Detection and characterization of seven novel protein S (PROS) gene

From www.bloodjournal.org by guest on November 18, 2014. For personal use only.
1995 86: 2632-2641
Detection and characterization of seven novel protein S (PROS) gene
lesions: evaluation of reverse transcript-polymerase chain reaction
as a mutation screening strategy
CJ Formstone, AI Wacey, LP Berg, S Rahman, D Bevan, M Rowley, J Voke, F Bernardi, C
Legnani and P Simioni
Updated information and services can be found at:
http://www.bloodjournal.org/content/86/7/2632.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.
From www.bloodjournal.org by guest on November 18, 2014. For personal use only.
Detection and Characterization of Seven Novel Protein S ( P R O S ) Gene
Lesions: Evaluation of Reverse Transcript-Polymerase Chain Reaction as a
Mutation Screening Strategy
By Caroline J. Formstone, Adam I. Wacey, Lutz-Peter Berg, Salman Rahman, David Bevan, Megan Rowley,
Jennifer Voke, Francesco Bernardi, Cristina Legnani, Paolo Simioni, Antonio Girolami, Edward G.D. Tuddenham,
Vijay V. Kakkar, and David N. Cooper
The molecular genetic analysis of protein S deficiency has
been hampered by the complexity of the protein S (PROS)
gene andby the existence of a homologous pseudogene. In
an attempt t o overcome these problems, a reverse transcript-polymerasechain reaction (RT-PCR) mutation screening procedure was developed. However, the application of
this mRNA-based strategy t o the detection of gene lesions
causing heterozygous type I protein S deficiency appears
limited owing to the high proportionof patients exhibiting
absence of mRNA derived from the mutation-bearing allele
(“allelic exclusion“). Nevertheless,this strategy remains extremely effective for rapid mutation detection in type 11/111
protein S deficiency. Using the RT-PCR technique, a G-to-A
transition was detected at position + l of the exon IV donor
splice site, which was associated with type I deficiency and
resulted in bothexon skipping and cryptic splice site utiliration. No abnormal protein S was detected in plasma from
this patient. A missense mutation (Asn 217 t o Ser), which
may interfere with calcium binding, was also detected in
exon Vlll in a patient with type 111 protein S deficiency. A
further three PROS gene lesionswere detected in three patients with type l deficiency by direct sequencing of exon
containing genomic PCR fragments: a single base-pair (bp]
deletion in exon XIV, a 2-bp deletion in exon VIII, and a Gto-A transition at position-1 of the exon X donor splice site
all resulted in the absence of mRNA expressed from the
disease allele. Thus, the RT-PCR methodology proved effective for further analysis of the resulting protein S-deficienl
phenotypes. A missense mutation (Met570t o Thr) in exon
XIV of afurther type Ill-deficient proband was subsequently
detected in this patient‘s cDNA. No PROS gene abnormalities were found in the remaining four subjects, three 01
whom exhibited allelic exclusion. However,the father of one
such patient exhibiting allelic exclusion was subsequently
shown to carry a nonsense
mutation (Gly448to Term) within
exon XII.
0 1995 by The American Society of Hematology.
T
tinguished phenotypically”: Type I deficiency is characterized by reduced total and free antigen levels together with
reduced anticoagulant activity. In type I1 deficiency, protein
S activity is reduced, although total and free antigen levels
are normal, whereas in type I11 deficiency, the total protein
S antigen level isnormalbut free protein S antigen and
activity are reduced. The diagnosis and classification of protein S deficiency are complicated both by the considerable
overlap in protein S antigen levels (free and total) between
protein S-deficient heterozygotes and normal controls” and
by the unreliability of protein S assays (antigen and activity).
The effect of oral anticoagulant treatment and the existence
of acquired protein S deficiency also contribute to diagnostic
uncertainty.” The study and characterization of mutations
underlying protein S deficiency promise to allow the construction of a rational classificatory system as well as improvement of diagnostic procedures.
Previously reported “small” protein S (PROS) gene lesions have been identified by polymerase chain reaction
(PCR)-direct sequencing of exon-containing DNA fragments.””* Such a strategy is rather laborious on account of
the structural complexity of the PROS gene (15 exons) and
complicated technically owing to the existence of a homologous pseudogene that exhibits 97% similarity to the PROS
In this report, we describe seven novel PROS gene
lesions, detected using a combination of two mutation
screening strategies: (1) PCWdirect sequencing and (2) an
alternative mRNA-based strategy.
HE IMPORTANT REGULATORY role of protein S in
hemostasis is reflected by the association of inherited
protein S deficiency with recurrent venous thrombosis.’ This
vitamin K-dependent protein serves as a cofactor to activated protein C (APC), enhancing the rate of inactivation of
factors Va and VIIIa in the coagulation ca~cade.’.~
Protein
S synthesis has been shown in a number of tissues, including
megakaryocytes and liver.637
Protein S occurs in two forms: approximately 60% of
protein S in plasma is noncovalently complexed with C4bbinding protein (C4bBP), a regulatory protein from the classical complement pathway.* The remainder (40%) is unbound (‘ ‘free”). The interaction of functional free protein S
with C4bBP abolishes the former’s anticoagulant activity.’
Three types of inherited protein S deficiency have been disFrom Charter Molecular Genetics Laboratory, Thrombosis Research Institute; the Haematology Department,St Georges Hospital:
Haemostasis Research Group, Clinical Sciences Centre, Hammersmith Hospital, London; the Haematology Department, Luton and
Dunstable Hospital, Luton, UK; Centro di Studi Biochimici della
Patologie del Genoma Umano, Universita di Ferrara, Ferrara; Servizio di Angiologia e Malattie della Coagulazione, Ospedale San
Orsola, Bologna; and Istituto di Semeiotica Medica, Universitci di
Padova, Padova, Italy.
Submitted January 17, 1995; accepted June 5, 1995.
Supported by the Thrombosis Research Trust and Charter plc.
Address reprint requests to David N. Cooper, PhD, Institute of
Medical Genetics, University of Wales College of Medicine, Heath
Park, Cardiff CF4 4XN. UK.
The publication costsof this article were defrayedin 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 1995 by The American Society of Hematology.
0006-4971/95/8607-0029$3.00/0
2632
MATERIALS AND METHODS
Patients
Initial laboratory diagnosis of protein S deficiency was made bq
assay of antigen levels (total and free) using a standard enzyme.
Blood, Vol 86, No 7 (October l ) , 1995 pp 2632-2641
From www.bloodjournal.org by guest on November 18, 2014. For personal use only.
MOLECULARGENETICS
A
2633
OF PROTEIN S DEFICIENCY
C
B
0& @ k i i x O ~ ~ & * ~ t :
T-14%
@
*b
*
T-70%
F-16%
D
M
E
1-
-
AG F-30%
T-74%
F-%%
A49%
F-18%
,
*Il"o
J- /
G
6 F-13%
F
* * IA
QE?%
AG
-
!
~
6
~
F-B%
AG
A allele lost
66
G allele lost
"p' *!IOJpTfg
A allele lost
T-64%
F-35%
H
G allele &C
M
[
1c
: GF-1
Fallele
M
~
I
I%"JLJ"
oflp
I%
%g%"2%
I
w
p
u
up^
I0
Gg$bTb
h'
F - w +
F-l6
A-58%
F-50%
NT
._.
AG
T-B%
" i t l F 4 9 %
F49%
A49%
wi"4"s0 I * F 0
0 ;:?:%bp
F??
or E:00
T-13&
F-S%
T-75%
-
F-10%
T-21%
< l296
F- GT-37%
allele
lost F-24% F-10%
AG
F-11%
0666
O
W bb6
G allele lost
F-=% F-17%
M
F-%%
F-11%
AG
Fig 1. Family pedigrees of patients 1 through 10 (A through J, respectively). Pedigree (B) has been published previously.m Semi-shaded
symbols denote clinically affected individuals. T, total protein S antigen. F, free protein S antigen. A, protein S activity. *Proven carrier of
mutation. (0)Proband studied by RT-PCR, PCRldirect sequencing of PROS gene exons, and Pro626 RFLP analysis ofplatelet cDNA. ( ) Patient
on anticoagulant therapy. AG denotes heterozygos" at Pro626 locus. AAlGG denote homozygosity at Pro626 locus. NT, not tested by PCRl
direct sequencing analysis. Data on allelic exclusion are also provided. C4b-BP levels for family C ware within thenormal range.
+
linked immunosorbent assay (ELISA, Dako, High Wycombe, UK)
and a rabbit polyclonal antibody. Protein S activity was measured
using the Stago kit (Diagnostica Stago, Asnitres, France). The hereditary nature of the deficiency state was established by identification
of additional family members with abnormally low levels of protein
S antigen (total and/or free) and activity. Twenty probands with an
inherited protein S deficiency and a personal and/or family history
of venous thrombosis were initially chosen for study. Ten of these
were studied in detail (Fig 1).
Plasma
Blood was collected in a tube with 1/10 v01 3.8% sodium citrate
and centrifuged at 2,OOOg for 20 minutes. The plasma was stored at
-80°C until use. Pooled normal plasma was prepared from eight
normal individuals.
Strategies for Mutation Screening and Analysis
Southern blot analysis. Southern blotting was performed as described." DNA was restricted with Sst I (Promega, Southampton,
UK) as recommended by the manufacturer. Bands homologous to
the PROS gene and its pseudogene were visualized using a full-
lengthPROS cDNA probe (gift from P.H. Reitsma, Leiden, The
Netherlands).
Pro626 RFLP analysis. Exon XVi3 and cDNA fragment PS9
(Table 1) were PCR amplified as described and digested with BstXI
(Boehringer Mannheim, Mannheim, Germany) according tothe
manufacturers' instructions. Restriction products were visualized as
for PCR fragments. The exon XV PCR fragment migrated at 415
bp.BsfXI restriction of an exon XV PCR fragment carrying the
Pro626 CCA allele yielded restriction fragments of 180 bp and 225
bp. An exon XV PCR fragment carrying the Pro626 CCG allele was
unaltered by BsfXI restriction and remained at 415 bp. Similarly for
PS9, the PCR fragment migrated at 180 bp. BstXI restriction of a
CCA-bearing PCR fragment yielded fragments of 130 bp and 50
bp.
Reverse Transcriptase-PCR (RT-PCR)
Platelets from protein S4eficient patients were separated from
whole blood" and RNA extracted from these ~ e l l s . Specific
~ ~ . ~ ~reverse transcription of platelet mRNA was performed using the Superscript RNase H RT kit (GIBCO-BRL, Middlesex, UK). Two distinct
cDNA templates were transcribed (PROS-specific primers 15B and
RevII; Fig 2, Table 1) from mRNA to avoid the possibility of ineffi-
From www.bloodjournal.org by guest on November 18, 2014. For personal use only.
2634
FORMSTONE ET AL
Table 1. Oligonucleotide Primers for RT of PROS cDNA, Amplification of First- (PSA-C) and Second-Round (PSI-9)PCR Fragments, and
Amplification Of the Donor Splice Siteof Exon 4, Exon 14, and the5' Region of the Human PROS Gene
otide
Sequence
Oligonucleotide
Primer Fragment
or
cDNA(i)
cDNA(ii)
A
B
C
PS1
PS2
PS3
PS4
PS5
PS6
PS7
PS8
PS9
4
PSPl5
14
"C
5'
Rev II
5' AACTGCTCCGCCAAGTAA 3'
15Bt
5' GATAGCAAGAGAAGTAAGAATTTC 3'
PSlNP
CGCCTCCGCGCCTTCGAA
-18
3'
PS3A
5' AGCTGTGCCACAAATGCTTGG 3'
PS3
5' CCATTCCAGACCAGTGTAGT 3'
PS6
5' ATCCATCTAGACGAGGGT 3'
PS6A
5' TGGCGGAGCAGTTTGCAGG 3'
ATAAGCAGAGAAAAGATGCC
Rev
5' end
3'
PSlNP
5' CGCCTCCGCGCCTTCGAA3'
PSlA
5' AATTTGCACGACGCTTCCTA 3'
PSZN
5' TTCCCGTCTCAGAGGCAAAC 3'
PS2A
5' TTCTCCTTGCCAACCTGGTT 3'
PS3A2
5' CTAAGAAGCTGTGTCAATG 3'
PS3N
TCAAAGAGCATKATCCACAT
5'
3'
PS4A
5' AGTTACCACTGTTCCTGTAA 3'
PS4N
GTTCAAGGGAAGGCACACTG
5' 876
3'
PS5N
5' GCTCAGCTTTGTGTCAATT 3'
PS5A
5' GTGATCGATAGATTCTGCGT 3'
PS6A
5' TGGCGGAGCAGTTTGCAGG 3'
PS6
5' ATCCATCTAGACGAGGGT 3'
5' GAAGCTGTGATGGATATA 3'
PS7A
5' CAGATGAGATTGTTGATC 3'
PS8A
5' CTGGTAACAACACAGTGCCCT 3'
PS8
5' TTCACTTCCATGCAGCCATT 3'
PS9A
5' TCCATTCAGTGCCACACCAGT 3'
ATAAGCAGAGAAAAGATGCC
Rev
5' end
2051 3'
Exon
Splice
TAGTTTATATTACCATGG
site
5' 3'
AH4
5' to
ATG
PSP23
codon
PSP35
PSP43
Exon
DC1
DC4
916 to 899
205 to 182t
to 1
-
645 to 625
347 to 366
1345 t o
905 t o
2051 to
--- 18 to
130 to 111
56 t o 75
456 t o 437
328 to
622 to
540 to
to
54 757 to
1038 to
905 to
1345 t o
1204 to
1725 t o
1583 t o
1928 to
1875 to
-95P
371V5' TGGAAGTTGTCTTGACCAGT
to
-391
3'
5' AGGAGACCGCCCGCTCCCAG
-160
3'
5' CTATCACCATAGTTCTTCC
-514
3'
5' AGACATCCTTCTTCACCA 3'
5' AACTCAAAAGTCACTCTTAA 3'
5' AAATATTATCGGTTTGAT 3'
1328
923
2032
1
346
602
560
857
715
1019
923
1328
1221
1708
1603
1909
1895
to 2032
to
56 to 7 5 t a
-
60
50
50
60
62
62
56
52
52
58
58
48
60
-179t'
to
t o -496$"
-240 t o -257$"
-53 to -73*b
43 to 60*b
52
49
* Nucleotide numbering of cDNA primers is derived from the published PROScDNA
sequence4' except where indicatedtS. Position1 denotes
A nucleotide of ATG (Met).
t Data from Reitsma et al.13
Numbering of position for genomic DNA primers is relative to (a) the
5' or 3' boundary of the exon 4 invariant GT of the donor splice site
(b) the 5' or 3' boundary of the donor or acceptor splice sites of exon 14, respectively, and (c) theATG translation initiation codon.
*
cient reverse transcription whichwouldleadtothe
production of
less than full-length cDNA species. RevII cDNA comprised only
exons I to IX of the gene (Fig 2).
Two successive rounds of PCR amplification (Fig 2) were used
to generate nine overlapping cDNA fragments, which together
spanned the entire PROS gene coding region. PCR primers were
specifically chosen to allow identification of aberrant mRNA splicing
involving any one of the PROS gene exons. The three large overlapping first-round PCR cDNA fragments PSA, B, and C (Fig 2) were
used as templates to generate the smaller second-round fragments
PSI-3, PS4,5, and PS6-9, respectively. PSA was amplified using
RevII cDNA as template. PSB and C were bothamplified from
cDNA generated using primer 15B. The primer sequences used and
their locations are presented in Table I .
PCR was performed in a total volume of 100 pL containing template DNA (typically 3 pL of reverse transcription reaction [cDNA
products derived from 2 pg RNA] or I to 5 pL first-round PCR
product), 500 ng appropriate primer, 100 p m o L each dNTP (Pharmacia, Uppsala, Sweden), 2 U Tuq DNA polymerase (Promega),
and Tuq polymerase buffer (1.5 mmoVL MgCI2; Promega). PCR
reactions were performed using either a Perkin Elmer-Cetus (Beaconsfield, UK) or Hybaid (Teddington, UK) Omnigene thermal cycler. Forty cycles of PCR (denaturation at 94°C for 1 minute, annealingat the appropriate temperature for I minute,and extension at
72°C for I minute) were then followed by a final elongation step at
72°C for 4 minutes. Templates for amplification of PSA and PSI
were initially denatured at 99°C for
15minutes, followed by 2 minutes at 94°C. At this point, dNTPs and Tuq DNA polymerase were
added. Annealing temperatures for each PCR reaction are presented
in Table 1. Second-round PCR fragments were visualized byUV
transillumination of ethidium bromide-stained agarose gels (2%).
Purification ofPCR products was performed using Geneclean I1
(BiolOl, Vista, CA). Purified PCR fragments were sequenced using
either the Sequenase kit (Amersham, Amersham, UK) or a cycle
sequencing system (GIBCO-BRL). For cycle sequencing, PCR products were purified by excision from a 1 % low-melting point agarose
gel and then added directly to the sequencing mix.
PCWdirect sequencing of exon-containing DNA fragments.
Genomic DNA was extracted from whole blood*' and DNA amplification was performed essentially as described." PCR amplification
From www.bloodjournal.org by guest on November 18, 2014. For personal use only.
MOLECULARGENETICS
UT PRE
nR TSR
EGF
PWGLP,
I
a
n
~
I
II
111 IV
2635
OF PROTEIN S DEFICIENCY
WJT
SHBG
II Ill
IV
a
n
v VIIVlll
VI
c
b
a
n
x
IX
XI
e
d
~
Q
f
a
~
m
m
Rev 1
I
PS A
160
PS c
FIRST
ROUND
PCR
PS B
l48
380
295
~
xv
XI1 XIV
Xlll
of factor X (A. Tulinsky, personal communication, November 1994)
as described previo~sly?~Residues (311120.5 (E205) and Am217
(N217) were rotamerized to reproduce the geometry of the analogous
calcium-binding side-chain carbonyl groups of Gln49 (449) and Hya
63 (P-hydroxylated Asp) of the first EGF domain of factor X. A
calcium ion was then introduced and orientated into the putative
binding pocket of the model.
~
n
r
r
346
-__
- _
ROUND
336
253
U1
522
PS2
PS3
PS.
PS5
PS6
SECOND
PS,
~
175
"
RESULTS
PCR
PS1
a
PS8
PS9
Southern blot analysis of genomic DNA derived from the
20 protein S-deficient patients under initial study failed to
Fig 2. Schematic representation of the RT-PCR strategy for the
show any gross abnormalities of the PROS gene.
detection of mutations in the PROS gene; shown are the location of
primers used for cDNA synthesis and the products of first-(PS A
In an attempt to avoid the laborious and technically probthrough Cl andsecond-round IPS 1 through 91 PCR amplification.
lematic method of screening each individual exon of the
The structure of the human protein S (PROSI gene coding region,
PROS gene, a two-step mRNA-based protocol was devised
comprising exons I through XV, is shown: 5'UT. 5' untranslated reand used to screen the 20 protein S-deficient patients. Table
gion; PRE, prepeptide; PRO, propeptide; G M , Gla domain; HR. hydrophobic region; TSR, thrombin-sensitive region; EGF, epidermal
2 (patients 1 , 2, 3, and 8) summarizes the results of this
growth factor-like modulesI through W
; SHBG, sexhormone binding
analysis.
globulin-like domains athrough g; 3'UT, 3' untranslated region. (I)
Positions of alternative primers for cDNA synthesis. Fragment sizes
are given in bp.
First-Stage RT-PCR
of overlapping DNA fragments encompassing DNA sequence 515
bp upstream of the translation initiation codon (ATG) was performed
using primers PSP35, PSP43 and PSP15, PSP23 (Table 1). DNA
template was initially denatured as for fragments PSA andPS1
above. Exons IV and XIV were amplified using primers PS4-5I3
and AH4 and DC1 and DC4 (Table l), respectively. Annealing
temperatures were as shown in Table 1. PClUdirect sequencing was
performed as described for RT-PCR. Mutations detected were confirmed by direct sequencing of the other strand.
Screening for factor V Leiden. APC resistance assays and
screening for the common factor V Leiden lesion (CGA to CAA
converting Arg506 to Gln) were performed essentially as de~cribed.~'~"
Immunoblot Analysis of Protein S
Protein S was partially purified from plasma essentially as in the
method of Friedberg and Pizzo2'except that 20 mmoVL benzamidine
(Sigma, Poole, Dorset, UK) was included in BaCI, precipitation and
washing procedures. The washed pellet was dissolved in 0.1 molL
Tris-HCI, 0.1 mom EDTA, 20 mmoVL benzamidine (pH 7.3, and
dialyzed overnight against 50 mmol/LTris-HC1, 0.1 m o a NaCl
( ~ H 7 . 3 Purified
. ~ ~ protein S (Calbiochem, Nottingham, UK) and
both pooled control and patient BaClz precipitated protein S fractions
were subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE)30using a 10% separating gel and transferred to nitrocellulose. Protein S was identified using a monoclonal
anti-protein S antibody (HS-4) (Sigma) and the presence of abnormal
protein S was evaluated using a polyclonal protein S antiserum
(Sigma).
Molecular Modeling of the Fourth EGF-like Domain of
Protein S
Molecular modeling of the fourth EGF-like domain of protein S
was performed on a Silicon Graphics Indigo Elan R4000 (Mountain
View, CA) using the BIOSYM software suite (BIOSYM Technologies Inc, San Diego, CA) and Ribbons 2.5 (M. Car~on).~'
Template
structures were derived from the calcium-binding residues of the
factor X first EGF-like domain (PDB accession number 1CCF32)
and the serine proteinase binding region of the second EGF domain
Nine overlapping PROS cDNA fragments PS1-9 (Fig 2)
were PCR amplified individually. For 18 of the 20 individuals, PS1-9 were of the expected size (380 bp). However, for
patient 1 two abnormally sized PS2 fragments of 428 bp and
294 bp were observed in addition to the 380-bp wild-type
fragment (Fig 3, tracks 2, 3, and 4). Some variation in the
extent of amplification from different PSA templates of the
smaller 294-bp PS2 fragment, and occasionally of the wildtype PS2 fragment, wasnoted. However, an intense PS2
band at 428 bp was consistently observed irrespective of the
fragment A template used. PS2 amplified directly from patient l Rev11 cDNA confirmed that this larger abnormal fragment at 428 bp was the predominant aberrant mRNA species
(Fig 3, track 5). The smaller 294-bp PS2 band was detected
at a significantly lower level than either the wild-type band
or the larger abnormal band. Second-round PCR amplification of fragment PS3 showed the existence ofan aberrant
Table 2. Summary of Mutations Detected in 11
Protein S-Deficient Individuals
Mutation Screening Strategy
RT-PCR
Patient
1
2
3
492 + 1 G + A
AAT AGT Asn217 + Ser
+
?
1908 del AC (Thr547/Pro548)
ACG (Met570 Thr)
1301 - l G + A
927 del G (Gly220)
4*
5
ATG
6'
7*
'
8
9*
9a
10'
Exon Sequencing
Insertion in intron 3?
4
+
?
?
GGA 4 TGA (Gly448
+
Term)
?
Mutations are designated by reference to the PROSgene sequence
presented by Schmidel et alZoaccording to the nomenclature recornmended by Beaudet and Tsui." Patient 9a is the father of patient 9.
* Exhibit allelic exclusion.
From www.bloodjournal.org by guest on November 18, 2014. For personal use only.
FORMSTONE ET AL
2636
1
2
3
4
5
PS3 band at 343 bp in addition to the wild-type (Fig 3, tracks
6 and 7).
PCWdirect sequencing of the two aberrant PS2 cDNA
fragments from patient 1 showed the loss of the 87-bp exon
IV from the smaller 294-bp fragment and inclusion of an
extra 48 bp of DNA sequence from the 5’ end of the intervening intron D, interposed between exons IV and V (data not
shown). Inspection of the sequence of the donor splice site
of exon IV within the 428-bp PS2 fragment showed a heterozygous G to A transition within the invariant GT dinucleotide. The abnormal PS3 fragment was also found to contain
a 48-bp intronic sequence identical to that observed in PS2
and inserted at the same position between exons IV and V
(data not shown). The sequences of the normal-sized PS2
and PS3 fragments were identical to that of the wild-type.
Confirmation of this putative PROS gene lesion was obtained
by PCWdirect sequencing of genomic DNA from patient 1
across the donor splice site region (data not shown). Inspection of the published PROS gene sequence” indicated the
presence of a cryptic splice site (taggtaaa) in intron D, 48
bp 3’ to the normal donor splice site. The first three bases
of this cryptic splice site (tag) were present at the junction
between the inserted intronic sequence and exon V in both
the aberrant 428-bp and 343-bp PS2and PS3 fragments,
respectively. This PROS gene lesion cosegregated with the
type I protein S-deficient phenotype in the family (Fig IA).
PS2 amplification from a second patient (patient 8, Fig
IH) also showed an abnormal band of slightly larger size
than wild-type (data not shown). PCWdirect sequencing of
this aberrant PS2 fragment showed the presence of a stretch
of DNA sequence between exons 111 and IV, the origin of
which remains unclear. The molecular basis of this rearrangement and its possible causative role in the deficiency
state are under further study.
Second-Stage RT-PCR
The second step in the RT-PCR mutation screening strategy used direct sequencing of nine PCR-amplified overlapping cDNA fragments. However, expression of both PROS
mRNA alleles is a requirement for successful mutation detection using this approach. To assess allele-specific mRNA
6
7
Fig 3. Agarose gel (2%) of PS2 (tracks 2 through
5) and PS3 (tracks 6 and 7) cDNA fragments. cDNA
synthesis was performed using Redl primer. Firstround PCR primers were PSlNP and PS3A. Primers
for PS2 and PS3 fragment PCR amplification were
PS2, PS2N and PS3N. PS3A2, respectively. Track
2
contains a PS2 fragment of the expected size (380
bp). Tracks 3 and 4 are second-round PS2fragments
amplified from different patient 1 fragment A templates and contain aberrantly sized DNA fragments.
Track 5 also shows PS2 PCR fragments amplified directlyfrom patient 1 Rewil cDNA. Tracks6 and 7 show
fragment PS3 amplified from control and patient 1
fragment A, respectively. For complete explanation,
see text. Track 1 is a d, X174/Hadll size marker. Numbers for size markers are 1353, 1078, 872, 603, 310,
281, and 271 bp when reading down from the top of
the gel.
expression in each of the 20 protein S-deficient individuals,
a silent restriction fragment length polymorphism (RFLP) at
Pro626 in exon XVof the PROS gene“4 wasused. This
variant represents the only reported RFLP within the PROS
coding region. However, only eight probands were heterozygous for the Pro626 RFLP at the genomic DNA level (patients 2 through 4, 6 through 10; data not shown). Of these
eight probands, only two (patients 2 and 3) were heterozygous at the cDNA level, ie, they exhibited mRNA expression
from both PROS alleles (mRNA analysis for patient 2 has
been reported p r e v i o ~ s l y ~The
~ ) . other six individuals exhibited loss ofmRNAfrom one allele (“allelic exclusion”).
Analysis of platelet cDNA from five normalcontrols heterozygous for the CCA/CCG polymorphism indicated that the
two alleles were normallyand codominantly transcribed
(data not shown). Patients 2 and 3 were subsequently analyzed by second-stage mutation screening.
PCWdirect sequencing of fragment PS4 amplified from
patient 2 cDNA showed an AAT to AGT transition within
exon VI11 of the PROS gene converting Asn217 to Ser (data
not shown). Confirmation of this missensemutation was
obtained by PCWdirect sequencing of exon VI11 from patient
2 genomic DNA. This lesion served to create an Mae111
restriction site (recognition sequence, GTNAC). Cosegregation of this lesion with the type 111 protein S deficiency in
this familyz4was confirmed by screening 23 family members
by Mae111 digestion; eight individuals were found to be heterozygous for the missense mutation. PCWdirect sequencing
of the remaining coding sequence of the patient’s PROS
genes failed to detect any other mutational lesion. No mutation was detected in the other patient (patient 3).
PCWDirect Sequencing of Exon-Containing DNA
Fragments
mRNA analysis of 20 protein S-deficient patients showed
the limited applicability of the RT-PCR technique to the
screening of inherited PROS gene lesions. Because PCW
direct sequencing of the PROS genes from the remaining
17 patients would clearly have been laborious, only certain
individuals were included in this nextphase of mutation
screening. Selection was made principally on the basis of
From www.bloodjournal.org by guest on November 18, 2014. For personal use only.
2637
MOLECULARGENETICS OF PROTEIN S DEFICIENCY
allelic exclusion, since the presence of a PROS gene lesion
could be inferred in these patients. The absence of mRNA
from one PROS allele was noted in six of the eight probands
(patients 4 , 6 through 10; Table 2) with CCNCCG heterozygosity. All six individuals manifested a type I deficiency.
One further proband (patient 5 ) was also included in this
analysis as part of a family study involving a case of compound heterozygosity.
PCR amplification of exons I through XV from patients
4 through 10 generated fragments of the expected size.I3
Table 2 summarizes the results of the subsequent PCWdirect
sequencing analysis (patients 4 through 10). Patient 7 exhibited a single base-pair deletion involving a G nucleotide in
codon Gly220 (exon VIII). This deletion creates an EcoRII
site [t CC(m)GG]. Cosegregation of this lesion with the
clinical phenotype in the affected family was confirmed by
EcoRII restriction enzyme digestion of exon VI11 PCR fragments from two family members (Fig 1G). A further deletion
of an AC dinucleotide, encompassing Thr547 andor Pro548
(exon XIV) of patient 4, was shown in one other affected
family member (Fig 1D) by PCWdirect sequencing analysis
of exon XIV amplified from the genomic DNA of this individual. Sequence analysis of PS4 and PS8, PCR amplified
during the RT-PCR study from the cDNA of patients 7 and
4, respectively, showed the absence, in both individuals, of
the deletion-bearing PROS allele (data not shown).
A G-to-A transition at position -1 (nucleotide 130lZ0)of
the donor splice site of exon X was detected in patient 6
and PCWdirect sequencing was again used to confirm the
presence of this lesion in two other affected family members
(Fig 1F). PCR amplification of fragment PS6 from patient
6 cDNA, again performed previously as part of the RT-PCR
study, had not shown the existence of an aberrant cDNA/
mRNA species.
Finally, patient 5 was shown to carry an ATG to ACG
transition in exon XIV converting Met570 to Thr. PCWdirect
sequencing of cDNA fragment PS8, amplified from patient
5 cDNA during the first stage of the RT-PCR protocol,
showed the existence of this missense mutation at the mRNA
level. PCWdirect sequencing was used to confirm the presence of this lesion within exon 14 from one other protein
Sdeficient family member (Fig 1E). Therefore, patient 5
has inherited a PROS gene lesion from his father not his
mother. PCWdirect sequencing of the remaining 14 PROS
exons failed to show a further PROS gene mutation.
The phase between the disease-associated allele, as determined by Pro626 RFLP analysis of genomic DNA from
different family members, and the mRNA species exhibiting
allelic exclusion (patients 4,and 6 through 10) was established by reference to the PS9 Pro626 cDNA RELP data
(Fig lD, F, G, H, I, and J).
In patients 8, 9, and 10, no abnormality within the PROS
gene sequence was identified. PCWdirect sequencing of a
patient 8 exon 11-containing PCR fragment showed a 3-bp
(tct) deletion at nucleotide position 74 bp 5’ to the acceptor
splice site of exon I1 [Intron A; 223-74 del TCT]. However,
this deletion didnot cosegregate with the protein S deficiency phenotype in the affected family (Fig 1H) and presumably represents a neutral polymorphism. DNA sequence
up to 515 bp 5‘ to the PROS transcription initiation sitezo
was PCR amplified from the genomic DNA of patients 8, 9,
and 10 and sequenced directly. No deviations from the wildtype DNA sequence were detected. For both patients 9 and
10, the CCG PROS gene allele, which was absent at the
mRNA level, did not cosegregate with the protein S-deficient
phenotype in other family members (Fig l , I and J). PCW
direct sequencing of 15 PROS gene exons amplified from
the genomic DNA of the father of patient 9 showed the
presence of a nonsense mutation (GGA to TGA) within exon
XI1 (Table 2, 9a). This Gly448 to Term conversion was
consistent with the type I deficiency state in this individual
(Fig 11). cDNA analysis was not undertaken for this individual. No further family members were available for genetic
analysis. Conversely, the sister and niece of patient 10 (Fig
1J) carried no abnormalities within the PROS gene sequence.
Factor V Leiden Screening
Because APC r e ~ i s t a n c ecould
~ ~ increase the probability
of protein S-deficient individuals coming to clinical attention, the presence of APC resistance in the 10 patients studied in detail was excluded either by screening for the causative factor V Leiden mutation directly (patients 1 and 3
through 10) or by means of the APC resistance assay (patient
2).
Analysis of Plasma Protein S From Patient I
SDS-PAGE of pure human protein S and BaCl, precipitated fractions of pooled control and patient plasma under
reducing conditions was followed by Western blot analysis
using a protein S monoclonal antibody (HS-4). A
migrating at 86 kD and 76 kD was revealed in all cases (data
not shown). A similar pattern was observed with a protein
S polyclonal antiserum. No abnormally sized band(s) were
detected in patient plasma (data not shown).
DISCUSSION
Using a combination of two mutation screening protocols,
seven novel PROS gene lesions were identified during the
detailed study of 10 unrelated protein S-deficient patients.
The RT-PCR strategy was originally adopted to avoid the
potential problems stemming from the presence of the protein S pseudogene” (this homologous sequence is not transcribed37),and to circumvent the laborious task of sequencing all 15 PROS exons individually. The protocol was
devised to allow the identification of intronic mutations that
might alter mRNA splicing as well as mutations within the
coding sequence of the PROS gene. Indeed, the first PROS
gene lesion to be identified, a point mutation that abolished
the donor splice site consensus sequence of exon IV, was
detected initially by PCR amplification ofan aberrant PS2
cDNA fragment. Thus, the effectiveness of the RT-PCR
strategy for the detection of splice site mutations that do not
influence the abundance of PROS mRNA was realized early
in the analysis.
An initial group of 20 protein S-deficient individuals
(mainly type I) was chosen for mutation screening analysis.
Progressively, however, during the different phases of the
From www.bloodjournal.org by guest on November 18, 2014. For personal use only.
2638
screening process, certain individuals were singled
out for
further detailed study. Obviously, expression of the mutation-bearing allele is necessary for successful mutation detection when using the RT-PCR technique. At present, the
Pro626 RFLP provides the only available means to determine allelic heterozygosity at both genomic DNA and cDNA
levels, thereby demonstrating the existence of mRNA species fromboth PROS alleles. From our studyof 20 protein Sdeficient individuals, only 8 were found to be heterozygous
(CCNCCG) at the Pro626 position. Further reports of informative intragenic PROS RFLPs in linkage disequilibrium
with Pro626 should ease the
identification of potentialcandidates for mutation analysis using RT-PCR. However, more
seriouswas the subsequentdemonstration of the loss of
mRNA from one allele for 6 of the 8 CCNCCG heterozygotes (“allelic exclusion”). All six individuals manifested
a type 1 deficiency. In a similar study reported by Reitsma
et al,” three mutations associated with type I protein S deficiencyweredetected
by PCWdirectsequencing of PROS
exons, but only one of these lesions was identified using an
mRNA-basedmutation screeningprotocol. Theothertwo
type I mutations resulted inthe absenceof mRNA transcripts
derived from the diseaseallele. Thus, of 7 type 1 PROS gene
lesionsreported to date (for whichanassociatedstudyof
mRNA abundance had been performed), 5 affect
the level
of themRNAspecies
bearing the PROS mutation.Such
a high frequency of allelic exclusion clearly questions the
feasibility of using the RT-PCR technique successfully as a
screening strategy in type I protein S deficiency.
However, using the RT-PCR protocol can provide valuable information on both the biochemical phenotype and the
molecularbasis of thetype I protein S deficiencystates
studied. For example, thepoint mutation at the invariant + 1
position of theexon IV donorsplice siteresultedin
the
generationof twoabnormal yetin-frame mRNA species.
The predominant mRNA resulted from utilization of a cryptic splice site 48 bp 3’ to the donor splicesite itself, whereas
the rarer mRNA species was generated by the skipping of
exon IV. Abnormal protein was not detected in plasma from
thispatient. The type I deficiency phenotype of patient 1
must therefore result from the inability of protein, translated
from either one orboth aberrant mRNA species, to fold into
a stable conformation. A mutation at position - 1 within the
donor splicesite of exon X (patient 6) resulted in the production of anunstable mRNA. Aberranttranscripts fromthe
mutation-bearing allele could not be detected when cDNA
from thispatient was analyzed by PCR amplification offragment PS6 (Fig 2). Reitsma et all3 have also
reported an exon
X donor splice site mutation at position +5 that is associated
with decreased mRNA stability.
Two short PROS gene deletions resulting in frameshifts
were also foundto influence mRNA stability. Thesingle
base-pair deletion within exon VI11 (patient 7) gave rise to
an in-frame TGA termination codon at amino acid position
250, whereas the 2-bp deletionin exon XIV (patient 4) generated an in-frame TAA termination codon just three amino
acids C-terminal to the lesion. Neither lesion was detected
when PCWdirect sequencing of second-round cDNA fragments, PCR amplified from patient cDNA during the RT-
FORMSTONE ET AL
PCR procedure, was performed. This is consistent with the
previously observed loss, in patients 4 and 7, of one PROS
mRNA allele. In eukaryotes, mRNAs that harbor a nonsense
codon often exhibit an abnormally low abundance.3x.3” We
report here the loss of stability of PROS mRNA associated
with the generation of a termination codon
close to the 3’
terminus of the PROS gene coding region.
The RT-PCR strategy may still be considered to be extremely effective when screening for lesions causing type I1
and I11 protein S deficiency. Second-stage RT-PCR analysis
enabled us to identify rapidly a missense mutation in fragment PS4 (exon VIII) which converted Asn2 l7 to Ser (patient 2). Classification of the protein S deficiency type resulting from this PROS gene lesion had originally proved
difficult owing to variation in protein S antigen levels24 exhibited by eight different members of the family. Theirmean
protein S activity, free antigen, and total antigen values were
45%, 58%, and 76%,respectively (free andtotal values were
incorrectly interchanged in the legend to Fig 1 in the article
cited”).By
comparison, 15 clinicallyandphenotypically
normal family members exhibited mean values of84%. 9 l%,
and 89%, respectively, whereas 9 unrelatedsubjects with
type I protein S deficiency exhibited protein S activity and
free and total antigen values of 25%. 21%, and 4596, respectively. From these data, we may conclude that patient 2 and
other affected family members manifest a type 111 protein S
deficiency.
It is known that protein S contains four very high-affinity
calcium-binding sites:” one or more of which have been
localized tothe third and fourth Asn217 (N217)-bearing
EGF-like domains.Byanalogy
with the calcium-binding
pockets of the first EGF domain of factors IX and X, the
mutated N217 residue is believed to play a significant role
in the binding of
(Fig The
4).
substitution of
Asn217 by Ser resultsin theloss of a y-carbonand the
replacement of the apical amide and reactive carbonyl group
by a less reactive hydroxyl group. The resultant reduction
in both steric and polaric contact between the mutated residue and the calcium ion would be expected to lead to loss
of calcium bindingaffinity. This proposal remainstobe
tested by the biochemicalcharacterization of an in vitroexpressed protein.
Previously, the allelic status of the mRNA from patient 5
was unknown; this patient is a Pro626 CCA homozygote.
However, patient 5 would have been a candidate for rapid
second-stage RT-PCR analysis because the Met570 to Thr
missense mutation was detected in his mRNA. Our inability
to incorporate this patient into the RT-PCR study highlights
the requirement for further informative PROS RFLPs. Because patient 5 is undergoing anticoagulant therapy, we must
assume that his protein S-deficient phenotype will resemble
that of his father, ie, type 111, who also carries the Met570
PROS gene lesion. Met570 is conserved in both rabbit4’ and
bovine protein S4’ and in the human sex hormone binding
globulin (SHBG).& Such sequenceconservation merits speculation as to the structural importance of this residue. The
amino acid environment of Met570 appears hydrophobic,
which suggests thatthis region may beinternal to theprotein.
The type I11 deficiency indicates an effect on C4bBPinterac-
From www.bloodjournal.org by guest on November 18, 2014. For personal use only.
MOLECULARGENETICS OF PROTEIN S DEFICIENCY
2639
A
1 .
/
F.
1
L
Y218
Fig 4. Molecular model of the fourth EGF-like domain of protein S and its putative Ca" binding pocket. (A) The secondarystructure of the
fourth EGF domain of protein S. p-Pleated sheets areshown as green arrows, random coil as a blue ribbon, disulphide bridges in gold, and
cysteine SG atoms in yellow. The side chainof residue Asn217is a ball and stick representation. (B) Close-up view of the wild-type protein S
Ca2+binding pocket. The backboneis represented by a cyan ribbon; residues areshown as stick and are coloredby atom. (C) Close-upview
of the mutant protein S C#+ binding pocket. The backboneis represented by a blue ribbon; residues are shown as stick and are colored by
atom. (D) Close-up view of the wild-type factor X Ca2+binding pocket. The backboneis represented by a purple ribbon; residues areshown
as stick and are colored by atom. The calcium-binding geometry ofthe fourthEGF-like domain has been inferred from NMR analysis of the
first EGF domain of factor X.= The backbone carbonyl of Gly47 and Gly64 and
the side-chain carbonyl oxygens ofHya63 and Gln49together
form a pocket capable of binding calcium (inset D). The replacement of Glu49 and Hya63 of factor X by the analogous Glu205 (E2051 and
Asn217 (N2171residues ofprotein S preservesthe location of the apical reactive groups the
of side chainwhile rotatingtheir geometry through
180' (inset B). This observation is consistent with experiments performed on the calcium-binding domain of factor IX EGF domain l?' The
orientation of the reactive side chains ofN217 and E205 may bring the carbonyl group of N217 into closer contact with thecalcium ion than
is observed in theNMR structure of factor X. This may accountfor the increased binding capacity of the protein Winding pocket overthat
of factor X. Substitution of Asn217 by Ser results in the loss of a pcarbon and the replacement of the apical amide and reactive carbonyl
group by a less reactivehydroxyl group. The resultant reduction in both steric and polariccontact between the mutated residue and calcium
ion wouldbe expected to lead to a loss of calcium binding affinity in this pocket consistent with the phenotype observed.
tion. Indeed, Lys571 is predicted to interfere with protein
S binding to C4bBP." However, further assessment of the
phenotypic effectof this mutation must again await the biochemical characterization of an in vitro-expressed protein
together with the determination of a crystallographic structure for the SHBG domain of protein S .
Our failure to detect PROS lesions in four patients (probands 3, 8, 9, and 10) is intriguing, but not without precedent.I3 Each proband exhibited a family history of protein
S
deficiency and venous thrombosis (Fig
1, C,H, I, and J).
We may speculate from the type E
l deficiency exhibited by
proband 3 and his family (Fig 1C; normal total antigen and
reduced free antigen and activity), that the genetic lesion
responsible may be nonallelic, eg, it could reside in one of
the genes encoding the two chains of C4bBp47 (C4bBP
levels, for all family members analyzed, were found to be
within the normal range). However, patients
8, 9, and 10
exhibited the lossof mRNA from one specificPROS allele,
which implies the existence of a PROS gene lesion. Firststage RT-PCR showed that patient 8 exhibited a low-level
aberrant mRNA transcript. The additional DNA sequence
interposed between exons 3 and 4 was of unknown origin.
This short stretchof DNA may represent unmapped protein
S intronicsequence,ahypothesisthatiscurrentlyunder
investigation. The loss of the Pro626 CCG allele in patient
8 (Fig1H)wasinaccordancewiththedemonstrationof
From www.bloodjournal.org by guest on November 18, 2014. For personal use only.
2640
FORMSTONE ET AL
cosegregation of the CCG allele and the disease phenotype
in affected family members. For patients 9 and 10, however,
loss of the CCG allele was observed, but the disease phenotype in other family members was found to cosegregate with
the CCA allele (Fig 1, I and J). For patient 9, PCWdirect
sequencing of his father’s PROS genes showed a nonsense
mutation (Gly448 to Term) within exon XI1 that was consistent with this individual’s type I deficiency state (Fig 11).
However, the explanation for the lack of concordance in
family H is unclear. Platelet mRNA may not be representative of liver PROS mRNA. However, Reitsma et all3 have
shown that the abundance of PROS mRNA is comparable
in liver, endothelial cells, and blood platelets. Furthermore,
analysis of the PS Heerlen polymorphismz9in one informative individual showed codominant transcription of the platelet PROS gene. We have confirmed this observation in five
normal individuals by studying the Pro626 polymorphism.
This study also provides data that argue against the use
of RT-PCR as a method for quantifying mRNA levels, particularly when employing a two-stage analy~is.~’
Patients 1
and 8 exhibited variation in the quantity of aberrant secondstage PCR fragments amplified from patient cDNA and, i n
the case of patient 1, variation in PCR amplification of the
wild-type fragment. However, such a bias in the amplification of different cDNA templates may prove advantageous
in the amplification of low-abundance mRNA species, such
as those evident in some cases of allelic exclusion.
In conclusion, RT-PCR provides a rapid effective methodology for mutation detection in type I1 and I11 protein S
deficiency states, but is inappropriate for screening for inherited type I protein S deficiency. Exon-by-exon sequencing
remains the most efficient and reliable screening protocol
for type I protein S deficiency. Indeed, our initial argument
against the use of exon-by-exon sequencing for reasons of
technical complexity has been weakened recentlyby a report
describing a homozygous exon I11 PROS gene mutation.”
Finally, analysis of mRNA abundance intypeI-deficient
patients using the RT-PCR strategy has been shown to be a
useful tool for further analysis of the resulting phenotype
manifested by the PROS gene lesions identified.
ACKNOWLEDGMENT
We thank P.H. Reitsma for providing thePROS gene cDNA clone
used in these studies and for PROS gene primer sequences, A.M.R.
Taylor for collecting normal
blood samples, V. Lindo for performing
protein S antigen assays on patient 7,P.J. Hallam for factor VLeiden
analysis of patient 5, G. Patel and C. Goodwin for advice on plasma
protein S analysis, A. Tulinsky for pre-release factor X coordinates
(EGFI), and D.S. Millar for critical reading of the manuscript.
REFERENCES
l . Henkens CMA, van der Meer J, Hillege JL,
van der Schaaf
W, Bom VJJ, Halie MR: The clinical expression
of hereditary protein
risk
C and protein S deficiency: A relation to clinical thrombotic
S? Blood Coag Fibrinol
factors and to levels of protein C and protein
4555, 1993
2. Dahlback B: Protein S and C4b-binding protein: Components
involved in the regulation of the protein C anticoagulant system.
Thromb Haemost 66:49, 1991
3. Heeb MJ, Mester RM, Tans G, Rosing J. Griffin JH: Binding
of proteinS to factor Va associated with inhibition
of prothrombinase
that is independent of activated protein C. J Biol Chem 268:2872,
1993
4. Heeb MJ, RosingJ, Bakker HM, Fernandez JA, Tans G,
Griffin
JH: Protein S binds to and inhibits Factor Xa. Proc Natl Acad Sci
USA 91: 2728, 1994
S. ShenL, Dtihlback B:FactorVandprotein
S as synergistic
cofactors to activated protein C
in degradation of factor VIIIa.
J
Biol Chem 269: 18735, 1994
6. Schwarz HP: Extra-hepaticsynthesis of protein S. Thromb
Haemost 64:333, 1990
7. Maillard C, Berruyer M, Serre CM, DeChavanne M, Delmas
PD: Protein S, avitaminK-dependantprotein,
is a bonematrix
protein component synthesized and secretedby osteoblasts. Endocrinology 130:1.599, 1992
8.DahlbackB,StenfloJ:Highmolecularweightcomplex
in
human plasma between vitamin K-dependent protein S and complementcomponentC4b-bindingprotein.ProcNatlAcadSci
USA
78:2512, 1981
9. Dahlback B: Inhibitionof protein C cofactor functionof human
and
bovine
protein
S by C4b-binding
protein.
J
Biol Chem
26 1:12022, 1986
IO. Scientific and Standardization Committee of the International
Society of Thrombosis and Haemostasis Subcommittee on protein
C and protein S, Munchen, Germany, July 1992. Stuttgart, Germany.
Schattauer
11. Engesser L, Broekmans AW, Briet E, Brommer EP, Bertina
RM: Hereditary protein S deficiency: Clinical manifestations. Ann
Intern Med 106:677, 1987
12. Malm J, Laurel1 M, Dahlback B: Changes in the plasma levels
of vitamin K-dependent proteins C andS and of C4b-binding protein
duringpregnancyandoralcontraception.BrJHaematol68:437,
1988
13. Reitsma PH, Ploos van Amstel HK, Bertina RM: Three novel
mutations in five unrelated subjects with hereditary protein S deficiency type I . J Clin Invest 93:486, 1994
14. Hayashi T, Nishioka J, Shigekiyo T. Saito S, Suzuki K: Protein S Tokushima: Abnormal molecule with a substitution of Glu
of
for Lys-IS5 in the second epidermal growth factor-like domain
protein S. Blood 83:683, 1994
IS. Yamazalu T, Sugiura 1, Matsushita T, Kojima T, Kagami K,
Takamatsu J, Saito H: A phenotypically neutral dimorphism of protein S: The substitution of Lys IS5
by Glu in the second EGF domain
Res 70:39S,
predicted by an A to G base change in the gene. Thromb
1993
16. Borgel D, Gandrille S, Gouault-Heilmann M, Aiach M: First
frameshift mutation in the active protein
S gene associated with a
quantitative hereditary deficiency. Blood Coag Fibrinol 5:593, 1994
17. Gomez E, Ledford MR, Pegelow CH, Reitsma PH, Bertina
RM: Homozygous protein S deficiency due to a one base-pair deletion that leads to stopcodon in exon Ill of the protein S gene.
Thromb Haemost 7 I :723, I994
I 8. Gandrille S, Borgel D, Eschwege-GuffletV, Aillaud M, Dreyfus M, Matheron C, Gaussem P, Abgrall IF, Jude B, Sie P, Toulon
P, Aiach M: Identification of IS differentcandidatecausal
point
mutationsandthreepolymorphisms
in 19patientswithprotein
S
deficiency using a scanning method for the analysis of the protein
S active gene. Blood SS:130, 1995
19. Ploos van Amstel HK , Reitsma PH, van der Logt PE, Bertina
RM: Intron-exon organisation of the active human protein
S gene
PS (Y anditspseudogene PS p: Duplicationandsilencingduring
primate evolution. Biochemistry 29:7853, 1990
20. Schmidel DK, Tatro AV, Phelps LC, Tomczak JA, Long GL:
Organisation of the human protein S genes. Biochemistry 29:7845,
1990
From www.bloodjournal.org by guest on November 18, 2014. For personal use only.
MOLECULARGENETICS OF PROTEIN S DEFICIENCY
2 I . Southern EM: Detection of specific sequences among DNA
fragments separated by gel electrophoresis. J Mol Biol98:503, 1975
22. Ginsberg D, Konkle BA, Gill JC, Montgomery RR, Bockenstedt PL, Johnson TA, Yang AY: Molecular basis of human von
Willebrand disease: Analysis of platelet von Willebrand factor
mRNA. Proc Natl Acad Sci USA 86:3723, 1989
23. Chomczynski P, Sacchi N: Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction.
Anal Biochem 162:156, 1987
24. Marchetti G, Legnani C, Patracchini P, Gemmati D, Ferrati
M, Palareti G, Coccheri S, Bernardi F: Study of a protein S gene
polymorphism at DNA and mRNA level in a family with symptomatic protein S deficiency. Br J Haematol 85:173, 1993
25. Kunkel LM, Smith KP, Boyer SH, Borganokar DS, Wachtel
SF, Miller OJ, Boseg WR, Jones HW, Rory RM: Analysis of human
Y-chromosome-specific reiterated DNAin chromosome variants.
Proc Natl Acad Sci USA 74:1245, 1975
26. Bertina R M , Koeleman BPC, Koster T, Rosendal FR, Dirren
RI, de Ronde H, van der Velden PA, Reitsma PH: Mutation in blood
coagulation factor V associated with resistance to activated protein
C. Nature 36954, 1994
27. Koster T, Rosendaal FR: Activated protein C resistance in
venous thrombosis. Lancet 342: 1503, 1994
28. Friedberg RC, Pizzo SV: Large scale purification of factor X
by hydrophobic chromatography. Prep Biochem 18:303, 1988
29. dertina RM, Ploos van Amstel HK, van Wijngaarden A,
Coenen J, Leemhuis MP, Deutz-Terlouw PP, van der Linden IK,
Reitsma PH: Heerlen polymorphism of protein S, an immunologic
polymorphism due to dimorphism of residue 460. Blood 76:538,
1990
30. Laemli UK: Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227:680, 1970
3 1. Carson M: Ribbon models of macromolecules. J Mol Graphics 5:103, 1987
32. Selander-Sunnerhagen M, Ullner M, Person E, Teleman 0,
Stenflo J, Drakenberg T: How an epidermal growth factor (EGF)like domain binds calcium. J Biol Chem 267:19642, 1992
33. Wacey AI, Krawczak M, Kakkar VV, Cooper DN: Determinants of the factor IX mutational spectrum in haemophilia B: An
analysis of missense mutations using a multi-domain molecular
model of the activated protein. Hum Genet 94:594, 1994
34. Diepstraten CM, Ploos van Amstel HK, Reitsma PH, Bertina
RM: A CCNCCG neutral dimorphism in the codon for Pro 626 of
thehuman protein S gene PS alpha (PROS1). Nucleic Acids Res
195091, 1991
35. Svensson PJ, Dahlback B: Resistance to activated protein C
is a basis for venous thrombosis. N Engl J Med 330:517, 1994
2641
36. Dahlback B: Purification of human vitamin K-dependant protein S and its limited proteolysis by thrombin. Biochem J 209:837,
1983
37. Ploos van Amstel HK, Diepstraten CM, Reitsma PH, Bertina
RM: Analysis of platelet protein S mRNA suggests silent alleles as
a frequent cause of hereditary protein S deficiency type I. Thromb
Haemost 659308, 1991
38. Belgrader P, Maquat LE: Nonsense but not missense mutations can decrease the abundance of nuclear mRNA for the mouse
major urinary protein, while both types of mutation can facilitate
exon skipping. Mol Cell Biol 14:6326, 1994
39. Cheng J, Belgrader P, Zhou X, Maquat LE: Introns are cis
effectors of the nonsense-codon-mediated reduction in nuclear
mRNA abundance. Mol Cell Biol 14:6317, 1994
40. Dahlback B, Hildebrand B, Linse S: Novel t y F of very high
affinity calcium-binding sites in 8-hydroxyasparagine-containing
epidermal growth factor-like domains in vitamin K-dependent protein S. J Biol Chem 265:18481, 1990
41. Handford PA, Mayhew M, Baron M, Winship PR, Campbell
ID, Brownlee GG: Key residues involved in calcium-binding motifs
in EGF-like domains. Nature 351:164, 1991
42. He X, Dahlback B: Molecular cloning, expression and functional characterization of rabbit anticoagulant vitamin K-dependent
protein S. Eur J Biochem 217:857, 1993
43. Dahlback B, Lundwall A, Stenflo J: Primary structure of bovine vitamin K-dependent protein S. Proc NatlAcad Sci USA
83:4199, 1986
44. Gershagen S, Fernlund P, Lundwall A: A cDNA coding for
humansex hormone binding globulin. Homology to vitamin Kdependant protein S. FEBS Lett 220:129, 1987
45. Greengard JS, Fernandez JA, Radtke K-P, Griffin JH: Identification of candidate residues for interaction of protein S with C4b
binding protein and activated protein C. Biochem J 305:397, 1995
46. Hillarp A, Pardo-Manuel F, Ramos Ruiz R, Rodriguez de
Cordoba S, Dahlback B: The human C4b-binding protein 8-chain
gene. J Biol Chem 268:15017, 1993
47. Femandez JA, Griffin JH: A protein S binding site on C4bbinding protein involves 8-chain residues. J Biol Chem 269:2535,
1994
48. Slomski R, Schlosser M, Berg L-P, Wagner M, Kakkar VV,
Cooper DN, Reiss J: Omission of exon 12 in CFTR gene transcripts.
Hum Genet 89:615, 1992
49. Hoskins JA, Norman DK, Beckmann RJ, Long GL: Cloning
and characterisation ofhuman liver cDNA encoding a protein S
precursor. Proc Natl Acad Sci USA 84:349, 1987
50. Beaudet AL, Tsui L-C: A suggested nomenclature for designating mutations. Hum Mutation 2:245, 1993
`