Mutations of Factor VIII Cleavage Sites in Hemophilia A

1022 Blood, Vol 72, No 3 (September), 1988: pp 1022-1028

By Jane Gitschier, Scott Kogan, Barbara Levinson, and Edward G.D. Tuddenham

Hemophilia A is caused by a defect in coagulation factor

VIII, a protein that undergoes extensive proteolysis during

its activation and inactivation. To determine whether some

cases of hemophilia are caused by mutations in important

cleavage sites, we screened patient DNA samples for

mutations in these sites by a two-step process. Regions of

interest were amplified from genomic DNA by repeated

rounds of primer-directed DNA synthesis. The amplified

DNAs were then screened for mutations by discriminant

hybridization using oligonucleotide probes. Two cleavage

site mutations were found in a survey of 21 5 patients. A

nonsense mutation in the activated protein C cleavage site

at amino acid 336 was discovered in a patient with severe

C LASSIC HEMOPHILIA, or hemophilia A, is an X-

linked bleeding disorder resulting from a defect in

blood coagulation factor VIII. The mature protein, consist-

ing of 2,332 amino acids,”2 serves as a cofactor to activated

factor IX and is stabilized by von Willebrand’s factor

(vWF). Studies on both the plasma-derived protein and the

recombinant DNA-derived protein have demonstrated that

proteolytic cleavages are needed for the activation and

subsequent inactivation of the molecule.36 As shown in Fig I,

the 95 kd B domain is released from factor VIII by cleavages

at amino acid positions 740 and 1648, yielding the 90 and 80

kd fragments. Maximum activity correlates with the appear-

ance of the 50, 40, and 73 kd species, and inactivation with

the appearance of the 43 and 67 kd fragments.5’6 In vitro

mutagenesis has also demonstrated that the arginine residues

at positions 336, 372, and 1689 are important for proper

factor VIII activation and inactivation.7 In addition, the

epitopes for inhibitory antibodies have been mapped to

regions between amino acids 336 and 372 and 1648 and

1689,�” further implicating these sites as vital for factor

VIII function.

Two classes of mutations have been found to cause hemo-

philia A: partial gene deletions and TaqI site mutations.’2’5

Each class constitutes roughly 5% of the hemophilia DNA

samples surveyed. TaqI sites appear to be “hot spots” for

From the Howard Hughes Medical institute and Department of

Medicine, University of California, San Francisco; and Hemostasis

Research Group, Clinical Research Centre, Middlesex, England.

Submitted April 13, 1988; accepted May 12. 1988.

Supported by the Howard Hughes Medical Institute and grants

to E.G.D. T. from North East Thames Regional Health Authority

and the Hemophilia Society.

E.G.D. T. is a member ofthe MRC, Clinical Science Staff J.G. isan assistant investigator with the Howard Hughes Medical Insti-

tute.

Address reprint requests to Jane Gitschier, PhD. Howard Hughes

Medical Institute, U426, Box 0724, University of California, San

Francisco, CA 94143.

The publication costs ofthis 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 toindicate this fact.

© 1988 by Grune & Stratton, Inc.‘WWi-4071/88/7203-0040$3.00/0

hemophilia. In another severely affected patient. a mis-

sense mutation results in a substitution of cysteine for

arginine in the thrombin activation site at amino acid 1689.

This defect is associated with no detectable factor VIII

activity. but with normal levels of factor VIII antigen. The

severe hemophilia in this patient was sporadic; analysis of

the mother suggested that the mutation originated in her

gametes or during her embryogenesis. The results demon-

strate that this approach can be used to identify factor VIII

gene mutations in regions of the molecule known to be

important for function.

a 1988 by Grune & Stratton, Inc.

mutagenesis because the recognition sequence includes a

CpG dinucleotide. The cytosine at CpG is often methylated

in mammalian DNA; a subsequent deamination of 5-methyl-

cytosine leads to an unrepaired transition to thymine. This C

to T transition causes nonsense and missense mutations at

five Taql sites in the factor VIII gene. The nonsense muta-

tions have been found in patients with severe hemophilia, and

the missense mutations in patients with moderate to mild

hemophilia, strongly implicating these mutations as the

cause of hemophilia.’6”7 Further evidence that TaqI muta-

tions are responsible for hemophilia is the coincident appear-

ance of hemophilia and the mutation in sporadic cases of

hemophilia A.’2”4

Codons for four of the six arginine residues present at

cleavage sites contain the CpG dinucleotide and thus are

candidates for cytosine to thymine transitions. The physio-

logical significance of cleavage sites suggests that even

missense mutations in these codons may lead to severe

hemophilia. Based on the TaqI mutation data, we reasoned

that a survey of hemophilia DNA samples should uncover

mutations at some of the arginine codons. We used a

two-step procedure to screen hemophilia DNA samples for

mutations at the four cleavage sites. Short segments of DNA

coding for the cleavage sites were amplified from hemophilia

DNA by the polymerase chain reaction technique.’8 These

amplified sequences were then screened for mutations by

discriminant hybridization of wild-type oligonucleotide

probes. This approach has led to the discovery of mutations

in activated protein C- and thrombin-cleavage sites in hemo-

philiacs.

MATERIALS AND METHODS

Samples. Subjects were hemophilia A patients at the Royal

Free Hospital, London, the University of California, San Francisco,

and Vanderbilt University, Nashville. Blood was collected in heparinor EDTA by venipuncture following informed consent, and DNAwas extracted from peripheral blood leukocytes as described previ-

ously.’9 DNA from two cell lines (4X and AL7) with normal factorVIII genes were used as controls.’�#{176}

Hematological assays. Factor VIII activity (VIII:C) was mea-sured on fresh plasma samples, collected in citrate, by the one-stage

assay as previously described.2’ The international reference prepara-tion supplied by the National Institute of Biological Standards (UK)was used as a primary standard and a pool of 20 normal plasmas was

used as the working standard. Factor VIII antigen (VIlI:Ag) was

For personal use only.on January 29, 2019. by guest www.bloodjournal.orgFrom

I SOkDa � 4OkDa

43 kDa

I� 73kDa

L 67k0a

FACTOR VIII CLEAVAGE SITE MUTATIONS 1023

mutation at codon 1689.

‘

�

II I IllE Al � � A2 I� B I I ‘� I c I C? I

I 90 kOa � 95 kDa � 80 kDa

ACTIVATiON

INACTIVATION

Fig 1 . Proteolytic cleavage sites in factor VIII. Mature factorVIII is a 2332 amino acid protein consisting of three types of

domains: a triplicated A domain (Al . A2. A3). a unique B domain.and a duplicated C domain (Cl and C2). Short acidic regions of

approximately 40 amino acids are located at the Al /A2 and B/A3boundaries. Factor Ila (thrombin) cleaves at amino acid positions372. 740. and 1 689. activating the molecule. Activated protein C

(APC) inactivates factor VIII by cleavage at amino acid 336. FactorXa cleaves at all four of these positions as well as at amino acid1 721 . The protein responsible for cleavage at position 1 648 is

unknown. but this cleavage occurs before or during secretion offactor VlII.7 The acidic spacers are bordered by the cleavage sitearginines at 336. 372 and 1 648. 1 689. Although all six cleavagesites are arginines. the codons for only four of them include the

CpG dinucleotide (336. 373. 1648. and 1689). Amino acid 1indicates the first amino acid of the mature protein. as describedby Wood et al.1 The cleavage products. and their correspondenceto activation and subsequent inactivation. are indicated below theprotein.

measured as described,22 using a polyclonal human antibody to

factor VIII in a one-stage immunoradiometric assay. The standard

used was a pool of 20 normal plasmas assumed to contain I unitVIII:Ag per milliliter-vWF antigen (vWF:Ag) was measured in a

fluid phase immunoradiometric assay using a polyclonal rabbit

antibody23 and the international reference preparation as the stan-dard.

DNA amplification. Short segments of genomic DNA were

amplified in vitro by a modification of the polymerase chain reaction

(PCR) technique.�24 Approximately 10 to 250 ng ofgenomic DNA

was incubated with 10 pmol of each oligonucleotide primer and 1

unit Taq polymerase (New England Biolabs, Beverly, MA) in a 25

�L reaction volume. Final concentration of the reagents in solution

were as follows: 17 mmol/L (NH4)2S04, 67 mmol/L TrisHCl, pH8.8, 6.7 mmol/L MgCl2, 170 �g/mL bovine serum albumin, 200

�mol/L each dGTP, dATP, dTTP, and dCTP. The mixture was

layered with 25 zL mineral oil to prevent evaporation. Following a

two minute denaturation at 94#{176}C,samples were amplified for 35cycles on a DNA Thermal Cycler (Perkin-Elmer Cetus, Norwalk,CT); each round consisted of I 5 seconds at 92#{176}C,one second at50#{176}C,and ten seconds at 70#{176}C.Occasionally a genomic DNA

sample was difficult to amplify. This problem was overcome by a

ten- to 20-fold dilution of the sample before amplification. One-tenth

of the amplified DNA was electrophoresed on a 5% polyacrylamide

minigel for inspection. For the carrier determination, one-tenth ofthe sample was digested in a volume of 30 �L. When samples from

individuals 1115 and 1110 were combined, one-twentieth of each

amplified sample was used.

Oligonucleotide hybridization. As described previously,24 20 �zL

amplified DNA was denatured by adding it to 180 �zL 0.4 N NaOH,25 mmol/L EDTA, and heating for two minutes at 95#{176}C.Sampleswere transfered to ice and distributed among four wells of a slot-blot

manifold onto nylon filters (Zetaprobe; Bio-Rad, Richmond, CA).Filters were hybridized to 32P-labeled 19-base oligonucleotide probes

(I to 2 million cpm/mL) in 5X SSPE, 5X Denhardt’s, and 0.5%

sodium dodecyl sulfate (SDS), for one to 18 hours at 41#{176}Cand

washed in 3 mol/L tetramethylammonium chloride (TMACL)

solution at 61#{176}C.25

Sequence analysis. The amplified 195-bp fragment from 4X

and H4 DNA was purified on a 5% polyacrylamide gel andsequenced directly using the method of Wong et al.26 Primer I 1.2

was end-labeled with gamma-32P-ATP using T4 polynucleotide

kinase. Approximately 0.3 pmol of amplified DNA was denaturedand annealed to 2 pmol of the labeled oligonucleotide by heating themixture to 95#{176}Cfor ten minutes in 50 mmol/L Tris HCI, pH 8.0, 5

mmol/L MgCl2, 50 mmol/L KC1, and 10 mmol/L dithiothreitol,

and then incubating at 37#{176}Cfor two minutes. The sample wassequenced by the dideoxy chain termination method27 using T7

DNA polymerase (Sequenase US Biochemical, Cleveland) accord-ing to the manufacturer’s protocol.

Amplified HlO and 4X control DNAs were digested with SstI andPstI (sites incorporated in the oligonucleotide primers), cloned into

an M I 3 vector, and sequenced using the Ti DNA polymerase

(Sequenase).

RESULTS

Table 1 lists the six arginine residues that are involved in

proteolysis of factor VIII. We have focused our attention on

Table 1 . Factor VIII Cleavage Sites

AminoAcid coiion and

Position Amino Acid Probable Mutation Protein Cleavage Physiological Role Location in Gene

336 CGA arg TGA stop

CAA gIn

APC. Xa Inactivation Exon 8

372 CGC arg TGC CyS

CAC his

Ia, Xa Activation Exon 8

740 AGA arg - Ila, Xa ?, 95 kd release Exon 14

1,648 CGG arg TGG trp

CAG gIn

? ?, 95 kd release Exon 14

1 689 CGC arg TGC CYS

CAC his

Ila, Xa Activation Exon 14

1,721 AGG arg - Xa ?, inactivation Exon 14/15

Probable mutation is the mutation expected by a C to T transition. A question mark indicates that the functional significance is not known. Evidence for

the role of cleavages is presented in Eaton et aIBB and Pittman and Kaufman.7 Amino acid 336 has been determined to be the site of cleavage by activated

protein C by N-terminal sequence analysis of cleaved peptides ID. Eaton, personal communication). Cleavage at this position is also proposed for factor

Xa and thrombin.�7 Sample H4 was found to contain the TGA nonsense codon at amino acid 336 and sample H1O contains the TGC cysteine missense

For personal use only.on January 29, 2019. by guest www.bloodjournal.orgFrom

the residues at positions 336, 372, 1650, and 1689 for several therefore were candidates for factor VIII cleavage site

reasons. First, the codons for amino acids 740 and 1721 do mutations.

not contain the CpG dinucleotide and therefore are unlikely The amplified H4 DNA did not hybridize to probe 1 1.3,

to be hot spots for mutation. In addition, in vitro mutagenesis designed to detect mutations in the activated protein-C

at these two positions had no effect on coagulant activity.7 cleavage site at amino acid 336 (Table 2, Fig 2). The sample

Furthermore, because codons 336 and 372 are clustered in hybridized to the three other probes. H4 DNA was re-

exon 8, and 1650 and 1689 in exon l4,20a11 four sites could be

examined by simultaneous amplification of two small

regions. By this strategy, each of the two target sites within

the amplified region serves as an internal positive control for

the amplification and hybridization of the other site.

The exon 8 and exon 14 regions of 195 bp and 207 bp,

respectively, were amplified with primers 1 1 .1 and I 1 .2, and

I 2.1 and 12.2 (Table 2). These primers anneal to their target

A B11.3 114 12.3 12.4

1 23 4 5 67 89 IOM H3 � #{149} �

.H4146 � #{149}�H7 � #{149}�H8 � #{149}�

207 bp (exon 14I......�

l95eeexon 8) H24 � �

sequences in denatured genomic DNA and direct DNA

synthesis in opposite and overlapping directions. By repeated H10 �

rounds of DNA denaturation, primer annealing, and DNA

synthesis at 70#{176}C,the target sequences, including possible

mutations, are faithfully replicated and amplified several

Hil � �

H12 �

H13

H14 -

hundred thousand-fold. Figure 2 shows the result ofsimulta- H16

neous amplification of the exon 8 and I 4 regions from

hemophilia DNAs.

Aliquots of the amplified DNAs were denatured and

dotted onto four separate nylon filters. Each of the four

filters was hybridized to one of the 32P-end-labeled oligonu-

Fig 2. Screen of hemophilia DNA samples for cleavage sitemutations. In the first step of this two-step procedure. two smallregions of the factor VIII gene are simultaneously amplified fromhemophilia DNA samples. Repeated rounds of oligonucleotide

primer-directed DNA synthesis result in 195-bp and 207-bp frag-

cleotide probes (1 1 .3, 1 1 .4, 1 2.3, and I 2.4; Table 2). The ments containing the cleavage site codons 336 and 372. and 1650

centers of the I 9-base oligonucleotides are positioned on the

CpG of the cleavage-site arginine codons. Following hybrid-

ization, the filters were washed at 61#{176}Cwith a 3 mol/L

tetramethylammonium chloride solution (TMACL).25 Sin-

and 1 689. respectively. The success of the amplification is moni-tored by visualizing the amplification products on a polyacrylamidemini-gel. an example of which is given in (A). Amplified samplesare shown in lanes 1 through 10. DNA amplified from a normalcontrol cell line is shown in lane 1. The H4 and HlO DNAs are

gle-base mismatches can be easily discriminated in TMACL

because only a perfectly matched hybrid of probe and DNA

is sufficiently stable to survive the 6 1#{176}Cdenaturation tem-

perature. A total of 2 1 5 hemophilia DNA samples were

shown in lanes 4 and 9. respectively. and the remaining lanes areother hemophilia samples. HaeIII-digested PhiX 1 74 DNA are themolecular weight standards in lane M. In the second step of the�reening proceciure (B). amplified DNA samples were dotted ontonylon membranes using a �sIot-blot’� manifold. The filters were

amplified and screened for mutations by testing their ability

to hybridize to the wild-type probes. Samples examined

included I 56 patients with severe hemophilia A (< I % nor-

mal factor VIII activity), nine with moderate hemophilia

incubated with �P-labeled. 1 9-base oligonucleotide probes ofwild-type sequence (Table 2). Under stringent washing conditionsof 61’C in a 3 mol/L tetramethylammonium chloride solution. onlyperfectly matched hybrids between probe and amplified DNA arestable. Mutations in the amplified samples result in dissociation of

(1% to 5% activity) and 24 with mild hemophilia (5% to the mismatched hybrid. Probes 11.3. 11.4. 12.3. and 12.4 were

50%). Twenty-nine of the severe patients produce anti-factor

VIII antibodies (inhibitors). The activity level in 26 patients

was unknown. In this population, two amplified DNAs, H4

found to bind stably in all amplified samples except H4 and HlO.implicating codons 336 and 1689 as mutation sites in these

samples. Shown are examples of two such screening experiments.Samples H3-H24 were screened together. and Hl0-H16 were

and H I 0, failed to hybridize to one of the four probes and screened together.

1024 GITSCHIER ET AL

Table 2. Oligonucleotide Primers and Probes

Exon Primers probes

8

14

1 1 . 1 5’GGAAGCTTATGTCAAAGTAG

1 1 .2 5’TAATGTACCCAAGTTTTAGG

1 2. 1 5’CTGAGCTCTCAAAACCCACC

1 2.2 5’CACTGCAGCAATAAAATAGT

1 1 .3 5’CCCCAACTACGAATGAAAA

PQ LR M K N

336

1 1 .4 5’ATCCAAATTCGCTCAGTTG

IQIRSVA

372

1 2.3 5’CGCCATCAACGGGAAATAA

RHQREIT

1648

1 2.4 5’CAGAGCCCCCGCAGClTrC

QSPRSFQ

1689

Arginine positions are ba sad on mature factor VIII amino acid positions as described by Wood et al.’

For personal use only.on January 29, 2019. by guest www.bloodjournal.orgFrom

A

B

C

II

Ill � �4I5

IV#{216}�

1 2

VIII:C% VIII:Ag% VWF:Ag%

1115(Propositus) 0 96 69

1110 (Mother) 74 127 94

113 (Sister) 84 84 89

11158AL7 1115 1110 1110 M

202 bp-

164bp -

;!�-J 1�t1

GATC GATC

CONTROL H1O

FACTOR VIII CLEAVAGE SITE MUTATIONS 1025

amplified using primers 1 1 . 1 and I I .2 and was sequenced

“directly”26 by the additon of 32P-labeled 1 1 .2 primer. Figure

3 demonstrates that the CGA arginine codon is mutated to a

TGA stop codon in the H4 DNA. This nonsense mutation

would result in a truncated protein of 335 amino acids.

Patient H4 has no factor VIII activity and does not produce

antibodies against factor VIII; the level of VIII:Ag was not

determined.

The H 10 amplified DNA sample failed to hybridize to the

probe designed to detect mutations in the thrombin activa-

tion site at position 1689 (probe 12.4, Table 2, Fig 2).Sequence analysis of the H10 amplified sample, presented in

Fig 3, confirmed that the H10 mutation lies in arginine

codon 1689. A C to T transition has resulted in a missense

mutation, replacing an arginine with a cysteine residue. This

mutation also generates a PstI restriction site that can beused for carrier determination (see below). The substitution

of any amino acid for an arginine at this position might beexpected to prevent cleavage by thrombin or factor Xa and

preclude activation of factor VIII. Hematological investiga-

tion of this case demonstrates that, while no factor VIII

activity is detectable in patient HlO, his level of VIII:Ag

approaches a normal value (Fig 4). This unusual activity/

antigen profile is likely to be the consequence of the missense

mutation of the arginine codon at the cleavage site. The

results indicate that the protein is synthesized and secreted at

normal levels, but cannot be activated by thrombin.

The hemophilia in patient H10 is a sporadic occurrence in

a large pedigree (Fig 4), and we attempted to determine

whether the patient’s mother is a carrier. The ratio of factor

�d�ji

�

GATC GATC

CONTROL H4

Fig 3. Sequence of H4 and Hl 0 mutations. Amplified H4. Hl 0.

and control DNAs were sequenced using two different methods.

The 1 95-bp. amplified exon-8 fragments from H4 and controlDNAs were gel-purified and sequenced by the “direct” method of

Wong et ala using oligonucleotide 1 1 .2 to prime DNA synthesis.Comparison of the normal control sequence with the H4 sequence

indicates a base substitution of an A for a G. The sequence aspresented is “anti-sense” and reflects a C to T transition in the

coding strand CGA codon 336 of the activated protein C cleavagesite. Exon-l 4 regions of Hi 0 and control DNAs were amplified.digested with Pad and Ssd. and cloned into an Ml 3 vector beforesequencing. The Hi 0 mutation results from a C to T transition atthe CGC arginine codon 1 689. The missense mutation caused acysteine substitution at the thrombin cleavage site. The Hi 0mutation also generates a Psd site; thus the control and Hi 0sequences diverge after the Pstl site. the sequence in the Hi 0sample being that of the Mi 3 vector. Sequence of two individualH10 clones gave the same result.

Fig 4. Analysis of the HiO pedigree. The HiO pedigree (A)consists of nonhemophilic males (0) and females (0). Only the

patient HiO (lll5. U) has hemophilia. Values for factor VIII activity(VIll:C). factor VIII antigen (VIll:Ag) and von WiIIebrand�s factorantigen (vWF:Ag; also known as factor VIII related antigen) arelisted as a percent of normal values (B). In (C) the 207-bp regioncorresponding to exon-i 4 was amplified from AL7. Ili 0. and llI5

DNAs using primers i2.i and 12.2 and was digested with Psd.Since a Pstl site is present in the oligonucleotide primer 1 2.1 #{149}theamplified DNAs are trimmed to 202-bp by Psd digestion. AmplifiedDNA from the patient (IlI5) is further cleaved to i64-bp and 38-bp(not visible) fragments as predicted from the sequence (Fig 3).Amplified DNAs from a normal control cell line (A17) and from themother (Ill 0) are not reduced to the 1 64-bp fragment by Pstl

digestion. In lane III5 & IIiO. amplified DNA from the patient and hismother were mixed before digestion to insure that the mother’s

amplified sample does not inhibit the restriction enzyme. Lane M

contains Haelll-digested PhiX 1 74 DNA.

VIII activity to vWF:Ag is used to predict carrier status.28 Inthe mother this ratio is 79% (74 of 94), giving a probability of86% that she is a carrier of severe hemophilia.29 For counsel-

ing purposes, it is assumed that two-thirds of mothers ofsporadic cases are carriers of hemophilia, thus reducing herrisk to 57%. The ratio of factor VIII activity to antigen (74 of

127 or 58%) suggests that the mother is heterozygous for the

missense mutation at codon 1 689. Given these data, we were

surprised to discover that the mutation is not present in DNA

For personal use only.on January 29, 2019. by guest www.bloodjournal.orgFrom

1026 GITSCHIER ET AL

isolated for the mother’s peripheral blood. DNAs from the

patient, the mother, and a normal control cell line were

amplified with primers 12. 1 and I 2.2, and the amplified

DNA was then digested with PstI (Fig 4). The patient’s

sample is cut by PstI, reflecting the C to T missensemutation, but the control sample and the mother’s sample

are not cleaved. Thus from the DNA analysis alone, we

would conclude that mother is not a carrier, although she

may be at risk for having another hemophilic son due to

germline mosaicism.’7’�#{176}’3’ Another explanation of the hema-

tological data is that the missense mutation originated

during the mother’s embryogenesis and that she is a somatic

mosaic.32’33 The patient’s sister was not available for DNAanalysis, but her activity and antigen levels indicate that she

is not a carrier of hemophilia.

DISCUSSION

We screened DNA isolated from hemophilia A patients

for mutations in factor VIII cleavage sites using DNA

amplification and oligonucleotide hybridization. Two muta-

tions were detected and both were C to T transitions. In

patient H4 this transition results in a nonsense mutation in

codon 336, truncating the mature factor VIII protein from

2,332 amino acids to only 335 amino acids. Previously

reported nonsense mutations have been observed in the 73-kd

domain, and all are found in severe patients.’2’4 The muta-

tion in patient H4 generates an even smaller peptide and is

undoubtedly responsible for his severe hemophilia.

A more interesting finding is seen in patient H 10 who has

an arginine to cysteine substitution at amino acid 1689, a site

necessary for activation of factor VIII by thrombin. The

patient has no factor VIII activity by coagulation assay, with

or without the addition of thrombin. However, his normallevels of VIII:Ag indicate that secretion, processing, and

stability of factor VIII are likely to be normal. These results

are consistent with those of Pittman and Kaufman7 who have

demonstrated by in vitro mutagenesis that an arginine toisoleucine substitution at position 1689 has no effect on the

level of factor VIII produced, but precludes its activation by

thrombin. Gel filtration studies show that the nonfunctional

VIII:Ag from patient HIO co-elutes with vWF suggesting

that the complex is formed normally (data not shown). A

preliminary experiment using human anti-factor VIII anti-

bodies indicates that factor VIII in patient HlO is not fully

released from vWF on thrombin activation. This result is also

consistent with the conclusions of Pittman and Kaufman34

and Foster et a135 who have demonstrated by deletion muta-

genesis and monoclonal antibody inhibition, respectively,

that vWF binds to the acidic region of the light chain. More

extensive studies using monoclonal antibodies specific for 90-

and 80-kd fragments are in progress to monitor the response

to thrombin and the binding ofvWF in this patient.

There are a number of issues germane to screening DNA

for hemophilia-causing mutations. The first involves the

biochemical characteristics of a particular group of patients.Most severe hemophiliacs have no detectable VlII:Ag, as

measured by an immunoradiometric technique,36’37 and are

classified as cross-reacting material-negative (CRM�5).

Factor VIII gene deletions and nonsense mutations have

been observed to cause hemophilia in some of these patients.

Most moderate and mild hemophiliacs have equivalent levels

of VIII:C and VIII:Ag and are therefore classified as

CRM�C��U�. Mutations that cause inefficient transcription or

translation or partial instability might be the basis of the

hemophilia in the CRM�U� patients. Cross-reacting posi-

tive (CRM”#{176}’)patients are those in whom factor VIII activity

is reduced relative to antigen by at least 30%. CRMIa�

patients are rare; for example only approximately 5% of

severe hemophiliacs are CRM”�.22’374’ In this study of the 34

patients whose VIII:Ag is known, three are CRM’� (includ-

ing patient HlO). Patient HlO is very unusual in having no

detectable VHI:C with normal VIII:Ag since CRM� hemo-

philiacs with normal VIII:Ag generally have 3% to 10%

VIII:C. This last group, the CRM�, are of great interest

since their variant proteins could give clues to sites in the

factor VIII protein essential for coagulant activity.

A second issue in screening for mutations is what parts of

the factor VIII molecule to explore. Factor VIII mRNA isproduced at low level, and its large coding sequence of 7.5 kb

is divided among 26 exons spanning almost 200 kb of

genomic DNA. It is reasonable to examine regions of factor

VIII that have been implicated as important for protein

function, as we have done here. However by limiting the

focus, mutations in domains of unsuspected functional

importance might be overlooked.

A final consideration is selecting a method to identify

these mutations. Mutations in the factor VIII gene were

initially detected by Southern blots, but only gene rearrange-

ments and muations altering restriction enzyme sites could

be detected. The ability to amplify genomic DNA by thepolymerase chain reaction now enables point mutations to be

identified. Currently, there are four alternatives for finding

single-base mutations in amplified genomic DNA: RNAse

cleavage,42 denaturing gradient gel electrophoresis,42 dis-

criminant oligonucleotide hybridization,43 and direct

sequencing.26 RNAse cleavage and denaturing gradient gels

offer the advantage that a large region can be surveyed for

mutations, but the disadvantage that not all mutations will

be detected. In contrast, oligonucleotide hybridization and

direct sequencing offer the advantage that probably all

mutations can be discerned, but that only small regions can

be examined at a time.

The approach we have taken is to screen an unselected

population for mutations in four specific sequences using

discriminant oligonucleotide hybridization. Other strategies

will undoubtedly uncover other types of mutations in the

factor VIII gene. Delineation of mutations that are responsi-

ble for partial or complete inactivation of the factor VIII

protein should provide new insights into what portions of themolecule are required for its activity.

ACKNOWLEDGMENT

We thank Marc Shuman, Marion Koerper, and Susan Karp at the

University of California, San Francisco, Robert ianco, iohn Phil-lips, and Pamela Orlando at Vanderbilt University, Nashville, and

Eleanor Goldman at the Royal Free Hospital, London, for collectingand providing the patient samples. We are grateful to Marie Dohertyfor preparation of oligonucleotides and Kevin Shannon for helpful

comments on this manuscript.

For personal use only.on January 29, 2019. by guest www.bloodjournal.orgFrom

FACTOR VIII CLEAVAGE SITE MUTATIONS 1027

REFERENCES

1. Wood WI, Capon DJ, Simonsen CC, Eaton DL, Gitschier J,Keyt B, Seeburg PH, Smith DH, Hollingshead P. Wion KL, Del-

wart E, Tuddenham EGD, Vehar GA, Lawn RM: Expression of

active human factor VIII from recombinant DNA clones. Nature312:330, 1984

2. Toole ii, Knopf JL, Wozney JM, Sultzman LA, Buecker JL,

Pittman DD, Kaufman RJ, Brown E, Shoemaker C, Orr EC,Amphlett GW, Foster WB, Coe ML, Knutson GJ, Fass DN, Hewick

RM: Molecular cloning of a cDNA encoding human antihaemo-

philic factor. Nature 3 1 2:342, 1984

3. Fulcher CA, Roberts JR. Zimmerman TS: Thrombin proteoly-

sis of purified factor VIII procoagulant protein: Correlation of

activation with generation of a specific polypeptide. Blood 61:807,

1983

4. Rotblat R, O’Brien DP, O’Brien FJ, Goodall AH, Tuddenham

EGD: Purification of human factor VIII:C and its characterization

by Western blotting using monoclonal antibodies. Biochemistry

24:4294, 19855. Eaton D, Rodriguez H, Vehar GA: Proteolytic processing of

human factor VIII. Correlation of specific cleavages by thrombin,

factor Xa, and activated protein C with activation and inactivation

of factor VIII coagulant activity. Biochemistry 25:505, 1986

6. Eaton DL, Hass PE, Riddle L, Mather J, Wiebe M, Gregory

T, Vehar GA: Characterization of recombinant human factor VIII.

J Biol Chem 262:3285, 1987

7. Pittman DD, Kaufman Ri: Proteolytic requirements for

thrombin activation of antihemophilic factor (factor VIII). Proc

Natl Acad Sci USA (in press)8. Fulcher CA, Roberts JR. Holland LZ, Zimmerman TS:

Human factor VIII procoagulant protein: Monoclonal antibodiesdefine precursor-product relationships and functional epitopes. J

Clin Invest 76:1 17, 19859. Shima M, Fulcher CA, Houghten RA, Zimmerman TS:

Localization ofthe binding site for a factor VIII activity neutralizing

antibody to amino acid residues 1663-1669. Blood 70:395a, 1987

(abstr)

10. Ware J, Toomey J, Stafford DW: Localization of a factor

VIII-inhibiting antibody epitope to a region between residues 338

and 362 of factor VIII heavy chain. Proc NatI Acad Sci USA

85:3165, 1988

1 1 . Foster PA, Fulcher CA, Houghten RA, de Graaf Mahoney 5,Zimmerman TS: Localization of the binding regions of a murine

monoclonal anti-FVIII antibody and a human anti-FVIII alloanti-

body, both of which inhibit FVIII procoagulant activity, to amino

acid residues thr35t-ser365 of the FY11! heavy chain. J Clin Invest (in

press)12. Gitschier J, Wood WI, Tuddenham EGD, Shuman MA,

Goralka TM, Chen EY, Lawn RM: Detection and sequence ofmutations in the factor VIII gene of haemophiliacs. Nature 315:427,

198513. Antonarakis SE, Waber PG, Kittur SD, Patel AS, Kazazian

HH, Mellis MA, Counts RB, Stamatoyannopoulos G, Bowie EJW,

Fass DN, Pittman DD, Wozney JM, Toole JJ: Hemophilia A:

Detection of molecular defects and of carriers by DNA analysis. N

EngI J Med 313:842, 198514. Youssoufian H, Kazazian HH, Phillips DG, Aronis 5, Tsiftis

G, Brown VA, Antonarakis SE: Recurrent mutations in haemophilia

A give evidence for CpG mutation hotspots. Nature 324:380, 1986

15. Youssoufian H, Antonarakis SE, Aronis 5, Tsiftis G, PhillipsDG, Kazazian HH: Characterization of five partial deletions of the

factor VIII gene. Proc Natl Acad Sci USA 84:3772, 198716. Gitschier J, Wood WE, Shuman MA, Lawn RM: Identifica-

tion of a missense mutation in the factor VIII gene of a mild

hemophilia. Science 232:1415, 1986

17. Levinson BL and Gitschier J: Unpublished observations,

1987

18. Saiki RK, ScharfS, Faboona F, Mullis KB, Horn GT, Erlich

HA, Arnheim N: Enzymatic amplification of beta-globin genomicsequences and restriction site analysis for diagnosis of sickle cell

anemia. Science 230:1350, 198519. Bell GI, Karam JH, Rutter WJ: Polymorphic DNA region

adjacent to the 5’ end of the human insulin gene. Proc NatI Acad Sci

USA 78:5759, 1981

20. Gitschier J, Wood WI, Goralka TM, Wion KL, Chen EY,

Eaton DH, Vehar GA, Capon DJ, Lawn RM: Characterization of

the human factor VIII gene. Nature 312:326, 1984

21 . Breckenridge RT, Ratnoff OD: Studies on the nature of the

circulating anticoagulant directed against anti-hemophilic factor

with notes on an assay for antihemophilic factor. Blood 20:137,

1962

22. Rotblat F, Tuddenham EGD: Immunologic studies of factorVIII coagulant activity (VIII:C) 1. Assays based on a haemophilicand an acquired antibody to VIII:C. Thromb Res 21 :43 1 , 1981

23. Hoyer LW: Immunologic studies of anti-haemophilic factor

(IV). Radioimmunoassay of AHF antigen. J Lab Clin Med 80:823,

197224. Kogan SC, Doherty M, Gitschier J: An improved method for

prenatal diagnosis of genetic diseases by analysis of amplified DNA

sequence: Application to hemophilia A. N Engl J Med 3 17:985,

1987

25. Wood WI, Gitschier J, Lasky LA, Lawn RM: Base composi-

tion-independent hybridization in tetramethylammonium chloride:

A method for oligonucleotide screening of highly complex gene

libraries. Proc Natl Acad Sci USA 82:1585, 1985

26. Wong C, Dowling CE, Saiki RK, Higuchi RG, Erlich HA,

Kazazian HH Jr: Characterization of beta-thalassaemia mutationsusing direct genomic sequencing of amplified single copy DNA.

Nature 330:384, 1987

27. Sanger F, Nicklen 5, Coulson AR: DNA sequencing with

chain-terminating inhibitors. Proc NatI Acad Sci USA 74:5463,

I 977

28. Zimmerman TS, Ratnoff OD, Littell AS: Detection of car-

riers ofclassic hemophilia using an immunologic assay for antihemo-

philic factor (factor VIII). J Clin Invest 50:255, 197129. Akhmetali MA, Aledort LM, Alexanian 5, et al: Methods for

the detection of haemophilic carriers: A memorandum. Bull WHO

55:675, 1977

30. Bakker E, Van Broeckhoven Ch, Bonten EJ, van de Vooren

Mi, Veenema H, Van Hul W, Van Ommen GJB, Vandenberghe A,Pearson PL: Germline mosaicism and Duchenne muscular dystro-

phy mutations. Nature 329:554, 1987

3 1 . Darras BT, Francke U: A partial deletion of the muscular

dystrophy gene transmitted twice by an unaffected male. Nature

329:556, 1987

32. Maddalena A, Sosnoski DM, Berry GT, Nussbaum RL:

Mosaicism for an intragenic X chromosome deletion in a male with

mild ornithine transcarbamylase deficiency. Am J Hum Genet

41:A227, 1987 (abstr)

33. Gitschier J: An unusual pattern of inheritance of rearrange-ments in the factor VIII gene in a family with hemophilia A. Am JHum Genet 39:A200, 1986 (abstr)

34. Pittman DD, Kaufman Ri: Internal deletions of factor VIII

identify potentially important peptide sequences for binding to vonWillebrand factor. Blood 70:392a, 1987 (abstr)

35. Foster PA, Fulcher CA, Houghten RA, Zimmerman TS: Animmunogenic region within residues valiB7O�glu��W� of the factor VIIIlight chain induces antibodies which inhibit binding of factor VIII to

von Willebrand factor. J Biol Chem 263:5230, 1988

For personal use only.on January 29, 2019. by guest www.bloodjournal.orgFrom

1028 GITSCHIER Er AL

36. Peake IR, Bloom AL: Immunoradiometric assay of procoagu-lant factor VIII antigen in plasma and serum and its reduction in

haemophilia. Lancet 1:473, 1978

37. Lazarchick J, Hoyer LW: Immunoradiometric measurementof the factor VIII procoagulant antigen. J Clin Invest 62:1048,

1978

38. Peake IR, Bloom AL, Giddings JC, Ludlam CA: An immu-

noradiometric assay for procagulant factor VIII antigen: Results in

haemophilia. Br J Haematol 42:269, 1979

39. Muller HP, Van Tilburg NH, Bertina RM, Terwiel JP,Veltkamp JJ: Immunoradiometric assay of procoagulant factor VIII

antigen (VIIICAG). Clin Chim Acta 107:1 1, 1980

40. Reisner HM, Price WA, Blatt PM, Barrow ES, Graham JB:

Factor VIII coagulant antigen in hemophilic plasma: A comparison

of five alloantibodies. Blood 56:61 5, 198041. Holmberg L, Borge L, Ljung R, Nilsson IM: Measurement of

antihaemophilic factor A antigen (VIII CAG) with a solid phaseimmunoradiometric method based on homologous non-haemophilic

antibodies. Scand J Haematol 23:17, 1979

42. Myers RM, Maniatis T: Recent advances in the development

of methods for detecting single-base substitutions associated with

human genetic diseases. CSH Symp Quan Biol 51:275, 198643. Conner BJ, Reyes AA, Morin C, Itakura K, Teplitz RL,

Wallace RB: Detection ofsickle cell beta5-globin allele by hybridiza-

tion with synthetic oligonucleotides. Proc Natl Acad Sci USA

80:278, 1983

For personal use only.on January 29, 2019. by guest www.bloodjournal.orgFrom

1988 72: 1022-1028  

J Gitschier, S Kogan, B Levinson and EG Tuddenham Mutations of factor VIII cleavage sites in hemophilia A 

http://www.bloodjournal.org/content/72/3/1022.full.htmlUpdated information and services can be found at:

Articles on similar topics can be found in the following Blood collections

http://www.bloodjournal.org/site/misc/rights.xhtml#repub_requestsInformation about reproducing this article in parts or in its entirety may be found online at:

http://www.bloodjournal.org/site/misc/rights.xhtml#reprintsInformation about ordering reprints may be found online at:

http://www.bloodjournal.org/site/subscriptions/index.xhtmlInformation about subscriptions and ASH membership may be found online at:

  Copyright 2011 by The American Society of Hematology; all rights reserved.Hematology, 2021 L St, NW, Suite 900, Washington DC 20036.Blood (print ISSN 0006-4971, online ISSN 1528-0020), is published weekly by the American Society of

For personal use only.on January 29, 2019. by guest www.bloodjournal.orgFrom