1. No category

From www.bloodjournal.org by guest on October 15, 2014. For personal use only.
1993 82: 151-158
Hemophilia B Leyden: substitution of thymine for guanine at position 21 results in a disruption of a hepatocyte nuclear factor 4 binding site
in the factor IX promoter
MJ Reijnen, K Peerlinck, D Maasdam, RM Bertina and PH Reitsma
Updated information and services can be found at:
http://www.bloodjournal.org/content/82/1/151.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 October 15, 2014. For personal use only.
Hemophilia B Leyden: Substitution of Thymine for Guanine at Position - 21
Results in a Disruption of a Hepatocyte Nuclear Factor 4 Binding Site in the
Factor IX Promoter
By Marlene J. Reijnen, Kathelijne Peerlinck, Diedka Maasdam, Rogier M. Bertina, and Pieter H. Reitsma
Hemophilia B Leyden is an X chromosome-linked bleeding
disorder characterized by an altered developmental expression of blood coagulation factor IX. This form of hemophilia B has been found t o be associated with a variety of
single point mutations in the factor IX promoter region. We
now describe a novel point mutation, T+G at position
21, in t w o related patients with the hemophilia B Leyden phenotype. This mutation lies within the factor IX promoter region ( 40 t o 9) that contains overlapping binding sites for hepatocyte nuclear factor 4 (HNF-4) and
androgen receptor. Transient transfection assays in
HepG2 cells show that the 21 mutation causes a significant reduction in factor IX promoter activity. Gel mobility
shift assays and transient cotransfection experiments revealed that the HNF-4-binding site but not the androgenresponsive element is disrupted by the 21 mutation. A
comparison of the
21 mutation with the previously described - 2 0 T+A mutation (associated with the hemophilia B Leyden phenotype) and - 26 G+C mutation (as-
sociated with severe hemophilia B throughout life) was
made. It shows that the -21 mutation reduced HNF-4
20
binding and transactivation t o a similar level as the
mutation, whereas the - 26 mutation completely abolished HNF-4 binding and transactivation. Mobility shift experiments indicate that there was no significant difference
in binding affinity of recombinant androgen receptor protein for oligonucleotides containing wild-type and 21 or
20 mutated DNA. The binding affinity for the oligonucleotide containing the - 2 6 mutation was twofold lower.
The results indicate that the disruption of the HNF-4-binding site by the - 2 1 T+G mutation is the cause of the
bleeding disorder in these t w o patients. This study adds
further support for the notion that the recovery from hemophilia at puberty may not only be related t o an intact androgen-responsive element but also t o the degree of disruption of the HNF-4-binding site.
0 1993 by The American Society of Hematology.
H
binding site but not the androgen-responsive element is
disrupted by the -2 1 mutation.
-
-
-
-
-
-
EMOPHILIA B is an X-linked bleeding disorder. It
results from a deficiency of clotting factor IX (fIX), a
vitamin K-dependent glycoprotein normally present in
plasma and an essential component of the clotting cascade.’
The majority of mutations causing hemophilia B has been
found in the coding sequence of the fIX gene, whereas all
mutations identified in a subgroup of patients with hemophilia B (hemophilia B Leyden) were detected in the fIX
promoter sequence (at -20, -6, -5, +6, t-8, and + 13*).
The characteristic aspect of hemophilia B Leyden is the
altered developmental expression of flX.334In the patients,
plasma fIX levels are less than 1% of normal during childhood, but after puberty, they gradually rise to a maximum
of 70% of normal.* Apparently, in these patients, the fIX
gene only becomes transcriptionally active after puberty;
probably under the influence of te~tosterone.~
The positions of the mutations associated with hemophilia B Leyden suggest that they disrupt binding sites for
transcription factors that are crucial for constitutive fIX
transcription. The observation that the mutations at -20
T+A and 13 A+G interfere with the binding of HNF-45
and C/EBP,6 respectively, further supports this view.
A recent publication reports that the fIX promoter contains a consensus androgen-responsive element (-36 to
- 2 2 ) that partially overlaps with the consensus HNF-4binding site (-27 to - 15).7 Both the androgen-responsive
element and the HNF-4-binding site were simultaneously
disrupted by a -26 G-tC mutation in the flX promoter.
This mutation was detected in a patient who has suffered
from severe hemophilia throughout
In this report, we describe a novel point mutation, T+G
at -2 1, in the fIX promoter of two related patients with the
hemophilia B Leyden phenotype. The -2 1 mutation, which
lies within overlapping binding sites for androgen receptor
and HNF-4 in the flX promoter, impairs transcription in
transient transfection assays. We show that the HNF-4-
+
Blood, VOI 82,NO 1 (July l), 1993:pp 151-158
-
-
-
PATIENTS, MATERIALS, AND METHODS
Patients. The family studied is of Belgium origin (Fig IA). Patient 111-5 had regular spontaneous hemarthroses at least once
weekly, usually in his ankles and right elbow and occasionally in his
knees till the age of 15 years. From that age onward, spontaneous
hemarthroses no longer occurred, and the patient only needed
transfusions after accidents or prophylactically before minor surgery. Patient 111-8 had very frequent, recurrent spontaneous hemarthroses as a child, especially in the right elbow. At least weekly
transfusions were needed till the age of 14. Thereafter, the frequency of spontaneous hemarthroses diminished, and at the age of
18, the spontaneous bleeding tendency disappeared.
Blood samples, collected in 0.1 volume of 0.1 1 mol/L sodium
citrate, were obtained from the patients (111-5 and 111-8) and the
family members (11-3,11-4, and 111-9). Plasma was obtained by cen-
From the Hemostasis and Thrombosis Research Center, the Department of Hematology, University Hospital, Leiden, The Netherlands; and the Centerfor Thrombosis and Vascular Research, University of Leuven, Belgium.
Submitted November 19, 1992; accepted February 19, 1993.
Supported by the NWOfoundation for Medical and Health Research Medigon (No. 900-526-087) and by a grant from the Stichting Hemophilia.
Address reprint requests to M.J. Reijnen, PhD, Hemostasis and
Thrombosis Research Center, Department of Hematology, Bldg 1:
C2-R, University Hospital, PO Box 9600, 2300 R C Leiden, The
Netherlands.
The publication costs of this article were defayed in part by page
charge payment. This article must therefore be hereby marked
“advertisement” in accordance with 18 U.S.C.section 1734 solely to
indicate this fact.
0 1993 by The American Society ofHematology.
0006-4971/93/8201-0035$3.00/0
151
From www.bloodjournal.org by guest on October 15, 2014. For personal use only.
REIJNEN ET AL
152
A
C
1
II
1
3
1'
1
1
* @
111
8
i 9
IV
10
B Factor IX activity
11
1
2
3 1
2
3 1
2
3 1
2
3
(%)
30
G
t
2s
S'--CTCAGCTTGTACTTTGGTACAACTA..3'
20
-
/
10
8
10
12
14
16
Age (years)
18
20
22
-21
111-5
- 111-8
1s
Fig 1. (A) Pedigree of the Belgian family with hemophilia B Leyden. Four generations of the family are shown. The two patients affected with hemophilia B Leyden are 111-5 and 111-8 ( d i d squares).
Half-solid circles denote carriers. (6) The flX activity levels in each of
the patients (111-5 and 111-8)in relation to age. Before puberty, no flX
was measurable. (C) Sanger dideoxy sequence analysis of the putative factor IX (flX) promoter region of the two patients, 111-5 (track 2)
and 111-8(track 3).and of one normal male (track 1). The arrows mark
21 in the patients.
the mutation T--G at
-
says were performed in HepG2 and HeLa cells seeded into 94-mm
trifugation and tested for flX clotting activities and antigen levels.
diameter Petri dishes. DNA (20 pg flX-CAT, 2 pg of HNF-4 expresflX clotting activities were measured by one-stage assay' and
sion vector [pLEN4St1]and 2 pg of &galactosidase expression vecadapted on an automate in coagulation testing (ACL-810. IL. Milan, Italy). flX antigen levels wcre assayed by the ELlSA method'
tor [pCH I IO"]) were transfected into the cells asdescrihed earlier.'
The CAT activity of each construct. which was normalized for 8-gaand measured automatically on a densitometric analysis microwell
lactosidase activity. was determined in three independent transfecsystem (Organon Teknika. Roxtel. The Netherlands). DNA was
tion experiments. Shown is the average normalized CAT activity of
prepared from the white cells. A DNA fragment (-230 bp) of the
three experiments.
putative flX promoter region spanning nucleotides - 193 to +40
was amplified by the polymerase chain reaction (PCR"). The seG d tnohi/ity shi/i a.s.saj: The analysis of protein-DNA comquenceand location ofthc oligonucleotide primers used in the PCR
plexes by mobility shin assays'and the preparation ofcrude nuclear
amplification are 5' ~ ' V ' ~ ~ ~ ~ C T3' ~ extracts
~ ~ from
~ rat~ livers
~ and
~ HG
~ ~ G
e 6 2~cells~were~ as described
previ-- '
and S'"ACGCGffGCATAACCTTTGC'" 3'. Amplified DNA
ously.'."." Labeled oligonucleotides containing the wild-type or
obtained from two separate amplification reactions was purified by
-21 mutated flX promoter region (-40 to -9: Table I ) were incuagarose gel electrophoresis and sequenced using the Sanger direct
bated with nuclear extracts from rat liver. HepG2 cells. or in vitro
sequencingstrategy. Sequencedata were obtained by sequenceanaltranslated HNF-4" either in the ahsence or presence ofcompetitor
ysis of both the sense and antisense strands of DNA. The analysis
DNA. Bacterially expressed androgen receptor-protein A fusion
was limited to determining the nucleotide sequence of the flX proprotein (a gin from Dr P. De Vos") was used in mobility shin assays
moter region between nucleotide - I92 and +40.
of androgen-responsive element containing oligonucleotides (oligo
"w". m-2 I , m-20, m-26 and ARE: Table I ). These experiments
Transicwt transkction. The wild-type. the -20 T-A. and the
were performed exactly as described by De Vos et ai."
-26 G+C mutated flX promoter-CAT constructs have been described previou~ly.~
The -2 l mutated flX promoter-CATplasmid.
For competition experiments. increasing amounts of doubledesignated - 193 -2 IT+G)CAT00. I. was similarly constructed.
stranded oligonucleotide competitors (Table I ) were added lo a
It contains the -21 T+G mutated human flX promoter region.
constant amount of nuclear extract and "P-labeled probe DNA.
Relative amounts of protein-DNA complexes were determined eispanning nucleotides - 192 to +38. cloned immediately upstream
ther by densitometric analysis (LKR Ultroscan XL dcnsitometer.
of the CAT gene. Relevant portions of the resulting reporter construct were verified by DNA sequencing. Transient transfection asPharmacia, Uppsala. Sweden) of autoradiographs or by counting
7
~
From www.bloodjournal.org by guest on October 15, 2014. For personal use only.
153
HEMOPHILIA B LEYDEN A N D HNF-4 BINDING
the amount of radioactivity (protein bound and free probe) from
the dried gels on the Phosphor Imager (Molecular Dynamics, Sunnyvale, CA) using the Image Quant analysis system. Shown is the
average of three experiments.
RESULTS
The factor IX coagulant activity levels (fIX:C) in the two
patients with hemophilia B Leyden (111-5 and 111-8) were
measured up to the age of 19 and 22 years, respectively (Fig
1B). The patients had a clinical history compatible with severe hemophilia B in the prepuberty years (flX:C levels below l % of normal). One of the patients needed prophylactic
treatment for frequent hemorrhages. After onset of puberty,
the f€X:C levels gradually rose to 12% to 14% and are still
rising. With the rise of fIX:C levels, the spontaneous bleeding tendency disappeared. In a recent blood sample of these
two hemophiliacs, factor IX antigen was found to be lo%,
with factor IX antigen/activity ratios of 0.96 and 0.7 1, respectively (indicating the synthesis of fully active factor IX
molecules).
A variety of single point mutations in the putative fIX
promoter region has been associated with hemophilia B
Leyden. Using the strategy of PCR amplification and the
Sanger direct sequencing analysis, we determined the sequence of the fIX promoter region of these two patients
with hemophilia B Leyden (Fig IC, 111-5 and 111-8). A novel
point mutation, T+G at position -2 1, of the flX gene was
detected. This mutation, in addition to a normal allele, was
also present in maternal DNA (11-3 and 11-4) and DNA prepared from the sister (111-9) of patient 111-8, thus confirming
their carrier status.
To determine whether the -2 1 T-tG mutation impairs
transcription from the fIX promoter, we analyzed the transcriptional activity of the wild-type and -2 1 mutated promoters. The fIX promoter-CAT constructs containing
wild-type and -2 1 T+G mutated fragments (extending
from - 192 to +38) were tested for expression in transiently
transfected HepG2 and HeLa cells. In agreement with previous observations, the wild-type promoter of the human
fix gene was capable of directing CAT transcription in
HepG2 cells but not in HeLa cells (Fig 2). When the transfections in HepG2 were carried out with the -21 T+G
mutated construct (- 192(-21T+G)CATOO.I), only a
background level of CAT activity was found. Similar experiments as described for the -21 mutation were previously
-
Table 1. Oligonucleotides Used in Gel Mobility Shift Assays
Sequence
Human factor IX gene
”wt“ ATACAGCTCAGCTTGTACTTTGGTACAACTAA
m-2 1 ATACAGCTCAGCTTGTACTgTGGTACAACTAA
17-20ATACAGCTCAGCTTGTACTTaGGTACAACTAA
m-26 ATACAGCTCAGCTTcTACTTTGGTACAACTAA
Rat prostatic binding protein gene
AREGATCATAGTACGTGATGTTCTCAAGATC
Chicken ovalbumin gene
COUP CTATGGTGTCAAAGGTCAAACTTCT
Position
-40
-40
-40
-40
to -9
to -9
to -9
to -9
core II
Intron seq
-90 to -65
The sequence and location of the top strand of the oligonucleotidesused in competition binding assays are given. Mutated nucleotides are indicated by lower case
letters.
done for the -20 T-A and -26 G+C mutated fIX promoters. Like the -21 mutation, they also produced only
background levels of CAT activity in HepG2 cells (Fig 2A).5
Because the T+G mutation at -21 lies within a region
that contains a HNF-4-binding site and an overlapping androgen-responsive element, we tested whether this mutation disrupts either of them. Oligonucleotides containing
the -40 to -9 region of the fIX promoter (wild type: oligo
“wt” and -2 1 T+G mutated: oligo m-21) were used in gel
mobility shift assays (Table 1). HNF-4 protein, either from
HepG2 cells (Fig 3A, lanes 1 and 3) or translated in vitro
from HNF-4 cDNA (lane 8), bound the labeled oligo “wt”
very well. When the experiments were carried out with labeled oligo m-2 1, reduced binding of HNF-4 to -2 1 mutated site was found (lane 2). The labeled HNF-4-DNA
complex was competed effectively by 40-fold molar excess
of unlabeled oligo “wt” (lanes 4 and 9) and to a lesser extent
by oligo m-21 (lanes 5 and IO). To confirm that indeed
HNF-4 binding is involved, we incubated the mobility shift
reaction mixtures with specific antibodies to HNF-4.” This
resulted in a “supershift” of the complex (lane 6). The addition of anti-NF- 1 antibody (lane 7, negative control16)had
no effect. Additional competition experiments were performed with increasing amounts of unlabeled competitor
DNA (oligo “wt”, m-2 1, m-20 or m-26; Table 1) added
to a constant amount of rat liver nuclear extract and labeled
oligo “wt” (Fig 3B). The experiments were repeated three
times. Both the -21 mutation and the -20 mutation
disrupt the HNF-4-binding site to a similar level (- I6
times more mutant DNA than wild-type DNA was needed
to reduce the formation of labeled protein-DNA complex
by 50%). The -26 mutated oligonucleotide (m-26) competed only marginally for HNF-4 binding. Similar observations were made with HepG2 nuclear extracts (data not
shown). These results indicate that the mutation at -21
interferes with HNF-4 binding to the fIX promoter. Moreover, although the -21 mutation and the -20 mutation
partially reduce HNF-4 binding, the -26 mutation seems to
completely disrupt the HNF-4-binding site.
To determine whether the androgen-responsive element
was also disrupted by the -2 1 mutation, we performed gel
mobility shift assays with a recombinant androgen receptor-protein A fusion protein.15When the experiments were
carried out with labeled oligo “wt” as probe, only a small
fraction ofthe probe was retarded (Fig 4A, lanes 1 and 2). In
contrast, the fusion protein bound very well to a probe containing the well-characterized androgen-responsive element
from the rat prostatic binding protein gene (oligo ARE,
lanes 3 and 4). Therefore, in an attempt to establish a difference in binding of the fusion protein to the wild-type and
-2 1 mutated flX promoter region, we performed competition experiments with oligo ARE as probe and increasing
concentrations of unlabeled oligo “wt” and oligo m-2 1 as
competitors (Fig 4B). Compared with the competition with
oligo ARE, large amounts (300- to 400-fold mol/L excess)
of unlabeled oligo “wt” and oligo m-2 I were required to
reduce the formation of labeled protein DNA complex by
50%(Fig 4B,C). There was no significant difference in competition for fusion protein binding between oligo “wt” and
oligo m-2 1. When the experiments were carried out with
From www.bloodjournal.org by guest on October 15, 2014. For personal use only.
154
REIJNEN ET AL
CAT activity/cpm per dish x 10,000
18
16
14
12
10
8
6
4
CATOO.l
-192CATOO.l
(-21 T-G)
(-26 G-C)
(-20 T-A)
HeLa cells
Fig 2. Effects of the - 21, - 26, and - 20 mutations on flX promoter activity in HepG2 cells (A) and Hela cells (B). Plasmid constructs
20 T+A
with the CAT reporter gene under control of wild-type ( - 192CATOO.l], - 21 T+G mutated, - 2 6 G - 4 mutated, and
mutated flX promoters were tested for CAT activity in the absence (m) or presence (B]of the HNF-4 expression vector pLEN4S." CATOO.l
was the promoterless control plasmid. CAT activity was normalized to the activity of a @-galactosidaseinternal standard. Shown is the
average normalized CAT activity of three experiments.
-
From www.bloodjournal.org by guest on October 15, 2014. For personal use only.
155
HEMOPHILIA B LEYDEN AND HNF-4 BINDING
A
comp:
,+
'
4 d
-
n6"
8 9
3 4 6
10
B
Residual Binding (%I
1207__-
I
A
F
60.
-c
WT
-+
M-21
-#+
M-26
M-20
4
0
20
10
Fold Molar Excess Competitor
-
Fig 3. Analysisofthe effect of the 21 T--G mutation on the binding affinity for hepatocyte nuclear factor 4 (HNF-4)by gel mobility shift
assays. (A) Crude nuclear extract from H e 6 2 cells (lanes 1 through 7) or in vitro translated HNF-4 (lanes 8 through 10) was incubated with
150-lmol labeled oligonucleotide "wt" (lanes 1 and 3 through 10) or oligonucleotide m 21 (lane 2) as probes. The assay was done in the
and 8) or presence ( ) of 40-fold mol/L excess of unlabeledoligonucleotides "wt" (lanes 4 and 9) or
absence ([ 1, lanes 1 through 3,6,7,
m - 21 (lanes 5 and 10)or in the presence of antiserum (lane 6, anti-HNF-4; lane 7, anti-NF1). (B) Competition mobility shift assays with
20, and - 26) factor IX promoter sequences ( - 40 to 9) using crude rat liver nuclear extracts. Increasing
wild-type and mutated ( 21,
amounts of unlabeledoligonucleotides "wt". m 21. m 26, and m - 20 were added to a constant amount of end-labeledoligonucleotide
"wt" as probe and rat liver nuclear extracts. Binding was quantitated by densitometric scans (LKB Ultroscan XL densitometer) of the
autoradiograms. The values of the formation of labeled HNF-&DNA complex throughout is shown relative to the value of labeled HNF-4DNA complex in the absence of competitor, which is arbitrarily set at 100. The experiments were carried out three times. The values
reported represent averages of the experiments.
-
+
-
-
-
-
unlabeled oligo m-20 as competitor. similar results were
obtained (Fig 4C). The affinity of the fusion protein for
oligo m-26 was lower since a twofold higher concentration
of DNA was needed to reduce the formation of labeled protein DNA complex by 50%. No competition was observed
with excess unlabeled oligonucleotide containing the
-
-
COUP-TF-binding sequence (COUP). These results indicate that the HNF-4-binding site but not the androgen-responsive element is disrupted by the -21 T+G mutation.
The functional consequence of the reduction in HNF-4
binding to the -21 mutated flX region was analyzed by
transient cotransfection experiments in H e 6 2 and HeLa
From www.bloodjournal.org by guest on October 15, 2014. For personal use only.
156
RELJNEN ET AL
B
A
probe
ARE
WT
probe ARE
M-21,
comp.
5 6 7 8 9 10 111213141516
1
2
3
4
C
Residual Binding (%I
120,
~
---
-
~
1
- - : <
;
..
+---
\
__
L
60
*
.\.
\.
.\,
I
1
I
-
1
--
WT
-.A
M-21
I
:
>
M-26
-;z M-20
L
ARE
40.
20
-+
',
c
1
COUP
\
!
L
I
I
0
!
t
'
L
40100
'
l
l
!
I
I
,
,
I
!
g
o
'
800
400
200
Fold Molar Excess Competitor
Fig 4. Analysis of androgen receptor binding t o wildtype and mutated factor IX promoter regions. (A) Mobility shift a m y with androgen
receptor-protein A fusion protein and labeled digo "wt" (lanes 1 and 2)or labeled oligo ARE (lanes 3 and 4) as probes. (B) Competition
mobility shift assay done with labeledoligo ARE as probe and fusion protein (1 50 fmol of labeled DNA and 0.4 pg of androgen receptor-protein A fusion protein) using oligo "wt" and oligo m 21 as competitors. Binding was inhibited with an increasing mol/L excess of unlabeled
400-,
and 800-fold mol/L excess). (C) Graph of the competition analysis done with androgen
oligo "wt" and oligo m - 21 (40-, loo-,200-.
receptor-protein A fusion protein using wild-type (oligo "wt") and mutated (oligos m 21,m 20,and m 26)factor IX promoter sequences, the COUP-TF-binding site of the chicken ovalbumin promoter (COUP) and the androgen-responsiveelement of the rat prostatic
binding protein (oligo ARE) as competitors (Table 1). Increasing amounts of unlabeledoligonucleotides "wt", m 21,m 26,m 20,and
ARE were added to a mixture of a fixed amount of end-labeledoligonucleotide ARE as probe and androgen receptor-protein A fusion protein.
The amounts of radioactivity (protein bound and free probe) were measuredon a Phosphor Imager using the Image Quant analysis system.
The values of the formation of labeled androgen receptor-DNA complex throughout is shown relative to the value of labeled androgen
receptor-DNA complex in the absence of competitor, which is arbitrarily set at 100.Shown is the average of three experiments.
-
-
-
-
-
-
-
From www.bloodjournal.org by guest on October 15, 2014. For personal use only.
157
HEMOPHILIA B LEYDEN AND HNF-4 BINDING
cells. In these experiments, an expression vector for HNF-4
(pLEN4S") was cotransfected with the reporter constructs
described above. The results presented in Fig 2A,B show
that HNF-4 transactivates the wild-type reporter construct
- 192CAT00.1) in both cell types quite well. Compared
with the wild-type construct, the -21 mutated plasmid
(- 192[-21 T+G]CATOO. 1) was expressed at a significant
lower level ( 1 8% in HepG2 cells and 15% in HeLa cells) in
response to HNF-4 transactivation. Similar observations
were made for the previously reported -20 T+A mutated
promoter (17% in HepG2 cells and 20% in HeLa cells),
whereas no transactivation was observed of the -26 G+C
mutated promoter.
DISCUSSION
A T+G base change at position -2 I was detected in the
fIX promoter region of two related patients with the hemophilia B Leyden phenotype. The mothers ofthe patients also
carried the defective allele suggesting that the original mutation has occurred in one of their parents or other forebears.
The T+G mutation at -2 I in both patients is associated
with severe bleeding symptoms and plasma fIX levels below
1% of normal during childhood. The severity of the fIX
deficiency in these two patients is similar to those observed
in the patients with mutations at position -20. Interestingly, the residues at -22, -21, and -20 (all thymine residues) are conserved in various mammalian species (ie, human, macaque, dog, rat, and mouseI7). It seems that the
residues at -2 1 and -20 are critical for transcription initiation.
The causal relationship between the -21 T+G base
change and the hemophilia B Leyden phenotype was demonstrated by transient transfection analysis and in vitro
binding studies. The mutation, which lies within overlapping binding sites for androgen receptor and HNF-4 in the
fIX promoter, causes a significant reduction in promoter
activity. We showed that the T+G mutation at -2 1 interferes with the binding of HNF-4 to its recognition site (between -34 and -10). Moreover, we show that although
HNF-4 transactivates the wild-type fIX promoter quite
well, it could activate the -21 T+G mutated promoter
only to a limited extent. The data presented here indicate
that the disruption of the HNF-4-binding site by the -21
T-G mutation is the cause of the severely impaired (hepatic) flX gene expression in these patients during childhood.
These results add further support for the notion that
HNF-4 plays a dominant role in controlling flX gene expression in normal individuals. In fact, the observation that
fIX promoter constructs are very efficiently activated by
HNF-4 in the nonhepatic HeLa cells suggest that HNF-4 is
the major determinant of the liver-specific production of
BX.
It was recently reported that the fIX promoter contains a
functional androgen-responsiveelement between -40 and
-9.7 These studies demonstrated that four copies of the
wild-type fIX promoter region (-42 to -17) confer androgen responsiveness to a herpes simplex thymidine kinase
(tk) promoter in a heterologous system. When the -26
G+C mutation was present no transactivation was ob-
served, whereas the -20 T-A mutation did not affect transactivation in HeLa cells. When we tested this candidate androgen-responsive element in gel mobility shift assays, the
binding affinity of the androgen receptor (ie, recombinant
androgen receptor-protein A fusion protein) for this site
was rather low when compared with the androgen-responsive element from the rat prostatic binding protein gene.
Furthermore, transient cotransfection experiments with several different androgen receptor coding plasmids in HepG2
cells showed that this potential androgen responsive element in the fIX promoter responded only poorly to the
activation by androgen^.^,^* Although such transactivation
experiments may not reflect the in vivo situation, this marginal response to androgens is not comparable with the significant increase in flX expression in patients with hemophilia B Leyden after puberty.
Assuming that the putative fIX androgen-responsive element is functional in the fIX promoter, it is interesting to
note that the -26 mutation causes only a minor difference
in binding affinity of the androgen receptor-protein A fusion protein. The fact that this mutation not only disrupts
the androgen-responsive element but also completely
blocks HNF-4 binding suggests a synergistic interaction between androgen receptor and HNF-4. The observation that
neither of the hemophilia B Leyden-like mutations at -2 1
and at -20 completely eliminates HNF-4 binding and transactivation suggests that remnant HNF-4 binding is required
for the androgen-mediated response after puberty. Both androgen receptor and HNF-4 are members ofthe steroid hormone receptor superfamily, a group of ligand-dependent
transcription factors that possess a high degree of similarity
in their DNA-binding domains. It is therefore possible that
HNF-4 forms heterodimers with the androgen receptor, as
has been seen for the thyroid hormone receptors and the
retinoic acid receptor^.'^^*^
ACKNOWLEDGMENT
The authors thank F.M. Sladek for the kind gift of specific antibodies to HNF-4. We thank Dr P. De Vos for providing bacterially
expressed androgen receptor-protein A fusion protein. We thank
E.G.C. Wojcik for measuring the fIX clotting activities and fIX
antigen levels.
REFERENCES
I. Thompson AR: Structure, function and molecular defects of
factor 1X. Blood 67:565, 1986
2. Giannelli F. Green PM, High KA, Sommer S, Lillicrap DP,
Ludwig M, Olek K. Reitsma PH, Goossens M, Yoshioka A, Brownlee GG: Haemophilia B: Database of point mutations and short
additions and deletions-third edition. Nucleic Acids Res 20:2027,
1992
3. Bnet E, Bertina RM, van Tilburg NH, Veltkamp J: Hemophilia B Leyden: A sex-linked hereditary disorder that improves
after puberty. N Engl J Med 306:788, 1982
4. Briet E, Merk MD, Wijnands MD, Veltkamp J: The prophylactic treatment of hemophilia B Leyden with anabolic steroids. J
Ann Intern Med 103:225, 1985
5. Reijnen JM, Bertina RM, Reitsma PH: Disruption of a binding site for hepatocyte nuclear factor 4 results in hemophilia B Leyden. Proc Natl Acad Sci USA 89:6300, 1992
6. Crossley M, Brownlee GG: Disruption of a C/EBP binding
From www.bloodjournal.org by guest on October 15, 2014. For personal use only.
158
site in the factor IX promoter is associated with haemophilia B.
Nature 345:444, 1990
7. Crossley M, Ludwig M, Stowell KM, de Vos P, Olek K,
Brownlee G G Recovery from hemophilia B Leyden: An androgenresponsive element in the fIX promoter. Science 257:377, 1992
8. Veltkamp JJ, Drion EF, Loeliger EA: Detection of the camer
state in hereditary coagulation disorder. Thromb Diath Haemorrh
19:279, 1968
9. Poort SR, van der Linden K, Krommenhoek-van Es C, Briet
E, Bertina RM: Rabbit polyclonal antibodies against the calciumdependent conformation of factor IX and their application in solid
phase immunoradiometric assays. Thromb Haemost 55: 122, 1986
IO. Saiki RK, Horn GT, Mullis KB, Erlich HA: Primer-directed
enzymatic amplification of DNA with a thermostable DNA polymerase. Science 239:487, 1988
11. Sladek FM, Zhong W, Lai E, Darnell Jr JE: Liver enriched
transcription factor HNF-4 is a novel member of the steroid hormone receptor superfamily. Genes Dev 4:235, 1990
12. Herbomel P, Bourachot B, Yamir M: Two distinct enhancers with different cell specificities coexist in the regulatory region of polyoma. Cell 39:653, 1984
13. Gorski K, Carneiro M, Schibler U: Tissue-specific in vitro
transcription from the mouse albumin promoter. Cell 47:767, I986
14. Dignam JD, Lebovitz RM, Roeder RG: Accurate transcrip-
REIJNEN ET AL
tion initiation by RNA polymerase I1 in a soluble extract from
isolated mammalian nuclei. Nucleic Acids Res 11:1475, 1983
15. De Vos P, Claessens F, Winderickx J, Van Dijck P, Celis L,
Peeters B, Rombauts W, Heyns W, Verhoeven G: Interaction of
androgen response elements with the DNA-binding domain of the
rat androgen receptor expressed in Escherichia coli. J Biol Chem
266:3439, 199 1
16. Rosenfeld PJ, Kelly T: Purification of Nuclear Factor I by
DNA recognition site affinity chromatography. J Biol Chem
261:1398, 1986
17. Pang C-P, Crossley M, Brownlee GG: Comparative sequence analysis of mammalian factor IX promoters. Nucleic Acids
Res 185731, 1990
18. Hirosawa S, Fahner JB, Salier J-P, Wu C-T, Lovrien EW,
Kurachi K Structural and functional basis of the developmental
regulation of human coagulation factor IX gene: Factor IX Leyden.
Proc Natl Acad Sci USA 87:442 I, 1990
19. Forman BM, Yan C-R, Au M, Casanova J, Ghysdael J, Samuels HH: A domain containing leucine-zipper-like motifs mediate
novel in vivo interactions between the thyroid hormone and retinoic acid receptors. Mol Endocrinol 3: I6 10, 1989
20. Glass CK, Lipkin SM, Devary OV, Rosenfeld MG: Positive
and negative regulation of gene transcription by a retinoic acid-thyroid hormone receptor heterodimer. Cell 59:697, I989