/
mL thrombin. At timed intervals, aliquots were removed and assayed. The activation quotient was determined by dividing the
thrombin-activated activity by the nonactivated activity. One unit of
factor V is that amount in 1 mLof normal pooled human plasma
before activation with thrombin. Factor Xa activation was performed
From www.bloodjournal.org by guest on December 22, 2014. For personal use only.
ROLE OF THE B DOMAIN
IN FACTORS V AND Vlll
by incubation with I pg/mL factor Xaat room temperature in 20
mmol/L imidazole, pH 7.4, and 0.15 mol/L NaCl containing the
phospholipid inosithin at I O 0 pg/mL and assayed at timed intervals.
RVV activation was performed at 2 p.g/mLRVV in 20 mmol/L
imidazole. pH 7.4. 0.15 mol/L NaCI, and 20 mmol/L CaCI2. Before
assay the samples were diluted into 20 mmol/L imidazole, pH 7.4,
0.15 mol/L NaCl and 2 mmol/L EGTA. Optimal RVV activity required the addition of 20 mmol/L CaCI2. Thereactions were diluted
into 2 mmol/L EGTAto reduce the calcium ion concentration to
prevent calcium activation in the clotting assay.
Ann/wis of fl~rotnbinnc/ivnfion and clrnvnge. At 60 hours posttransfection, cells were metabolically labeled with ["SI-methionine
(250 pCilmL) as described above and conditioned medium was harvested." Factor V was immunoprecipitated withan anti-factor V
polyclonal antibody. The immunoprecipitates were resuspended in
buffer A, aliquoted, andtreatedwith increasing concentrations of
thrombin at37°C for I hour. Factor VI11 was immunoprecipitated
with the MoAbF8 coupled to CL4B sepharose. The immunoprecipitated protein was resuspended in buffer A and treated with 1.5 U/
mL thrombin at room temperature for the indicated periods of time.
Reactions were terminated by the addition of 2-mercaptoethanol
(2.5%) and SDS (0.5%) and heating at 85°C. The resulting polypeptides were separated by SDS-polyacrylamide gel electrophoresis
(SDS-PAGE) and visualized after fluorography by treatment with
En'Hance.
4217
CM
CE
30' pulse
4 h chase
i
-1
Y
Mock V
0
$>:"-
Vlll
-+-+ - +
Ila
RNA
kDa
-
.200
.92
1
2
3
.69
,46
RESULTS
Factor Vlll, compared with factor
V, is ineflciently secreted from COS-] monkey cells. Factor V and factor VI11
mRNA and protein expression were compared by transient
transfection in COS-l monkey cells. The factor V and factor
VI11 cDNA expression vectors were transfected into COS-l
cells and, after 60 hours, factor V and factor VI11mRNA
expression was measured by Northern blot analysis of total
RNA. Because DHFR sequences are present in the 3' end
of the mRNAs derived from the expression vector pMT2,
hybridization to a single "P-labeledDHFR probe permits
comparison of mRNA levels from vectors having different
cDNA inserts. The results showed that the level of factor V
rnRNA was approximately twofold greater than factor VI11
(Fig 2, lanes 2 and 3, and Table l ) . Factor V and factor VI11
protein synthesis were evaluated by immunoprecipitation of
cell extracts prepared from ['5S]-methionine pulse-labeled
cells. Analysis of extracts from cells transfected with pMT2V detected a prominent polypeptide migrating at 330 kD
(Fig 2, lane 5) thatwasnotpresent
in cells thatdidnot
receive DNA (Fig 2, lane 4). Similar analysis of extracts
frompulse-labeled cells transfected withpMT2-VI11 detected a single-chain polypeptide of 280 kD on irnmunoprecipitation with anti-factor VI11 antibody (Fig 2, lane 6). Comparison of the primary translation products for factor VI11
and factor V showed that factor V synthesis was approximately twofold greater than factor VI11 (Fig 2, lanes 5 and
6). The factor V and factor VI11 primary translation products
contain 52 and 60 methionine residues, respectively. Thus,
the difference in translation rates was proportional to the
different mRNA levels.
The secreted factor V and factor VI11 were analyzed by
immunoprecipitation of conditioned medium after a 4-hour
chase period in medium containing excess unlabeled methionine. Factor V was recovered in the conditioned medium as
4
5 6 7 8 9 1 0 1112
Fig 2. Expression of factorV and factor Vlll in COS-lcells. Expression vectors encoding factor Vlll or factor V were transfected into
COS-l monkeycellsand at 60 hoursposttransfection,cells were
pulse-labeled with [35S1-rnethionine for30 minutes and cell extracts
ICE) were prepared. In parallel, labeled cells were chased for 4 hours
in medium containing excess unlabeled methionine and conditioned
medium (CM) was harvested. Samples were immunoprecipitated
with anti-factor VIII- (lanes 6, 11, and 12)or anti-factor V-specific
antibody (lanes 4, 5, and 7 through 10) for analysis by SDS-PAGE.
Aliquots of precipitated protein were digested with thrombin (Ila)
before SDS-PAGE (lanes8,10, and 12).Total RNA was preparedfrom
cells transfected in parallel and analyzed byNorthern blot hybridization to a DHFR probe that hybridizes to the 3' end of both the factor
Vlll and factor V mRNAs derived from the expression vector. The
molecular weight markers are shown on the right.
an abundant polypeptide migrating at 330 kD (Fig 2, lane
9). Thrombin digestion of the immunoprecipitated protein
before electrophoresis yielded the expected cleavage products representing the 150-, 94-, and 74-kD fragments (Fig
2, lane IO) that are characteristic of plasma-derived factor
V.'5In contrast, low amounts of factor VI11 were detected
in the conditioned medium that migrated as a 200-kD heavy
chain and a 80-kD light chain (Fig 2, lane 1 l). Thrombin
digestion of the immunoprecipitated factor VI11 yielded the
expected cleavage products migrating at 73, 50, and 43 kD
(Fig 2, lane 12).2'.37In this analysis, the 73-kD polypeptide
migrates just below the 69-kD marker. Quantitation of data
averaged fromsix
independent transfection experiments
showed thatthe amount of radiolabeled factor V polypeptide
secreted into the conditioned mediumwas IO-fold greater
than observed for factor VI11 (Table I ) . These results show
that factor V was more efficiently expressed than factor VI11
in COS-l cells. The increased factor V expression resulted
From www.bloodjournal.org by guest on December 22, 2014. For personal use only.
4218
PITTMAN,MARQUETTE,ANDKAUFMAN
Table 1. Relative mRNA Levels, Protein Expression, and Activity of
Wild-Type, B-Domain Deletion, and Chimeric Molecules
Mean 2 SD (n)
mRNA
Primary Translation
Product Protein
~V/MVlIl*
2.2 ? 1.1 (4)
1.63 2 1.5 (3)
Secreted
10.2 ? 3.6
(6)
Primary
Translation
Product
mRNA
Plasmid
Secreted Protein
Activity
1.o
1.3
2.0
1.o
1.o
1.o
1 .o
2.3
0.63
1.o
0.34
0.25
~
Factor Vlllt
pWtVlll
pLA-VIII
2.0
pVlllB5
1.o
1 .o
15
17
1.9
2.1
1.3
Factor VS
PMV
3.0
PV9W4
pVB8
1 .o
3.4
0.4
* COS-l cells were transfected with pMT2-V or pMT2-VIII and at
60 hours posttransfection expression was analyzed as described in
Materials and Methods. Total RNA was isolated and Northern blot
analysis performed byhybridization to aDHFR probe. Band intensities
were quantitated and the ratio of factor V mRNA to factor Vlll mRNA
is shown withthe standard deviation (SD) for four independent transfection experiments. In parallel, cells were pulse-labeled with [%]methionine for15 minutes and cell extracts were prepared for immunoprecipitation and SDS-PAGE analysis. The band intensities for the
primary translationproducts were determined for three independent
experiments and the ratio of factor V to factor Vlll is shown. Protein
secreted into the conditioned mediumwas quantitated by [35Sl-methionine pulse-labeling with a4-hour chase in mediumcontaining excess
unlabeled methionine. Protein was immunoprecipitated and digested
with thrombinbefore SDS-PAGE. For more consistent quantitation of
the secreted protein, the band intensities of the thrombin cleaved
light chain for factor V and factor Vlll were scanned by a densitometer.
The factor V and factor Vlll light chains contain 20 and 19 methionine
residues, respectively. The ratio of secreted factor V to factor Vlll
protein is presented from six independent transfection experiments.
t Factor VIll secreted protein and the mRNA levels were quantitated
from the data in Figs 3 and 4 as described above. Quantitation of
primay translation products was from data not shown and activity
was from data shown in Table 2.
Factor V secreted protein and the mRNA levels were quantitated
from the data in Figs 3 and 4 as described above, except that the
quantitation of secreted protein was performed before thrombin digestion by adding the band intensities for all radiolabeled factor Vrelated polypeptides in the conditioned medium. All values presented
are from the same transfection experiment and are relative to the
values obtained for wild-type factor Vlll and factor V. Independent
transfection experiments yielded data consistent with those presented for factor Vlll and factor V. The activity values presented are
results from factor V and factor Vlll clotting assays as described in
Materials and Methods.
difference in expression. To characterize sequences within
factor VI11 and factorV that were responsible for differences
in mRNA expression and/or secretion, B-domain deletions
as well as chimeric cDNAswith exchanged B domains were
I , bottom). The boundaries of
constructed (depicted in Fig
the B-domain chimeras were chosen to coincide with known
proteolytic cleavage sites. To facilitate construction of the
hybrid cDNAs, first unique MZu I restriction sites, that encode Thr-Arg, wereintroduced into the cDNAsof both factor
VI11 (residue 739-740) and factorV (residue 708-709)at the
borderof thrombin cleavage siteswithin the heavychain
and at the thrombin cleavage site within factor V light chain
(residue1544-1545)and attheintracellular
cleavage site
between the heavy and thelight chain in factor V111 (residue
1647-1648). The mutations of Pro739Thr and Gln,,,Thr in
factor VI11 and of Ile,o,Thr and LeulSuThr in factor V did
not affect either thesynthesis, secretion, thrombin cleavage,
or activity of the expressedmolecule (data not shown).
Chimeric cDNAs were then constructed by exchanging the
B domain encoding MZu I fragments between the factor V
and factor VI11 expression plasmids. A factor V cDNA was
constructed in which the factor VI11 B domain (residues 740
to 1648) replaced the factor VBdomain (residues 709 to
1545) and was designated VB8. A factor VI11 cDNA was
constructed in which the factor V B domain (residues 709
to 1545) replaced the factor VI11 B domain (residues 740 to
1648) and was designated VIIIB5. In addition, afactor V
B-domain deletion molecule was alsoconstructed and designated VY4,74ras described in Materials and Methods.The
factor VI11 B-domain deletion molecule designated LA-VI11
with residues 759 to 1639 deleted was previously characterized,28,'?.'8
Plasmid DNA expressionvectors harboring wild-type, Bdomain deletion mutants of factors V and VI11 and chimeric
cDNAs weretransfected into COS-l cells. The synthesisand
processing of these molecules was studied by metabolically
labeling cells with [?3]-methionine and immunoprecipitation of the conditioned medium with either an anti-Factor
VI11 heavy chain or ananti-factorVlightchainantibody
and reducing SDS-PAGE. Comparedwith secretion of wildtype factor VI11 (Fig 3, lane I), secretion of wild-type factor
V was increasedsixfold(Fig
3. lane 7). Secretion of the
factor VI11 deletion molecule LA-VI11 wasincreased 1.3fold compared with wild-typefactor VI11 (Fig 3, lane 3).
Expression of the VIIIB5 chimera was twofold greater than
wild-type factor VI11 (Fig 3, lane 3). The chimeric protein
VIIIB5 was secreted asasingle-chainpolypeptide
of the
expected size 330 kD. Thrombin digestion of the immunoprecipitated VIIIB5 protein generated the 73-kD light-chain
fragment and the 50-kD and 43-kD heavy-chain fragments
that comigrated with those derived from thrombin digestion
of wild-type factor VI11 and LA-VI11 (Fig 3, lanes 2. 4, and
6) and a fragment migrating below the200-kD marker, likely
from an approximately twofold increase in mRNA level and representing the factor V B domain.
In contrast to wild-type factor V (Fig 3, lane 7), immunoa fivefold increase in secretion efficiency.
precipitation analysis ofthe VB8 chimera using factor V
B-domain sequences do not inhibit secretion of factor VIII.
light-chain-specificantibodydetectedanapproximately
Because the B domains of factors V and VI11 are evolution3 10-kD polypeptide, likely representing single-chain
VB8.
arily the least conserved domains of the molecule, we asked
and the cleaved light chain, migrating at 74 kD (Fig 3, lane
whether the B-domain sequences contribute to the 10-fold
*
From www.bloodjournal.org by guest on December 22, 2014. For personal use only.
ROLE OF THE B DOMAIN IN FACTORS V AND Vlll
4219
VI11 LA-VIII VlllB5
V
VB8 94/74 Mock
n
n
n
n
n
n
n
Ila :
Fig 3. Expression ofwild-type and chimeric molecules in COS-l cells. PlasmidDNA expression vectors
encoding the indicated molecules were transfected
into COS-l cells. [35S1-methionine-labeledconditioned medium was prepared from transfected COS1 cells at 60 hours posttransfectionas described in
Materials and Methods. Factor Vlll and factorV polypeptides were quantitatively immunoprecipitated
using F-8 MoAb" for factor Vlll (lanes 1 through 6)
or E-9 MoAb" for factor V (lanes 7 through 141. Immunoprecipitatedproteins were resolved by reducing SDS-PAGE.Beforeelectrophoresis, portions of
the immunoprecipitateswere subjected to thrombin
digestion (+lla). Mock represents cells that did not
receive plasmid DNA. The migration of polypeptides
representingwild-type factor V (V) and wild-type factor Vlll (VIIII are indicated at the right as single chain
(SCI, heavy chain (HCI, and light chain (LC). The migration of the 73-,50-, and 43-kD thrombin cleavage
products of wild-type factor Vlll are also indicated
on the right. Molecular weight markers are shown
on the left.
-
-+ "-+-+-+-+-+
F-
'
V
- sc
-
Vlll
-
kDa
- HC
200 -
- HC
92 -
- LC
- LC
- 73
69 -
- 50
46
-
- 43
1
2
9). Secretion of the single-chain V98 chimera (Fig 3, lane
9) was slightly reduced compared with wild-type factor V
(Fig 3, lane 7). However. secretion of the B-domain deletion
molecule
(Fig 3, lane I I ) was increased 2.3-fold compared with wild-type factor V. Upon thrombin digestion of
VB8, the single-chain polypeptide was cleaved to lower molecular weight fragments, including one that comigrated approximately with the 94-kD factor V heavy chain (Fig 3, lanes
8 and IO). In contrast,themajority
ofthesecreted
Vgdn4
B-domain deletion molecule was resistant to thrombin cleavage (Fig 3, lane 12). In addition to analysis by immunoprecipitation of the conditioned media described above, factor
VI11 and factor V wild-type and mutant proteins were detected in the total conditioned medium on SDS-PAGE analysis. This analysis yielded quantitatively similar results to
those obtained on immunoprecipitation (data not shown).
This finding indicated thatthe differences in expression were
not caused by differences in antibody reactivity.
Replacement ?f factor V B-domain sequences into factor
VI11 increases mRNA expression. The above results and
other data not shown demonstrated a reproducible increase
in expression of the VIIIBS chimera compared with wildtype factor VIII. In addition, the expression of B-domaindeleted factor V was also increased, whereas expression of
V98 was slightly reduced compared with wild-type factor
V. To determine if the differences in expression were caused
by mRNA levels, translation, or secretion efficiencies, RNA
was prepared from transfected cells and analyzed by Northem blot analysis using a DHFR probe to permit comparison
ofmRNA levels for the different expression vectors. The
factor V mRNAwas expressed at a fivefold greater level
than the factor VI11 mRNA in this experiment (Fig 4A, compare lanes 2 and 7). The VB8 and VIIIBSchimeric molecules
yieldedmRNA levels that were intermediate between the
levels for wild-type factor V and wild-type factor VI11 (Fig
3
4
5
6
7
8
9
IO
II
12
1314
4A, lanes 4 and S). Deletion of the B domain of either factor
VI11 or factor V increased the mRNA level significantly, ie,
17- and 3.4-fold, respectively (Fig 4, lanes 3 and 6 , and
Table l ) . Hybridization to a @-actin probe showed that the
RNA was loaded equivalently in all lanes (Fig 4B). Independent transfection experiments gave results that were consistent with those presented in Figs 3 and 4. Analysis of extracts
from["SI-methionine pulse-labeled cells showed thatthe
level of the primary translation products were proportional
to the mRNA levels for all wild-type and mutant molecules
(data not shown, summarized in Table l ) . Whereas deletion
of the B domain in factor V increased mRNA and secreted
protein, deletion of the factor VIII B domain (LA-VIII) increased mRNAlevel 17-fold, but onlyyielded a 1.3-fold
increase in secreted protein. These results show that the levels of secreted protein for the B-domain chimeras and the
B-domain-deleted factor V94nA were proportional to the levels of their mRNAs. In contrast, although B-domain-deleted
factor VI11 yielded a significant increase in mRNA and primary translation product compared with wild-type factor
VIII, there was little increase in the amount of protein secreted into the conditioned medium (Table l ) . Although the
quantitation in Table 1 relies on immunoprecipitation with
specific antibodies, the analysis of [?3j-rnethionine-Iabeled
polypeptides by SDS-PAGE analysis of the totalconditioned
medium also yielded data that were consistent with these
conclusions (data not shown). Thus, the reduced expression
of LA-VI11 compared with factor Vqan4 resulted from a reduced secretion efficiency.
The factor VI11 chimera harboring the.factor V B domain
i s jmctional. Factor VI11 activity in the conditioned medium was determined at 60 hours posttransfection. Expression ofLA-VI11 yielded a slight increase in activity compared
with
wild-type
factor VIII, proportional to
the
increased secretion (Table 2). Expression of VIIIBS yielded
From www.bloodjournal.org by guest on December 22, 2014. For personal use only.
4220
PITMAN, MARQUElTE, AND KAUFMAN
B
I
2
3
4
5
6
7
Fig 4. Northern blot analysis of transfected COS-l cells. At 60
hours posttransfection, total cellRNA was isolated andanalyzed by
Northern blot hybridization using a DHFR-specific probe t o detect
vector-derived mRNAs (A). Subsequently, the blot was hybridized
to
a chicken p-actin probet o ensure equivalent RNA loading (B).
lanes 9 through 14) and of the VIIIBS chimera (Fig S , lanes
15 through 21), the rate of appearance of the 73-kD lightchain- and SO- and 43-kD heavy-chain-derived fragments
were similar. In addition, the rate of appearance and disappearance of the 90-kD heavy-chainintermediate was also
similar. This finding shows that thetwoproteinswere
equally susceptible to thrombin cleavage.
The B domain of factor V is required for thrombin octivcrtion. Conditioned medium was harvested from transfected
COS-l cells and factor V activity was measured in a factor
V clotting assay. The activity of the VB8 and Vojnd mutants
were threefold to fourfold lower compared with wild-type
factorV (Table 2). Correctingfor theamountofprotein
secreted into the conditioned medium, the specific activity
of the VsJn4 mutant was reduced sevenfold and that of VB8
was reduced 6.4-fold compared with wild-type factor V. In
contrast to wild-type factor V, the VpJnJand VB8 molecules
were not activatable on thrombin treatment (Table 2). Both
molecules also displayed significantly reduced activation by
factor Xa (Table 2). Interestingly, both VB8 and Vyjnd were
activated by RVV to almost a similar degree as wild-type
factor V (Table 2). These results show that at least a portion
of VB8 and Vgdn4molecules were folded in a manner to be
activated by RVV.
The susceptibility to thrombin cleavage of wild-type factor
VandtheB-domaindeletionmutant
Vsdn4 was compared
Table 2. Functional Activity ofWild-Type, B-Domain
Deletion, and Chimeric Molecules
Factor Vlll Activity
twofold greater factor VllI activity in the conditioned medium compared with the wild-type factor VIII control when
measured by either the Kabi Coatest factor Xa generation
assay or clottingassayusing
factor VIII-deficientplasma
(Table 2). Analysis of four independent transfection experiments showed the reproducibility of the transfection assay
(Table 2). Comparison of the amount of radiolabeled polypeptide present (Table 1) and the amount of activity in the
conditioned medium (Table2) showed that the specific activity for the wild-type, LA-VIII, and VIIIBS factor VI11 molecules were comparable. Factor VIII, the B-domain deletion
molecule LA-VIII previously characterized,32 and the VIIIBS
chimera all exhibited a similar extentof activation by thrombin (Table 2). In addition, the timecourseof thrombin activation for wild-type factor VlIl and VIIIB5 wereparallel, both
showing peak activation at 1 minute (data not shown).
The susceptibility of wild-type and VIIIB5 factor VI11 to
thrombin cleavagewascompared by immunoprecipitation
of [35S]-methionine-labeledconditioned medium and treatment of the immunoprecipitated protein with thrombin for
increasing periods of time before SDS-PAGE. Comparison
to thebackground bands observedin cells that did not receive
DNA (Fig S, lanes 1 through 7) showed that wild-type factor
Vlll migrated as a 200-kD polypeptide and a 80-kD light
chain before thrombin digestion (Fig S , lane 8). In contrast.
the VIIIBS chimera migrated as a 330-kD polypeptide (Fig
5, lane IS). Upon thrombin treatment of
wild-type (Fig S,
Coatest Assay
Plasmid DNA
(mUlmL; % wt 2 SDI
pwtVlll
PIA-VIII
pVIIIB5
62
157 (100)
207 (132 ? 6)
334
(215
z 28)
Clotting
Activity
(mUlrnLI
Fold Activation
665
ND
1,491
34
Factor V Activity
Fold Activation
Plasmid DNA
Clotting Activity
(mUlmL; % wt 2 SDI
Ila
R W
Xa
PwtV
PV9'"l
pVB8
185 (100)
62 (36 ? 13)
46 (31 ? 9.6)
5.5
1.1
1.5
6.8
4.5
5.8
8
3
2
Plasmid DNA was transfected into COS-l cells. After 40 hours, cells
were fed fresh complete medium. After 24 hours, conditioned media
samples were collected for factor Vlll and factor V activity assays as
indicated and the cells were labeled with [35S1-methionineto study
the synthesis and secretion as shown in Fig 3. Thrombin (Ha) and RVV
activation coefficients were determined as described in Materials and
Methods. The activity values are from the same transfection experiment analyzed in Fig 3. The mean activity values relative to wildtype (% w t ) and the standard deviation (SD) are presented for four
independent transfection experiments with two independent preparations of DNA. Activity from mock-transfected cells was less than 1
mU/mL for factor Vlll (coatest assay) and less than 10 mUlmL for
factor V (clotting assay).
Abbreviation: ND, not determined here but was previously rep~rted.~'.'~
From www.bloodjournal.org by guest on December 22, 2014. For personal use only.
ROLE OF THE
B
DOMAIN IN FACTORS V AND VI11
4221
v111
Mock
Fig 5. Time course of thrombin cleavage for wild-type factor
Vlll and VlllB5. COS-l cells were
transfected with wild-type factor
Vlll (lanes 8 through 14) or VlllB5
(lanes 15 through 21) expression
vectors and after 60 hours were
labeled with ["SI-methionine for
2 hours, followedby a 4-hour
chase in medium containing excess unlabeled methionine. Conditioned medium was harvested
and immunoprecipitated with
anti-factor Vlll antibody F-8. Immunoprecipitates
were
resuspended and treated with thrombin (Ma; 1.5 UlmL) for increasing
periods of time as shown. Samples were analyzed by SDSPAGE and fluorography. Molecular weight markers are shown on
the left. Lanes 1through 7 represent cells that did not receive
DNA.
I
11
1 0 5 IO 15 20 40 60
0
- 92
- 69
I
"
- 45
"-
- 30
I
2
3
4
5
6
7
by metabolically labeling transfected cells with [3sS]-methionine and analyzing the secreted protein by immunoprecipitationwith a factor V-specific antibody. Incubation of the
immunoprecipitated proteinwith
increasing amounts of
thrombinbefore reducing SDS-PAGE generated the expected cleavage products that likely represent the light chain
of 74 kD, a heterogeneous heavy chain of 94 kD, and the Bdomain fragment of 150 kD characteristic of plasma-derived
factor V (Fig 6, lanes 2 through 5). In contrast, upon treatment
of
the
Vyjf14
immunoprecipitate with increasing
amounts of thrombin, there was minimal cleavage to yield
theheavy chain andlight chain polypeptides comigrating
with wild-type thrombin-cleaved factor V (Fig 6, lanes 6
through IO). At 100 U/mL there were still significant
amounts of single-chain VYjfl4(Fig 6, lane IO).
v
Vlll85
-l
94/74
8
9
IO
I1
12 13 14
15
The factor V R domain is required for heavy and light
chain association. Because the chimera VB8 was secreted
primarily as a two-chain polypeptide because of intracellular
proteolytic processing within the B domain of factor VIII,
we investigated the association of the heavy and light chains
by a coimmunoprecipitation assay. Immunoprecipitation
with the factor V light-chain-specific MoAbE9 detected
single-chain factor V in conditioned mediumfromwildtype factor V-transfected cells (Fig 7. lane 1). Subsequent
immunoprecipitation of the supernatant of the E9 precipitation with a polyclonal anti-factor V antibody identified residual amounts of immunoprecipitable wild-type factor V (Fig
7, lane 2). In contrast, immunoprecipitation of conditioned
medium from cells transfected with VB8 detected the factor
V light chain migrating at 74 kDand some single-chain
Mock
r
.-
kDo
200 -
- 330
- 120
- 94 (HC)
- 74 (LC)
69 I 2 3
4 5 6 7 8 9 IO
16 17 18 19 20 21
II 12 1314 15
Fig 6. Thrombin cleavage of
wild-type factor V and
[3sSl-methionine radiolabeled
factor V was prepared from cells
transfected withthe indicated
expression vectors encoding factor V (V, lanes 1 through 5) or
Vgrn4 (lanes 6 through 10). After
immunoprecipitation withthe
anti-factor V MoAb, precipitated
proteinwas
resuspended and
subjected t o thrombin digestion
with increasing amounts
of
thrombin (lla). Samples were
then analyzed by SDS-PAGE.
The 330-kD single chain, the 150kD activation peptide, the 94-kD
heavy chain (HC), and the 74-kD
light chain (LC) are indicated.
Molecular weight markers are
shown on the right. Mock samples did not receive DNA (lanes
11 through 15).
From www.bloodjournal.org by guest on December 22, 2014. For personal use only.
PITMAN, MARQUETTE, ANDKAUFMAN
4222
V
VB8 Mock
"
m
Ab: E9 V E9 V E9 V
kDa
-
200 -
"
L
-
92 69 -
46 I
2
3
4
5
6
Fig 7. The heavy and light chains in the VB8chimera are not
associated. COS-l cells were transfected with wild-type factor V (V)
or the chimera containing the B-domain of factor Vlll (VB5) expression vectors. Cells were labeled with 135Sl-methionine as described
in Materials and Methods and immunoprecipitated with anti-factor
V MoAb (E9; lanes 1,3, and 51. The supernatants from the immunoprecipitationreactionswere subsequently immunoprecipitatedwith
polyclonal anti-factorV antibody (V; lanes 2,4, and 61. Immunoprecipitated proteins were analyzed by SDS-PAGE. Mock samples did not
receive DNA (lanes 5 and 6). Molecular weight markers are depicted
on the left.
polypeptide migrating at 310 kD (Fig 7, lane 3). In addition,
a heterogeneous smear migrating at approximately 120 kD
was detected and may represent the factor VB8 light chain
that was cleaved at the intracellular processing site within
the factor VI11 B domain at residue 1313. Subsequent immunoprecipitation of the supernatant with an anti-factor V polyclonal antibody detected a significant amount of a 200-kD
polypeptide, likely representing the heavy chain extending
into the B domain (Fig 7, lane 4). Thus. this analysis did
not detect any association of the processed heavy chain with
the processed light chain of factor VB8.
DISCUSSION
Expression of factor VI11 was 10-fold less efficient than
factor V in transfected COS- 1 monkey kidney cells. Because
the A and C domains of factor V and VI11 are 40% identical
and the B domains exhibit little sequence identity, we studied
the influenceof B-domain sequences on expression by analysis of B-domain4eleted and chimeric proteins. In addition,
because the B domains of both cofactors are encoded by
single large exons it is likely they evolved by divergence
from an ancestral exon; thus, it was of interest to evaluate
whether they exhibit similar properties when contained in
chimeric molecules. In this report, we have characterized the
effect of the B domain on expression and functional properties of the secreted factors V and VIII.
Analysis of the levels of mRNA and secreted protein obtained from expression of wild-type factor VI11 and factor
V in transiently transfected COS-l cells provided insights
into factors responsible for the inefficient expression of factor VIII. The steady-state level of factor V mRNA was approximately twofold greater than factor VI11 mRNA. Analysis of primary translation products indicated thatboth
mRNAs were translated with equal efficiency; however, the
factor V translation productwasfivefoldmore
efficiently
secreted into the conditioned medium. The increased secretion efficiency of factor V compared with factor VI11 was
also observed in Chinese hamster ovary cells (our unpublished data), indicating that the differences in secretion were
not unique to the COS- 1 cell transfection system. The difference in secretion efficiency indicates that these two homologous proteins exhibit different requirements for transport
through the secretory pathway. Whereas newly synthesized
factor VI11 significantly binds the protein chaperone BiP
within the lumen of the endoplasmic
recent
studies have shown that factor V expressed in C H 0 cells
does not detectably bindBiP."
In addition, depletion of
intracellular ATP inhibited factor VI11 secretion."" with no
effect on factor V secretion."
Analysis of the expression of chimeric cDNAs having the
B domains exchanged indicatedthatthe levels of mRNA
expressed were intermediate to those of wild-type factor V
and factor VIII, suggesting that sequences within the B domain were partially responsible for the difference in mRNA
levels expressed for factors V and VIII. Deletion of the B
domain within either the factor V or factor VI11 cDNA increased mRNA expression 3.4-and17-fold,
respectively.
This finding shows that in this expression system the Bdomain sequences are detrimental for factor V mRNA expressionand.to
a greater extent. for factor Vlll mRNA
expression. For the wild-type and mutant factor V and factor
VI11 molecules, the steady-state levels of mRNA were proportional to the levels of primary translation products, indicating that there were no significant differences in translational efficiency for the different mRNAs. The factor VI11
chimera containing the factor V B domain was secreted with
a similar efficiency to wild-type factor VIII. Similarly, the
factor V chimera containing the factor VI11 B domain was
secreted with a similar efficiency to wild-type factor V. Thus,
this analysis did not detect any differences in secretion efficiency that could be attributed to unique properties of the B
domains of factors V and/or VIII. In addition, deletion of
the B domain in factor V increased the amount of secreted
protein proportional to the increase in mRNA, showing that
wild-type and B-domain-deleted factor V are secreted with
equally high efficiency. In contrast, the amount of secreted
protein for the B-domain-deleted factor VI11 did not increase proportionally to the increase in mRNA. These results
suggest that regions outside the B domain are responsible
From www.bloodjournal.org by guest on December 22, 2014. For personal use only.
4223
ROLE OF THE B DOMAIN IN FACTORS V AND Vlll
wild-type factor V, suggesting that a portion of the chimeric
for BiP interaction and significantly reduce the secretion
molecules were folded in a functional manner. Because the
efficiency of factor VIII. We are presently constructing addiprotease responsible for factor V activation in RVV cleaves
tional chimeras to identify regions responsible for the reonce after residue 1545," these results suggest that a single
duced secretion efficiency of factor VIII.
cleavage at 1545 can activate the chimeric protein.
Factor V is secreted as a single-chain molecule, whereas
Deletion of the factor V B domain from residue 710 to
factor VIII is proteolytically processed intracellularly at two
1545
resulted in a molecule that was very efficiently exsites within the B domain to yield a heterodimer.4' A factor
pressed but had a sevenfold reduced specific activity comVI11 chimera that had the factor V B domain was secreted
pared with wild-type factor V and was not effectively
as a single-chain polypeptide at a twofold greater level than
cleaved by thrombin to an active form. However, this molewild-type factor VIII. The chimeric protein had a specific
cule was activated by RVV, showing that a portion of the
activity similar to wild-type factor VIII. In addition, there
secreted molecules were folded in a manner that were activawas no detectable difference in thrombin cleavage and actitable and functional after RVV treatment, similar to the facvation, Previous studies on B-domain-deleted factor VI11
tor V B-domain chimeric molecule. Kane et
described
molecules indicated that removal of the B domain altered
the sensitivity to thrombin activation and/or ~ l e a v a g e . ~ ~ , ~a' .factor
~ ~ V B-domain deletion molecule that retained susceptibility to thrombin cleavage and was activatable by RVV.33
For example, factor VI11 B-domain deletion molecules exThe mutant described by Kane et a133 contained an additional
hibited either increased sensitivity to thrombin activation?9
168 amino acids within the B domain that were lacking in
increased sensitivity to thrombin cleavage of the light
the mutant described here. Because these extra amino acids
chain,32or an increased thrombin a~tivation.~'The results
include a region rich in acidic amino acids with three potenreported here show that substitution of the factor V B domain
tial sites of tyrosine sulfation at residues 1494, 1510, and
into factor VI11 restored the wild-type pattern of thrombin
1515,49 wepropose that this region is important for thrombin
activation. We suggest that the context of the thrombin cleavinteraction for cleavage at residue 1545, similar to that obage site is more accurately presented in the hybrid protein
than in the B domain deletion molecules. In addition, beserved for the carboxy-terminal acidic amino acid-rich region of hirudin.50 Consistent with this hypothesis, nonsulcause there is little homology between the factor V and factor
VI11 B domains, these results suggest that there is no unique
fated factor V exhibits delayed thrombin cleavage and
primary amino acid sequence requirement for the B domain
activation, whereas factor Xa cleavage and activation were
and that B-domain sequences in factor VI11 may simply
not affe~ted.~'
The loss of activity for the factor V B-domain
function to separate the heavy and light chains before activadeletion molecule that juxtaposes two thrombin cleavages
tion.
sites at 709 and 1545 are in contrast to observations with the
In contrast to factor VIII, several observations reported
factor VI11 B-domain deletions that juxtapose the thrombin
here support that the B domain of factor V is required for
cleavage sites at 740 and 1689 and yield functional and
factor V structure and activation. First, replacement of the
thrombin activatable molecule^.*^*^^^^^ The results described
factor VI11 B domain for the factor V B domain resulted in
here show that sequences within the B domain of factor V
a molecule with a 6.5-fold reduced specific activity that was
are required for proper thrombin- and, to a lesser extent,
resistant to thrombin activation. The chimeric protein was
factor Xa-mediated activation of factor V. In contrast, RVV
secreted as two forms comprising a single-chain polypeptide
did not exhibit any requirement for sequences within the B
and cleaved and dissociated heavy- and light-chain polypepdomain for factor V activation. This finding suggests that
tides. The reduced specific activity of the chimera was likely
RVV activation does not require the acidic region within the
caused by dissociation of the heavy and light chains. The
factor V B domain (residues 1493-1545). Although there
secretion of dissociated chains suggests that a further reshould be caution from drawing conclusions from mutant
quirement for the B domain of factor V may be to facilitate
proteins, the results from analysis of the factor V B-domain
protein folding in such a way as to promote heavy- and lightdeletion molecule are consistent with those of the factor V/
chain association. However, itisnot possible to rule out
factor VI11 B-domain chimeric molecule and both suggest
that the heterologous factor VI11 B domain affected protein
that the factor V B domain is required for thrombin activation
folding to prevent chain association. The single-chain polybut not for RVV-mediated activation.
peptide of the factor V chimera was sensitive to thrombin
cleavage, but did not display thrombin activation of procoagREFERENCES
ulant activity. In addition, the factor V chimera displayed
1.
Davie
EW,
Fujikawa
K,KisielW:Thecoagulationcascade:
significantly reduced activation by factor Xa. In a detailed
Initiation,
maintenance,
and
regulation.Biochemistry 30:10363,
study of the kinetics of factor Xa and thrombin activation
1991
of human factor V it was concluded that factor Xa cleaves
2. Mann KG, Jenny W,Krishnaswamy S : Cofactorproteinsin
initially after residue I018 andthen after 709, whereas
the assembly and expression of blood clotting enzyme complexes.
thrombin cleaves initially after residue 709 and then after
AnnuRev Biochem 57:915, 1988
1018.25The absence of the cleavage site at 1018 or alteration
3. Kane WH, Davie EW: Cloning of a cDNA coding for human
in the conformation of the cleavage site at 709 within the
factor V, a blood coagulation factor homologous to factor VIII and
chimeric protein may account for the loss of activation by
ceruloplasmin. Proc Natl Acad Sci USA 83:6800, 1986
these proteases. However, RVV treatment increased the ac4. Jenny W,Pittman DD, Toole JJ, Kriz RW. Aldape RA, Hewick
tivity of the chimeric protein to a fold similar to that of
R M , Kaufman
Mann KG: Complete cDNA and derived amino
RT.
From www.bloodjournal.org by guest on December 22, 2014. For personal use only.
4224
acid sequence of human factorV. Proc Natl Acad Sci USA 84:4846,
1987
5. Toole JJ, Knopf JL, Wozney JM, Sultzman LA, Buecker J,
PittmanDD,KaufmanRJ,BrownE,Shoemaker
C, Orr EC,
Amphlett GW, Foster WB, Coe ML, Knutson GJ, Fass DN, Hewick
RM: Molecular cloningof a cDNA encoding human antihaemophilic
factor. Nature 312:342, 1984
6. Vehar CA, KeytB,Eaton
D, Rodriguez H, O’BrienDPO,
Rotblat F, Oppermann H, KeckR, Wood WI, Harkins RW, Tuddenham EGD, Lawn RM, Capon DJ: Structure of human factor VIII.
Nature 312:337, 1984
F W : Structuralmodel of
7. OrtelTL,TakahashiN,Putnamm
human ceruloplasmin based on internal triplication, hydrophilicihydrophobic character, and secondary structure of domains. Proc Natl
Acad Sci USA 81:4761, 1984
8. Stubbs ID, Lekutis C, SingerKL, Bui A, Yuzuki D, Srinivasan
U, Parry G: cDNAcloningofamousemammaryepithelialcell
surface protein reveals the existenceof epidermal growth factor-like
domains linked to factor VIII-like sequences. Proc Natl Acad Sci
USA 87:8417, 1990
9.GuintoER,EsmonCT,MannKG,MacGillivrayRTA:The
complete cDNA sequence of bovine coagulation factor
V. J Biol
Chem 267:2971, 1992
IO. Gitschier J, Wood W1, Goralka TM, Wion KL, Chen EY,
Eaton DH, Vehar CA, Capon DJ, Lawn RM: Characterization
of
the human factor VI11 gene. Nature 312326, 1984
1 1 . Cripe LD, Moore KD, Kane WH: Structure of the gene for
human coagulation factor V. Biochemistry 31:3777, 1992
12. Kaufman RJ, Wasley LC, Domer AJ: Synthesis, processing,
and secretion of recombinant human factor VI11 expressed in mammalian cells. J Biol Chem 263:6352, 1988
13. Weiss HJ, Sussman 11, Hoyer LW: Stabilization of factor VI11
in plasma by the von Willebrand factor. Studies on posttransfusion
and dissociated factor VI11 and in patients with von Willebrand’s
disease. J Clin Invest 60:390, 1977
14. DahlbackB:HumancoagulationfactorVpurificationand
thrombin-catalyzed activation. J Clin Invest 66:583, 1980
15. Kane WH, Majerus PW: Purification and characterization of
human coagulation factor V. J Biol Chem 256:1002, 1981
16. Esmon CT: The subunit structure of thrombin-activated factor
V. Isolation of activated factor V, separation of subunits, and reconstitution of biological activity. J Biol Chem 254:964, 1979
17. Nesheim ME, Myrmel KH, Hibbard L, Mann KG: Isolation
and characterization of single chain bovine factor V. J
Biol Chem
25508, 1979
18. Katzmann JA, Nesheim ME, Hibbard LS, Mann KG: Isolation
of functionalhumancoagulationfactorVbyusingahybridoma
antibody. Proc Natl Acad Sci USA 78:162, 1981
19. Fulcher CA, Zimmeman TS: Characterization of the human
factor VI11 procoagulant protein with a heterologous precipitating
antibody. Proc Natl Acad Sci USA 79:1648, 1982
20. Rotblat F, O’Brien DP, 0-Brien FJ, Goodall AH, Tuddenham
EGD: Purification of human factor VII1:C and its characterization
by Westernblottingusingmonoclonalantibodies.Biochemistry
24:4294,1985
21, Eaton DL, Rodriguez HR, Vehar CA: Proteolytic processing
of human factor VIII. Correlation of specific cleavages by thrombin,
factor Xa, and activated protein C with activation and inactivation
of factor VI11 coagulant activity. Biochemistry 25:505, 1986
22. Lollar P, Parker CC: pH-dependent denaturation of thrombinactivated porcine factor VIII. J Biol Chem 265:1688, 1990
23. Fay PJ, Haidaris PJ, Smudzin TM: Human factor VIIIa subunit structure. Reconstruction of factor VIIIa from the isolated AI/
A3-CI-C2 dimer and A2 subunit. J Biol Chem 266:8957, 1991
24. Pittman DD, Millenson M, Marquette K, Bauer K, Kaufman
PITMAN, MARQUETTE, AND KAUFMAN
RJ: A2 domain of human recombinant-derived factor
Vlll is required
forprocoagulantactivitybutnotforthrombincleavage.
Blood
79:389,1992
25. Monkovic DD, Tracy PB: Activation of human factor V by
factor Xa and thrombin. Biochemistry 29:1118, 1990
26. Krishnaswamy S, Russell CD, Mann KG: The reassociation
of factor Va from its isolated subunits. J Biol Chem 264:3160, 1989
27. Laue TM, Lu R, Krieg UC, Esmon CT, Johnson AE: Ca”dependent structural changes in bovine blood coagulation factor Va
and its subunits. Biochemistry 28:4762, 1989
28. Toole JJ. Pittman DD, Orr EC, Murtha P, Wasley LC, Kaufman RJ: A large region (approximately equal to 95 kDa) of human
factor VI11 is dispensablefor in vitro procoagulantactivity. Proc
Natl Acad Sci USA 83:5939, 1986
29.EatonDL,WoodWI,EatonD,Hass
PE, Hollingshead P,
Wion K, Mather J . Lawn RM, Vehar CA, Gorman C: Construction
andcharacterization of anactivefactor
VI11 variant lacking the
central one-third of the molecule. Biochemistry 25:8343, 1986
30. Meulien P, Faure T, Mischler F, Harrer H, Ulrich P, Bouderbala B, Dott K, SainteMarie M, Mazurier C, Cazenave J-P, Courtney
M, PaviraniA: A new recombinantprocoagulantproteinderived
fromthecDNAencodinghumanfactor
VIII. Protein Eng 2:301,
I988
31. Domer AJ, Bole DC, Kaufman RJ: The relationship
of Nlinkedglycosylationandheavychain-bindingproteinassociation
with the secretion of glycoproteins. J Cell Biol 105:2665, 1987
32. Pittman DD, Alderman EA, Tomkinson KN, Wang JH, Giles
AR, Kaufman, RJ: Biochemical, immunological, and
in vivo funcBlood
tional characterization of B-domain-deletedfactorVIII.
81:2925,1993
33. Kane WH, Devore-Carter D, Ortel TL: Expression and characterization of recombinant human factor V and mutant lacking a
major portion of the connecting region. Biochemistry 29:6762, 1990
34. Sambrook J, Fritsch EF, Maniatis T: Molecular Cloning: A
Laboratory Manual (ed 2). Cold Spring Harbor, NY, Cold Spring
Harbor Laboratory, 1989
35.Sanger F, Nicklen S, Coulson AR: DNAsequencingwith
chain-terminating inhibitors. Proc Natl Acad Sci USA 74:5463,
1977
36.KaufmanRJ:Vectorsforexpression
in mammaliancells.
Methods Enzymol 185:487, 1990
37. Pittman DD, Kaufman RJ: The proteolytic requirements for
thrombin activation of anti-hemophilic factor (factorVIII). Proc Natl
Acad Sci USA 85:2429, 1988
38. Dorner AJ, Kaufman RJ: Analysis of synthesis, processing,
and secretion of proteins expressed in mammalian cells. Methods
Enzymol 185577, 1990
39. Dreyfuss G, Adam SA, Choi YD: Physical change in cytoplasmic messenger ribonucleoproteinsin cells treated with inhibitors
of mRNA transcription. Mol Cell Biol 4:415, 1984
Isola40. Chirgwin JM, Przybyla AE, MacDonald RJ, Rutter WJ:
tion of biologically active ribonucleic acid from sources enriched in
ribonuclease.Biochemistry 185294, 1979
41.Derman E, Krauter K, Walling L, Weinberger C, Ray M,
Darnell JR: Transcriptional controlin the production of liver-specific
mRNAs. Cell 23:73 I , 1981
42. Cleveland DW, Lopata MA, MacDonald RJ, Cowan NJ, Rutter WJ, Kirschner MW: Number and evolutionary conservation of
alpha-andbeta-tubulinandcytoplasmicbeta-andgamma-actin
genes using specific cloned cDNA probes. Cell 20:95, 1980
43. Nesheim ME, Katzmann JA, Tracy PB, Mann KG: Factor V.
Methods Enzymol 80:249, 1981
44.Domer AJ, WasleyLC,KaufmanRJ:Overexpressionof
GRP78 mitigates stress induction of glucose regulated proteins and
blocks secretion of selective proteins
in Chinese hamster ovary cells.
EMBO J 11:1563,1992
From www.bloodjournal.org by guest on December 22, 2014. For personal use only.
ROLE OF THE B DOMAIN IN FACTORS V AND Vlll
45. Pittman DD, Tomkinson KN, Kaufman U: Post-translational
requirements for functional factor V and factor VI11 secretion in
mammalian cells. J Biol Chem 269:17329, 1994
46. Domer A J , Wasley LC, Kaufman RI: Protein dissociation
from GRP78 and secretion is blocked by depletion of cellular ATP
levels. Proc Natl Acad Sci USA 87:7429, 1990
47. Pittman DD, Wang JH, Kaufman U:Identification and functional importance of tyrosine sulfate residues within recombinant
factor VIII. Biochemistry 31:3315, 1992
48. Esmon CT: in Mann KG, Taylor F Jr (eds): The Regulation
of Coagulation. NewYork, NY, Elsevier/North Holland, 1980, p
137
49. Hortin GL: Sulfation of tyrosine residues in coagulation factor
V. Blood 76:946, 1990
50. Maraganore JM, Chao B, Joseph ML, Jablonski J, Ramachan-
4225
dran KL: Anticoagulant activity of synthetic hirudin peptides. J Biol
Chem 264:8692, 1989
51. Pittman DD, Tomkinson KN, Michnick D, Selighsohn U,
Kaufman €U:Post-translational sulfation of factor V is required for
efficient thrombin cleavage and activation and for full procoagulant
activity. Biochemistry 33:6952, 1994
52. Nesheim ME, Pittman DD, Giles AR, Fass DN, Wang JH,
Slonosky D, Kaufman RJ: The effect of plasma von Willebrand
factor on the binding of human factor VI11 to thrombin-activated
human platelets. J Biol Chem 266:17815, 1991
53. Leyte A, van Schijndel HB, Niehrs C, Huttner WB, Verbeet
MPh, Mertens K, van Mourik JA: Sulfation of Tyr1680 of human
blood coagulation factor VI11 is essential for the interaction of
factor VI11 with von Willebrandfactor.JBiol
Chem 266:740,
1991
From www.bloodjournal.org by guest on December 22, 2014. For personal use only.
1994 84: 4214-4225
Role of the B domain for factor VIII and factor V expression and
function
DD Pittman, KA Marquette and RJ Kaufman
Updated information and services can be found at:
http://www.bloodjournal.org/content/84/12/4214.full.html
Articles on similar topics can be found in the following Blood collections
Information about reproducing this article in parts or in its entirety may be found online at:
http://www.bloodjournal.org/site/misc/rights.xhtml#repub_requests
Information about ordering reprints may be found online at:
http://www.bloodjournal.org/site/misc/rights.xhtml#reprints
Information about subscriptions and ASH membership may be found online at:
http://www.bloodjournal.org/site/subscriptions/index.xhtml
Blood (print ISSN 0006-4971, online ISSN 1528-0020), is published weekly by the American
Society of Hematology, 2021 L St, NW, Suite 900, Washington DC 20036.
Copyright 2011 by The American Society of Hematology; all rights reserved.