A 5' splice-region G → C mutation in exon 1 of the human β-globin gene inhibits pre-mRNA splicing: A mechanism for β+-thalassemia

Abstract

We have characterized a Mediterranean beta-thalassemia allele containing a sequence change at codon 30 that alters both beta-globin pre-mRNA splicing and the structure of the hemoglobin product. Presumably, this G----C transversion at position -1 of intron 1 reduces severely the utilization of the normal 5' splice site since the level of the Arg----Thr mutant hemoglobin (designated hemoglobin Kairouan) found in the erythrocytes of the patient is very low (2% of total hemoglobin). Since no natural mutations of the guanine located at position -1 of the CAG/GTAAGT consensus sequence had been isolated previously, we investigated the role of this nucleotide in the constitution of an active 5' splice site by studying the splicing of the pre-mRNA in cell-free extracts. We demonstrate that correct splicing of the mutant pre-mRNA is 98% inhibited. Our results provide further insights into the mechanisms of pre-mRNA maturation by revealing that the last residue of the exon plays a role at least equivalent to that of the intron residue at position +5.

Full text

Proc.
Nail.
Acad.
Sci.
USA
Vol.
86,
pp.
1041-1045,
February
1989
Medical
Sciences
A
5'
splice-region
G
->
C
mutation
in
exon
1
of
the
human
fi-globin
gene
inhibits
pre-mRNA
splicing:
A
mechanism
for
fB+-thalassemia
MICHEL
VIDAUD*,
RENATA
GATTONIt,
JAMES
STEVENINt,
DOMINIQUE
VIDAUD*,
SERGE
AMSELEM*,
JEMNI
CHIBANIf,
JEAN
RoSA*,
AND
MICHEL
GOOSSENS*§
*Institut
National
de
la
Santd
et
de
la
Recherche
M6dicale
U.91,
H6pital
Henri
Mondor,
94010
Creteil,
France;
tInstitut
National
de
la
Santd
et
de
la
Recherche
Mddicale,
U.184,
Laboratoire
de
Gdndtique
Mol6culaire
du
Centre
National
de
la
Recherche
Scientifique,
Facultd
de
M6decine,
67085
Strasbourg,
France;
and
tFacultd
de
Pharmacie,
5019
Monastir,
Tunisia
Communicated
by
Y.
W.
Kan,
September
22,
1988
(received
for
review
July
14,
1988)
ABSTRACT
We
have
characterized
a
Mediterranean
,B-
thalassemia
allele
containing
a
sequence
change
at
codon
30
that
alters
both
,B-globin
pre-mRNA
splicing
and
the
structure
of
the
hemoglobin
product.
Presumably,
this
G
-.
C
transver-
sion
at
position
-1
of
intron
1
reduces
severely
the
utilization
of
the
normal
5'
splice
site
since
the
level
of
the
Arg
-.
Thr
mutant
hemoglobin
(designated
hemoglobin
Kairouan)
found
in
the
erythrocytes
of
the
patient
is
very
low
(2%
of
total
hemoglobin).
Since
no
natural
mutations
of
the
guanine
located
at
position
-1
of
the
CAG/GTAAGT
consensus
sequence
had
been
isolated
previously,
we
investigated
the
role
of
this
nucleotide
in
the
constitution
of
an
active
5'
splice
site
by
studying
the
splicing
of
the
pre-mRNA
in
cell-free
extracts.
We
demonstrate
that
correct
splicing
of
the
mutant
pre-mRNA
is
98%
inhibited.
Our
results
provide
further
insights
into
the
mechanisms
of
pre-mRNA
maturation
by
revealing
that
the
last
residue
of
the
exon
plays
a
role
at
least
equivalent
to
that
of
the
intron
residue
at
position
+5.
The
study
of
naturally
occurring
mutations
in
the
human
,B-globin
gene
has
contributed
greatly
to
the
understanding
of
the
mechanisms
of
normal
transcription
and
mRNA
process-
ing.
These
mutations
give
rise
to
two
main
classes
of
disorders:
,B°-thalassemiai
in
which
no
3-globin
chains
are
produced,
and
f3-thalassemia,
in
which
8-globin
chains
are
synthesized,
but
at
reduced
levels
(1).
The
underlying
mo-
lecular
lesions
are
highly
heterogeneous
and
interfere
with
virtually
every
stage
of
/3-globin
gene
expression.
Several
,3-thalassemia
genes
have
been
shown
to
contain
nucleotide
substitutions
that
adversely
affect
RNA
splicing
in
diverse
ways
(for
reviews,
see
refs.
2
and
3).
The
splicing
of
mRNA
precursors
requires
at
least
two
conserved
sequences
at
the
ends
of
introns:
the
5'
splice-site
consensus
sequence
CAG/GTAAGT,
which
shows
a
striking
complementarity
to
the
5'-terminal
region
of
U1
small
nuclear
RNA
(snRNA),
and
the
3'
splice-site
consensus
sequence
Y"NCAG/
(4-6).
Most
of
the
naturally
occurring
mutations
and
in
vitro-generated
mutations
that
reduce
the
complemen-
tarity
with
U1
snRNA
result
in
authentic
5'
splice-site
inactivation
(for
reviews, see
refs.
7
and
8)
and
can
activate
cryptic
sites
(9-11).
Biochemical
data
and
complementation
experiments
also
support
the
proposal
that
the
5'
end
of
U1
snRNA
interacts
with
the
5'
splice
site
in
mRNA
precursors
(12-15).
However,
it
is
not
possible
to
predict
with
confidence
the
effect
of
one
mutation
at
a
given
5'
splice
site.
For
instance,
a
number
of
point
mutations
in
the
5'
splice
site
of
the
large
rabbit
3-globin
intron
do
not
severely
inhibit
splicing
in
transfected
Hela
cells
or
in
vitro
(10,
16,
17).
These
results
could
be
interpreted
by
proposing
that
5'
splice
sites
whose
sequences
match
best
with
the
consensus
can
undergo
point
mutation
without
severe
inactivation
(18,
19).
Nevertheless,
the
strength
of
the
5'
splice
sites
is
probably
not
the
only
factor
involved
in
the
splice-site
selection,
which
can
be
less
dependent
on
match
to
the
consensus
than
on
the
sequence
context
in
which
the
5'
splice
site
resides
(20).
A
possible
explanation
for
these
divergent
observations
could
be
that
the
individual
contribution
of
exon
or
intron
residues
to
the
5'
splice
site
is
not
equivalent.
However,
the
role
of
the
nucleotides
preceding
the
5'
splice
site,
which
are
moderately
conserved,
has
not
been
extensively
documented
(8).
It
is
striking
that
no
natural
mutations
of
the
guanine
at
position
-1
(present
in
80%
of
cases
for
human
genes)
have
been
isolated.
Here
we
describe
the
isolation
and
characterization
of
a
G
C
mutation
in
the
consensus
sequence
at
the
last
nucleotide
of
exon
1
of
the
human
f3-globin
gene
that
results
in
a
considerable
reduction
of
RNA
splicing
at
the
5'
splice
site
of
intervening
sequence
1
(IVS1).
We
have
analyzed
the
products
of
this
mutant
gene
in
vivo
and
determined
the
effects
of
this
mutation
on
the
in
vitro
splicing
of
the
pre-mRNA.
MATERIALS
AND
METHODS
Patient.
A
North
African
woman,
daughter
of
a
Tunisian
mother
and
a
Libyan
father,
having
mild
/B3-thalassemia
was
the
subject
of
our
study.
She
was
not
transfusion-dependent
and
at
age
24,
after
splenectomy,
her
hemoglobin
level
was
9-10
g/dl;
a
peripheral
blood
smear
showed
abnormal
cells
with
1500
erythroblasts
per
100
leukocytes.
Hemoglobin
electrophoresis
revealed
85%
Hb
F,
11%
Hb
A,
and
4%
Hb
A2.
Peripheral
blood
DNA
analysis
was
performed
as
de-
scribed
(21).
Pedigree
analysis
(Fig.
1)
established
that
the
patient
had
inherited
a
/3+-thalassemia
gene
associated
with
a
yet
undescribed
,8-thalassemia
chromosome
haplotype
(22)
from
her
father
and
a
$+-thalassemnia
gene
linked
to
Medi-
terranean
haplotype
I
(23)
from
her
mother.
Cloning
and
Sequencing.
The
(B-globin
genes
were
cloned
after
fragment
enrichment
as
described
by
Nicholls
et
al.
(24)
in
the
vector
pUC19.
DNA
sequence
analysis
was
performed
by
the
dideoxy
chain-termination
method
of
Sanger
et
al.
(25).
Genotype
Determination.
DNA
samples
from
the
family
members
were
enzymatically
amplified
by
the
polymerase
chain
reaction
(PCR)
(26, 27),
dot-blotted
onto
nylon
filters,
and
analyZed
by
hybridization
with
allele-specific
oligode-
ox
'nucleQtide
probes
homologous
to
the
normal
and
mutated
seqUences
as
described
(28):
,B30
normal
probe,
5'-CCCTG-
Abbreviations:
IVS,
intervening
sequence
(intron);
snRNA,
small
nuclear
RNA;
PCR,
polymerase
chain
reaction;
ddNTP,
dideoxy-
nucleoside
triphosphate;
nt,
nucleotide(s).
§To
whom
reprint
requests
should
be
addressed.
1041
The
publication
costs
of
this
article
were
defrayed
in
part
by
page
charge
payment.
This
article
must
therefore
be
hereby
marked
"advertisement"
in
accordance
with
18
U.S.C.
§1734
solely
to
indicate
this
fact.

1042
Medical
Sciences:
Vidaud
et
al.
2
3
4
F
Ac
K
\-
A
7--
-F
-S
*
-A2
g30
Probe
Normal
*
Mutant
*
*
1
2
3
4
Normal
control
FIG.
1.
Phenotype
and
genotype
analysis
of
patients
with
hemo-
globin
Kairouan.
(Top)
Pedigree
of
a
Tunisian
family
in
which
the
.30
splicing
mutation
(black)
is
present
in
the
proband
(1-4)
and
in
three
other
individuals.
(Middle)
Analysis
of
the
hemoglobin
patterns
by
isoelectric
focusing.
The
presence
of
hemoglobin
Kairouan
is
un-
masked
in
the
subject
(11-1)
bearing
another
abnormal
3-globin
allele
((8S)
and
manifests
as
a
band
migrating
similarly
to
Hib
A.
(Bottom)
Dot
blot
analysis
of
PCR-amplified
genomic
DNA
from
the
same
individuals
by
hybridization
with
oligonucleotide
probes
for
the
,830
normal
and
thalassemia
sequences.
GGCAGGTTGGTATC-3';
(30
mutant
probe,
5'-CCCTGG-
GCACGTTGGTATC-3'
Plasmid
Construction.
The
SP6
promoter-containing
plas-
mids
were
prepared
by
inserting
the
Rsa
I
(position
-127)-
BamHI
(+477)
fragment
of
the
human
P-globin
gene
between
the
HindIII
and
BamHI
sites
of
pOP65,
after
ligation
of
a
HindIll
linker
to
the
Rsa
I
end.
The
H,8G
plasmid contains
the
wild
type
globin
sequence.
The
IVS1(-1C)
mutant,
in
which
the
guanine
at
position
-1
from
the
5'
splice
site
is
changed
to
a
cytosine,
and
the
IVS1(+2G)
mutant,
in
which
the
thymine
at
position
+2
is
changed
to
a
guanine
(29),
were
constructed
by
replacing
the
wild-type
sequence
between
the
Nco
I
(position
+48)
and
BamHI
sites
by
the
mutated
fragments.
Transcription
and
in
Vitro
Splicing
Reaction.
Labeled
tran-
scripts
were
synthesized
by
run-off
transcription
of
templates
linearized
at
the
Sma
I
site,
and
the
in
vitro
splicing
reaction
was
carried
out
according
to
Krainer
et
al.
(30).
Assays
have
been
done
in
the
presence
of
KCl
(42-60
mM)
and
MgCl2
(1.6-3.2
mM)
at
30'C
or
350C
(31).
The
processed
RNAs
were
purified
by
proteinase
K
treatment
and
phenol/chloroform
extraction
and
analyzed
by
electrophoresis
in
6%
or
8%
acrylamide
sequencing
gels
(31).
Primer
Extension
Analysis.
Primer
extension
analysis
of
the
total
processed
RNA
from
each
transcript
was
performed
in
the
presence
of
a
dideoxynucleoside
triphosphate
(ddXTP)
and
the
three
other
dNTPs
(32).
A
23-residue
oligonucleotide
complementary
to
the
5'
end
of
exon
2
was
used
as
primer.
The
5'-end-labeled
primer
was
hybridized
to
RNA
overnight
and
the
reverse
transcription
was
performed
as
described
(31)
except
that
0.2
mM
ddATP
or
ddGTP
was
used
and
the
three
other
dNTPs
were
present
at
0.08
mM
(32).
RESULTS
Cloning
and
Sequence
Analysis
of
fi-Thalasseniia
Genes.
Haplotype
analysis
of
the
patient
revealed
heterozygosity
for
Mediterranean
haplotype
I
and
for
a
novel
haplotype,
Na
(-
-
-
-
+
+
-
+),
yet
undescribed
in
83-thalassemia
chromo-
somes
of
Mediterranean
subjects
(22).
The
patient's
DNA
was
analyzed
by
hybridization
with
allele-specific
oligonu-
cleotide
probes
corresponding
to
the
most
frequent
mutations
encountered
in
the
Mediterranean
basin.
All
mutations
tested
were
shown
to
be
absent
in
this
patient,
thereby
making
these
genes
candidates
for
a
new
mutation.
Nucleotide
sequence
analysis
of
the
thalassemia
gene
associated
with
haplotype
Na
revealed
a
C
--
T
substitution
at
position
-88
in
the
promoter
region
of
the
gene,
contributing
to
the
,3+-
thalassemia
phenotype
observed
in
the
patient.
The
thalassemia
allele
linked
to
haplotype
I
was
shown
by
DNA
sequence
analysis
to
have
a
G
-*
C
transversion
in
exon
1,
at
the
second
position
of
codon
30
(which
is
split
at
this
site
by
IVS1),
as
the
only
abnormality.
Codon-30
Mutation
Results
in
fi+-Thalassemia.
In
vivo
characterization
of
codon-30
mutation.
As
it
was
difficult
to
predict
the
nature-f3+
or
f30-of
this
type
of
defect
that
modifies
the
5'
consensus
sequence
of
the
first
exon/intron
junction
of
the
13-globin
gene,
additional
studies
were
per-
formed
to
explore
its
phenotypical
consequences
in
vivo
and
in
vitro.
The
codon-30
mutation
alters
the
coding
potential
of
exon
1:
the
sequence
AGG
(Arg)
is
changed
to
ACG
(Thr)
and
is
thus
expected
to
lead
to
the
production
of
a
structurally
abnormal
hemoglobin
if
some
normal
(3-globin
mRNA
is
produced.
However,
no
hemoglobin
variant
was
observed
by
standard
hemoglobin
isoelectric
focusing
analysis.
A
possible
explanation,
compatible
with
the
production
of a
low
level
of
mature,
translatable
/3-globin
mRNA,
could
be
related
to
the
instability
of
the
abnormal
protein
product.
Indeed,
codon
30
is
located
in
an
area
of
contact
between
the
hemoglobin
subunits
a,
and
P1,
and
examples
of
unstable
f3
mutants
at
this
site,
which
might
affect
the
ai/13l
contacts,
are
known
(2):
for
example,
hemoglobin
Tacoma
((830
Arg
-*
Ser)
(33).
We
first
confirmed
the
presence
or
absence
of
the
codon-30
mutation
in
family
members
by
analysis
of
PCR-amplified
(8-globirq
DNA
with
oligonucleotide
probes
specific
for
the
normal
and
mutated
sequences.
The
hematologic
data
of
simple
het-
erozygotes
for
this
mutation
were
typical
of
a
83-thalassemia
trait
and
were
inconsistent
with
the
presence
of
a
significant
hemolytic
process
due
to
an
unstable
hemoglobin.
Hemoglo-
bin
analysis
was
performed
on
fresh
blood
subsequently
obtained
from
the
patient
and
from
another
member
of
the
family,
who
had
inherited
the
same
thalassemia
allele
along
with
the
/3s-globin
gene,
in
order
to
detect
any
hemoglobin
mutant
that
might
comigrate
with
either
Hb
A
or
acetylated
Hb
F.
The
pattern
of
the
hemoglobins
and
of
the
globin
chains
was
analyzed
by
isoelectric
focusing
(34)
(Fig.
1)
and
elec-
trophoresis
in
Triton/acid/urea
(35)
(data
not
shown),
re-
spectively.
These
methods
revealed
that
a
structurally
ab-
normal
hemoglobin,
migrating
on
isoelectric
focusing
gels
at
a
position
slightly
anodic
to
Hb
A,
which
we
called
hemo-
globin
Kairouan,
is
produced
at
a very
low
level
(<2%).
This
indicates
that
this
G
--
C
substitution,
which
is
immediately
upstream
of
the
normal
5'
cleavage
site,
does
not
totally
inactivate
the
IVS1
donor
site.
In
vitro
splicing
in
the
cellfree
extract.
Taken
together,
the
previous
observations
suggested
that
the
mutation
of
the
guanine
at
position
-1
relative
to
the
5'
splice
site
of
IVS1
could
result
in
a
strong
impairment
of
the
splicing
reaction
of
the
first
intron.
To
analyze
this
possibility,
we
used
an
in
vitro
system,
which
reproduces
the
effects
of
mutations
on
splicing
of
human
83-globin
transcripts
in
vivo
(11,
30),
and
RNA
transcripts
originating
from
the
5'
portion
of
the
(8-globin
gene.
The
normal
83-globin
transcripts,
the
mutant
IVS1-
(-1C),
and
another
mutant,
IVS1(+2G),
which
contains
a
T
-+
G
transversion
at
position
+2
(29),
were
tested.
The
identity
of
intermediates
and
final
products
was
determined
by
the
following
criteria:
size,
aberrant
electrophoretic
mo-
bility,
kinetic
behavior,
and
analysis
of
the
mRNA
species
by
primer
extension
experiments.
With
wild-type
(H(3G)
transcripts
(Fig.
2A),
we
observed
two
intermediates
that
accumulated
within
30
mi
of
incu-
bation.
The
RNA
species
of
291
nt
is
similar
in
length
to
normal
exon
1
(Nor
El).
The
second
intermediate,
IVS-exon
Proc.
Natl.
Acad.
Sci.
USA
86
(1989)

Proc.
Natl.
Acad.
Sci.
USA
86
(1989)
1043
,_
H
vG
IVS1(-iC)
IVS1(+2G)
M
0
0.5
1
2
0
0.5
1
2
0
0.5
1
2
(h)
_~~~~~~~~~-9-
+ _
-
b
3U
Cr.
3
E
21
-N
o
r
E
E
--500
-Cr.2
j
'485
-Cr.2
IVSIU
-433
----Nor
IVS9
-390
-w-ww-
Nor
F3
-
-Cr.
3
[
Cr.2gE
-
x
-
307
-291
--277
-
257
'
Cr.2
IVS
-
_---
Nor
IVS
SP6
Tr.
(625-nt)
mRNAs
Cr.l:
457-nt
5'S.S.
I
I~~~~~~~~~~~~~
Rsal
+1
Cr.1
Cr.2
Nor
Cr.3
4
4
"',
kI
3'S.S.
BamHI
Exon
1
IVS
Exon
2
245
I
168
212
Cr.2:
479-nt
267
146
212
o-r:
495-nt
283
1
Cr.3:
507-nt
295
212
FIG.
2.
Time
course
of
in
vitro
splicing
of
SP6-generated
normal
(HlfG)
and
mutant
human
f3-globin
transcripts.
(A)
32P-labeled
transcripts
were
incubated
in
splicing
extracts
in
the
presence
of
42
mM
KCI
and
2.6
mM
MgCI2,
in
order
to
obtain
maximum
splicing
efficiency
(31).
Times
of
reaction
are
indicated
in
hours
(h)
above
each
lane.
The
identity
of
each
of
the
RNA
species
is
indicated
(El
and
E2,
exons
1
and
2;
Nor,
normal;
Cr.,
cryptic;
see
text).
The
lengths
of
the
products
(in
nucleotides,
nt)
generated
from
the
three
transcripts
are
indicated
at
right.
With
all
transcripts,
we
detected
a
linear
fragment
of
257
nt
(X)
that
could
be
due
to
nonspecific
RNA
cleavage,
as
observed
previously
(30).
Lane
M,
mixture
of
32p-
labeled
Taq
I
and
Msp
I
fragments
of
pBR322;
these
markers
range
from
the
doublet
at
622-616
nt
to
the
band
at
147
nt.
(B)
Schematic
representation
of
the
SP6-generated
P-globin
transcript
(SP6
Tr).
Cross-hatched
boxes,
transcribed
pSP64
sequences;
hatched
boxes,
the
human
sequences
that
behave
as
exons;
open
box,
the
normal
,8-globin
IVS1.
Position
+1
refers
to
the
cap
site
of
the
P-globin
pre-mRNA.
The
positions
of
normal
(Nor)
and
cryptic
(Cr.
1,
2,
and
3)
5'
splice
sites
are
indicated.
Lengths
of
the
resulting
mRNAs
and
of
the
exons
and
introns
are
indicated.
2,
has
an
apparent
length
of
390
nt
and
displays
an
aberrant
mobility
when
analyzed
in
gels
of
higher
concentrations.
The
two
other
products,
accumulating
later,
have
apparent
lengths
of
500
and
140
nt,
exhibiting
the
expected
character-
istics
for
normal
mRNA
(Nor
E1-E2)
and
excised
intron
(Nor
IVS).
It
has
been
shown
in
vivo
(9)
and
in
vitro
(30)
that
inactivation
of
the
normal
5'
splice
site
of
the
human
B-globin
gene
results
in
the
activation
of
three
cryptic
5'
splice
sites
located
16
or
38
nt
upstream
and
13
nt
downstream
of
the
normal
site
[cryptic
sites
(Cr.)
1,
2,
and
3,
respectively
(Fig.
2B)],
with
cryptic
splice
site
2
being
the
most
efficient.
When
we
used
the
IVS1(+2G)
mutant,
where
the
invariant
GT
dinucleotide
is
replaced
by
GG,
the
four
normal
splicing
products
were
absent but
four
other
products
were
detected,
indicating
that
another
splicing
reaction
had
occurred
(Fig.
2A).
The
exon
1
species
(Cr.2
El)
and
the
mRNA
species
(Cr.2
mRNA)
are
-15
nt
shorter
than
the
wild-type
products,
indicating
that
the
cryptic
5'
splice
site
at
position
-16
is
used
predominantly
under
our
conditions.
The
two
other
products
(IVS-E2
and
IVS)
display
the
characteristics
expected
for
the
use
of
this
cryptic
splice
site.
The
IVS1(-lC)
mutant
showed
a
more
complicated
splic-
ing
pattern.
The
same
four
spliced
products
observed
with
the
previous
mutant
were
obtained,
but
they
were
produced
less
efficiently
(Fig.
2A).
In
addition
to
the
Cr.2
mRNA
of
483
nt,
we
detected
two
faint
RNA
species
of
500
and
515
nt
whose
presence
is
consistent
with
the
utilization
of
the
normal
5'
splice
site
and
another
site
located
-15
nt
down-
stream,
respectively.
The
structure
and
amount
of
the
differently
spliced
mRNAs
were
determined
by
primer
extension
analysis
in
the
presence
of
a
saturating
concentration
of
one
ddNTP,
allow-
ing
a
complete
(with
ddATP)
or
almost
complete
(with
ddGTP)
termination
at
the
first
complementary
residue
encountered
on
the
template
RNA.
The
position
of
expected
stops
is
marked
in
Fig.
3B
for
the
putative
mRNA
species
and
unspliced
transcript,
and
the
identity
of
these
RNAs
is
directly
indicated
in
Fig.
3A.
With
the
HpG
transcript,
only
the
normal
mRNA
was
detected.
As
expected
from
direct
RNA
analysis
(Fig.
2A),
a
very
low
amount
of
normally
spliced
mRNA
was
obtained
with
the
IVS1(-lC)
mutant
(Fig.
3A).
In
addition,
Cr.2
and
Cr.3
mRNAs
were
also
well
detected.
With
the
IVS1(+2G)
mutant,
absolutely
no
stop
corresponding
to
the
normal
mRNA
was
detected,
whereas
the
Cr.2
mRNA
was
still
more
efficiently
spliced
than
with
the
other
mutant
(Fig.
3A).
Finally,
upon
longer
exposure
of
the
gel
(data
not
shown),
the
splicing
reaction
at
the
cryptic
5'
splice
site
(Cr.1)
was
also
faintly
detected
with
both
mutant
transcripts.
To
obtain
more
precise
information
about
the
use
of
the
normal
or
cryptic
5'
splice
sites,
we
analyzed
the
intensity
of
all
cDNA
bands
detected
in
the
autoradiogram
of
Fig.
3A
by
densitometry.
The
frequencies
of
utilization
of
the
different
5'
splice
sites
are
given
in
Table
1.
The
mutation
of
the
guanine
at
position
-1
results
in
an
almost
complete
(98%)
inactivation
of
the
normal
5'
splice
site,
whereas
the
inacti-
vation
is
complete
with
the
IVS1(+2G)
mutant;
these
results
were
confirmed
by
the
direct
quantification
of
the
mRNA
bands
shown
in
Fig.
2.
Strikingly,
as
observed
in
Figs.
2A
and
3A,
the
degree
of
activation
of
the
cryptic
sites
is
dependent
on
the
position
of
the
mutations
in
the
wild-type
5'
splice
site
(Table
1).
DISCUSSION
Here
we
describe
the
isolation
and
characterization
of
a
mutation
resulting
in
l3+-thalassemia.
The
Mediterranean
patient
studied
here
has
a
mild
thalassemia
due
to
the
interaction
of
two
f8-thalassemia
genes.
One
gene
has
a
C
--
T
substitution
at
position
-88
relative
to
the
cap
site,
a
mutation
already
described
in
Black
(37)
and
Asian
Indian
(38)
patients
as
leading
to
only
mild
impairment
of
8-globin
expression.
The
finding
of
this
molecular
lesion
in
our
patient
from
Tunisia,
associated
with
a
different
p-globin
gene
framework
(23),
illustrates
further
the
occurrence
of
the
same
defect
in
different
ethnic
groups
and
can
best
be
explained
by
independent
origins
of
the
same
mutation
(38).
The
other
Medical
Sciences:
Vidaud
et
al.

1044
Medical
Sciences:
Vidaud
et
al.
A
m
20_3
15-
1-
1
0-
go
4-
3
-_--
2-
H
13G
C.U
IVS1
(-Ic)
:C
:
U
C
r.2
Cr.2m
-..aNor.
-Nor.
Tr.
L>
a
Z~~~~~10
___r-
Xr_
Cr.3---
sAw3v
Nor.w-q
<Tr.
)
<C
C
r.2
Pr.
(
)
Pr.
)
Pr.
HBG
Transcript
axon
1
IVS1
(130-nt)
ACCXGUGAUGCAGUGGUG
,'
GUGAGGCCCUGGGCAGC7
guugguaucaag
Iquuaca
---
---
-
__ __
__
__
_----
I___
__
__
Ir
No
5S
C.
4-Cr.1
Cr.2
Nor
555s
CO.
HBG
Tr.
Nor
mRNA:
CUGGGCAG
t
tf
(U+8)
(C+4)
|IVS1
(-lc)
Tr-.
Cr.3
mRNA:
gguaucaaq
(u+6)
(c+5)
"Nor'mRNA:
CUGGGCI.
Cr2
mRNA:
(U+8)
(C+2)
ACGUGGAUGAAGUUGGUG,
(C+18)
(U+3)
IVSI
(+2G)
Tr.
=
1VS1
(-
C)
for
Cr.2
and
Cr.3
mRNAs
FIG.
3.
Primer
extension
analysis
of
the
splic
normal
and
mutant
transcripts.
(A)
The
transcripts
vitro
for
2
hr,
and
the
resulting
RNAs
served
as
temp
transcription
using
32P-labeled
primer
complemental
24
of
exon
2
(see
B).
Primer
extension
was
in
the
pre
(lanes
or
of
ddATP
(lanes
U).
For
each
assay,
rt
tion
of
unprocessed
transcript
(Tr.)
results
in
termi
by
open
arrowheads.
Second
residual
termination
M
unprocessed
transcript
or
from
correctly
spliced
mR
dots
at
positions
7
or
9
nucleotides
upstream
fro
primer,
respectively.
Lane
M,
products
of
primer
initial
transcript
in
the
presence
of
a
mixture
of
the
ddNTPs.
(B)
Schematic
representation
of
the
c
according
to
the
utilization
of
normal
or
cryptic
5'
mutant
gene
is
studied
in
this
paper
and
is
ir
number
of
reasons.
VS
1
(+2G)
Until
now,
only
one
mutation
has
been
revealed
in
the
exon
_-
T__-uresidues
of
the
5'
splice
sites
of
the
P-globin
gene
(39),
whereas
several
point
mutations
have
been
detected
in
the
intron
residues.
The
G
--
C
transversion
at
position
-1
virtually
abolishes
(98%
inhibition)
the
utilization
of
the
normal
5'
splice
site.
Significantly,
the
level
of
the
correctly
spliced
mRNA
is
comparable
to
that
of
hemoglobin
carrying
the
Arg
-*
Thr
mutation
found
in
the
erythrocytes
of
the
patient.
The
severity
of
the
mutation
is
unexpected
if
we
consider
that
the
mutated
5'
splice
site
still
contains
a
6-nt
match
with
the
consensus.
This
results
in
a
score
of
67
(Table
1),
a
value
found
for
certain
authentic
5'
splice
sites
(6).
S
i
However,
two
observations
might
explain
the
importance
of
the
inactivation
of
the
mutated
5'
splice
site.
First,
as
the
6
=<
nt
that
match
the
consensus
are
not
contiguous,
the
Gibbs
I.wCr.2
energy
value
for
the
Ul
snRNA-5'
splice
site
interactions
is
considerably
lower
with
the
mutated
site
(AG
=
-2.5
(U
kcal/mol)
than
with
the
wild-type
site
(AG
=
-7.9
kcal/mol)
(Table
1).
Second,
competition
of
neighboring
cryptic
sites
with
the
mutated
site
for
splicing
could
accentuate
the
inhibition
of
the
mutated
5'
site.
exon
2
On
the
whole,
it
appears
that
the
drastic
inhibition
of
the
cccuua.iAE
normal
splice
site
caused
by
the
G
C
transversion
at
[E!Pr:me
position
-1
is
similar
to
or
higher
than
that
found
by
in
vivo
(+6),
f(u+4)
or
in
vitro
analysis
of
mutations
of
the
guanine
at
position
+5
cccuua5GCGC=U
(9,
11,
30,
40).
Therefore,
our
results
indicate
that
the
----------
Pr.
contribution
of
the
last
nucleotide
of
the
5'
exon
is
at
least
as
(c+6))
(u+4)
important,
if
not
more
so,
than
that
of
the
nucleotide
at
cccuuagGCUGCU
position
+5.
---*~~~
Pr.
Up
to
now,
the
indications
that
exon
residues
of
the
5'
-GCUGCU
splice
site
play
a
role
equivalent
to
that
of
the
intron
residues
------------
Pr.
were
rather
indirect.
Iida
and
Sasaki
(41)
described
a
set
of
,GCUGCU
four
5'
minimal
consensus
sequences
(/GTAAGT,
RG/GT-
--------
XPr.
GAG,
AG/GTNNGT,
and
AG/GTA)
that
shows
that
the
decreased
homology
to
the
consensus
in
the
intron
portion
is
compensated
by
increased
homology
to
the
exon
part
of
the
ing
products
of
5'
splice
site.
However,
of
the
1389
5'
splice-site
sequences
were
spliced
in~
recently
analyzed,
only
58%
corresponded
to
at
least
one
of
lates
for
reverse
these
four
minimal
consensus
sequences
(6).
The
most
direct
ry
to
positions
2-
evidence
of
the
contribution
of
exon
residues
to
the
consti-
esence
of
ddGTP
tution
of
an
active
5'
splice
site
could
be
deduced
from
the
everse
transcrip-
analysis
of
j3-thalassemia.
It
has
been
shown
that
a
T
-*
G
ination
indicated
mutation
at
position
705
of
IVS2
activates
a
cryptic
5'
splice
vith
ddGTP
from
site,
AG/GTAAGA,
in
which
the
new
guanine
is
located
at
NA
is
marked
by
position
-1
(42,
43).
Similarly,
the
cryptic
5'
splice
site
2
in
Em
the
5'
end
of
exon
1
can
be
activated
by
a
T
-+
A
mutation
at
position
-2
extension
of
the
of
the
5'
splice
site
(44).
Therefore,
results
and
the
above
expected
cDNA
observations
are
consistent
with
the
proposal
that
the
exon
splice
sites.
and
intron
residues
of
5'
splice
site
can
play
a
similar
role
in
the
constitution
of
an
efficient
splice
site.
iteresting
for
a
An
important
point
that
is
not
well
understood
is
the
existence
of
a
hierarchy
in
the
5'
splice
sites,
which
has
been
Table
1.
Efficiency
of
utilization
of
the
various
5'
splice
sites
in
normal
and
mutant
transcripts
AG,
Relative
efficiency,
%
5'
splice
site
Sequence
Score
kcal/mol
Normal
IVS1(-1C)
IVS1(+2G)
Normal
IVS1
C
A
GIG
T
T
G
79
-7.9
100
IVS1(-1C)
A
C/CI
T
C
67
-2.5
1.7
IVS1(+2G)
C
A
G
G
T
G
70
-4.9
0
Cryptic
1
(-38)
A
A
GIG
T
G
A
A
C
76
-4.5
<0.1
0.1
0.5
Cryptic
2
(-16)
G
T
G
AG
T
G
84
-4.5
<0.1
7.6
15.1
Cryptic
3
(+13)
A
A.GLG.T
T
A
C
A
70
-3.3
<0.1
3.2
1.3
Normal
IVS2
A
G
GLG
T
G
A
G
T
86
-8.1
Consensus
C
A
C/G
T
A
A
G
T
100
-9.9
The
potential
base
pairing
between
the
5'
splice
region
and
the
U1
snRNA
is
indicated
by
underlining
the
nucleotides
of
the
5'
splice
site.
G-U
base
pairs
are
indicated
with
dots.
The
AG
values
(in
kcal/mol;
1
kcal
=
4184
J)
for
the
5'
splice
region-Ul
snRNA
interaction
were
determined
according
to
Freier
et
al.
(36)
except
that
Gibbs
energy
increments
for
unpaired
terminal
nucleotides
were
not
included.
Calculation
of
the
score
(between
0
and
100)
was
according
to
Shapiro
and
Senapathy
(6).
The
relative
5'
splice
site
efficiencies
for
each
transcript
are
given
as
a
percentage
of
that
of
the
correct
spliced
mRNA
formed
from
normal
(HOG)
transcript.
Proc.
Natl.
Acad
Sci.
USA
86
(1989)

Proc.
Natl.
Acad.
Sci.
USA
86
(1989)
1045
demonstrated
after
competition
assays
(18,
20).
Suboptimal
splice
sites
might
be
required
to
regulate
alternative
splicing,
as
is
likely
the
case
for
adenoviral
ElA
or
simian
virus
40
early
pre-mRNA
(15,
19).
However,
there
is
no
clear
advan-
tage
for
a
highly
conserved
gene
such
as
the
f-globin
gene.
It
could
be
imagined
that
the
existence
of
an
intrinsically
suboptimal
5'
splice
site
like
that
of
the
IVS1
of
human
,8-globin
could
be
compensated
for
by
the
intervention
of
other
determinants
such
as
sequence
context
(20).
This
is
not
the
case,
since
a
single
mutation
that
decreases
the
match
of
the
5'
splice
site
to
the
consensus
(ref.
9
and
this
work)
or
that
increases
the
match
of
a
cryptic
5'
splice
site
(44,
45)
drastically
alters
the
correct
splicing
of
the
f3-globin
pre-
mRNA.
We
thank
Evelyne
Brouquet
for
her
assistance
with
preparation
of
the
manuscript
and
John
White
for
critically
reading
it.
We
are
indebted
to
F.
Ellouze
and
M.
Farhat
for
their
cooperation
during
the
course
of
these
studies.
We
thank
F.
Plassa,
G.
Hildwein,
J.
Martin,
and
C.
Godard
for
skilled
technical
assistance.
This
work
was
supported
by
the
Institut
National
de
la
Sante
et
de
la
Recherche
Medicale,
the
Centre
National
de
la
Recherche
Scientifique,
and
the
program
"Hdmoglobinopathies
en
Tunisie"
of
the
French
Ministere
de
la
Cooperation.
1.
Weatherall,
D.
J.
&
Clegg,
J.
B.,
eds.
(1981)
The
Thalassemia
Syndromes
(Blackwell,
Oxford).
2.
Bunn,
H.
F.
&
Forget,
B.
G.
(1986)
Hemoglobin:
Molecular
Genetic
and
Clinical
Aspects
(Saunders,
Philadelphia).
3.
Stamatoyannopoulos,
G.,
Nienhuis,
A.
W.,
Leder,
P.
&
Ma-
jerus,
P.
W.
(1987)
The
Molecular
Basis
of
Blood
Diseases
(Saunders,
Philadelphia).
4.
Mount,
S.
M.
(1982)
Nucleic
Acids
Res.
10,
459-472.
5.
Oshima,
Y.
&
Gotoh,
Y.
(1987)
J.
Mol.
Biol.
195,
247-259.
6.
Shapiro,
M.
B.
&
Senapathy,
P.
(1987)
Nucleic
Acids
Res.
15,
7155-7175.
7.
Green,
M.
R.
(1986)
Annu.
Rev.
Genet.
20,
671-708.
8.
Padgett,
R.
A.,
Grabowski,
P.
J.,
Konarska,
M.
M.,
Seiler,
S.
&
Sharp,
P.
A.
(1986)
Annu.
Rev.
Biochem.
55,
1119-1150.
9.
Treisman,
R.,
Orkin,
S.
H.
&
Maniatis,
T.
(1983)
Nature
(London)
302,
591-596.
10.
Wieringa,
B.,
Meyer,
F.,
Reiser,
J.
&
Weissman,
C.
(1983)
Nature
(London)
301,
38-43.
11.
Atweh,
G.
F.,
Wong,
C.,
Reed,
R.,
Antonarakis,
S.
E.,
Ding-
er,
Z.,
Ghosh,
P.
K.,
Maniatis,
T.,
Forget,
B.
G.
&
Kazazian,
H.
H.
(1987)
Blood
70,
147-151.
12.
Kramer,
A.,
Keller,
W.,
Appel,
B.
&
Luhrmann,
R.
(1984)
Cell
38,
299-307.
13.
Chabot,
B.,
Black,
D.
L.,
Le
Master,
D.
M.
&
Steitz,
J.
A.
(1985)
Science
230,
1344-1349.
14.
Krainer,
A.
R.
&
Maniatis,
T.
(1985)
Cell
42,
725-736.
15.
Zhuang,
Y.
&
Weiner,
A.
M.
(1986)
Cell
46,
827-835.
16.
Aebi,
M.,
Hornig,
H.,
Padgett,
R.
A.,
Reiser,
J.
&
Weissmann,
C.
(1986)
Cell
47,
555-565.
17.
Aebi,
M.,
Hornig,
H.
&
Weissmann,
C.
(1987)
Cell
50,
237-246.
18.
Eperon,
L.
P.,
Estibeiro,
J.
P.
&
Eperon,
I.
C.
(1986)
Nature
(London)
324,
280-282.
19.
Zhuang,
Y.,
Leung,
H.
&
Weiner,
A.
M.
(1987)
Mol.
Cell.
Biol.
7,
3018-3020.
20.
Nelson,
K.
K.
&
Green,
M.
R.
(1988)
Genes
Dev.
2,
319-329.
21.
Goossens,
M.
&
Kan,
Y.
W.
(1981)
Methods
Enzymol.
76,
805-
817.
22.
Chibani,
J.,
Vidaud,
M.,
Duquesnoy,
P.,
Berg&-Lefranc,
J.
L.,
Pirastu,
M.,
Ellouze,
F.,
Rosa,
J.
&
Goossens,
M.
(1988)
Hum.
Genet.
78,
190-192.
23.
Orkin,
S.
H.,
Kazazian,
H.
H.,
Antonarakis,
S.
E.,
Goff,
S.
C.,
Boehm,
C.
D.,
Sexton,
J.
P.,
Waber,
P.
G.
&
Giardina,
P.
J.
V.
(1982)
Nature
(London)
296,
627-631.
24.
Nicholls,
R.
D.,
Hill,
A.
V.
S.,
Clegg,
J.
B.
&
Higgs,
D.
R.
(1985)
Nucleic
Acids
Res.
13,
7569-7578.
25.
Sanger,
F.,
Nicklen,
S.
&
Coulson,
A.
R.
(1977)
Proc.
Natl.
Acad.
Sci.
USA
74,
5463-5467.
26.
Saiki,
R.
K.,
Bugawan,
T.
L.,
Horn,
G.
T.,
Mullis,
K.
B.
&
Erlich,
H.
A.
(1986)
Nature
(London)
324,
163-166.
27.
Saiki,
R.
K.,
Gelfand,
D.
H.,
Stoffel,
S.,
Scharf,
S.
J.,
Higuchi,
R.,
Horn,
G.
T.,
Mullis,
K.
B.
&
Erlich,
H.
A.
(1988)
Science
239,
487-491.
28.
Amselem,
S.,
Nunes,
V.,
Vidaud,
M.,
Estivill,
X.,
Wong,
C.,
d'Auriol,
L.,
Vidaud,
D.,
Galibert,
F.,
Baiget,
M.
&
Goossens,
M.
(1988)
Am.
J.
Hum.
Genet.
43,
95-100.
29.
Vidaud,
M.,
Vidaud,
D.,
Chibani,
J.,
Amselem,
S.,
Duquesnoy,
P.,
Rosa,
J.
&
Goossens,
M.
(1987)
Blood
70,
Suppl.,
69A
(abstr.).
30.
Krainer,
A.
R.,
Maniatis,
T.,
Ruskin,
B.
&
Green,
M.
R.
(1984)
Cell
36,
993-1005.
31.
Schmitt,
P.,
Gattoni,
R.,
Keohavong,
P.
&
Stevenin,
J.
(1987)
Cell
50,
31-39.
32.
Pikielny,
C.
W.
&
Rosbash,
M.
(1985)
Cell
41,
119-126.
33.
Brimhall,
B.,
Jones,
R.
T.,
Baur,
E.
W.
&
Motulsky,
A.
G.
(1969)
Biochemistry
8,
2125-2129.
34.
Dubart,
A.,
Goossens,
M.,
Beuzard,
Y.,
Monplaisir,
N.,
Testa,
U.,
Basset,
P.
&
Rosa,
J.
(1980)
Blood
56,
1092-1099.
35.
Alter,
B.,
Goff,
S.
C.,
Efremov,
G.
D.,
Gravely,
M.
E.
&
Huisman,
T.
H.
J.
(1980)
Br.
J.
Haematol.
44,
527-534.
36.
Freier,
S.
M.,
Kierzek,
R.,
Jaeger,
J.
A.,
Sugimoto,
N.,
Caru-
thers,
M.
H.,
Neilson,
T.
&
Turner,
D.
H.
(1986)
Proc.
Natl.
Acad.
Sci.
USA
83,
9373-9377.
37.
Orkin,
S.
H.,
Antonarakis,
S.
E.
&
Kazazian,
H.
H.
(1984)
J.
Biol.
Chem.
259,
8679-8681.
38.
Wong,
C.,
Antonarakis,
S.
E.,
Goff,
S.
C.,
Orkin,
S.
H.,
Boehm,
C.
D.
&
Kazazian,
H.
H.
(1986)
Proc.
Natl.
Acad.
Sci.
USA
83,
6529-6532.
39.
Chehab,
F.
F.,
Der
Kaloustian,
V.,
Khoury,
F.
P.,
Deeb,
S.
S.
&
Kan,
Y.
W.
(1987)
Blood
69,
1141-1145.
40.
Lapoumeroulie,
C.,
Acuto,
S.,
Rouabhi,
F.,
Labie,
D.,
Krish-
namoorthy,
R.
&
Bank,
A.
(1987)
Nucleic
Acids
Res.
15,
8195-
8204.
41.
lida,
Y.
&
Sasaki,
F.
(1983)
J.
Biochem.
94,
1731-1738.
42.
Dobkin,
C.,
Pergolizzi,
R.
G.,
Bahre,
P.
&
Bank,
A.
(1983)
Proc.
Natl.
Acad.
Sci.
USA
80,
1184-1188.
43.
Dobkin,
C.
&
Bank,
A.
(1985)
J.
Biol.
Chem.
260,
16332-16337.
44.
Goldsmith,
G.
E.,
Humphries,
R.
K.,
Ley,
T.,
Cline,
A.,
Kantor,
J.
&
Nienhuis,
A.
W.
(1983)
Proc.
Natl.
Acad.
Sci.
USA
80,
2318-2322.
45.
Orkin,
S.
H.,
Kazazian,
H.
H.,
Antonarakis,
S.
E.,
Ostrer,
H.,
Goff,
S.
C.
&
Sexton,
J.
P.
(1982)
Nature
(London)
300,
768-
769.
Medical
Sciences:
Vidaud
et
al.