close

Вход

Забыли?

вход по аккаунту

?

Homozygous hereditary C1q deficiency and systemic lupus erythematosusA new family and the molecular basis of C1q deficiency in three families.

код для вставкиСкачать
ARTHRITIS & RHEUMATISM
Vol. 39, No.4, April 1996, pp 663-670
0 1996, American College of Rheumatology
663
HOMOZYGOUS HEREDITARY C l q DEFICIENCY AND
SYSTEMIC LUPUS ERYTHEMATOSUS
A New Family and the Molecular Basis of Clq Deficiency in Three Families
JASON H. SLINGSBY, PETER NORSWORTHY, GLEN PEARCE, AKSHAY K. VAISHNAW,
HELEN ISSLER, BERNARD J. MORLEY, and MARK J. WALPORT
Objective. To describe a new kindred with Clq
deficiency and to identify the molecular lesions responsible for complete functional Clq deficiency in this and
2 other previously described kindreds.
Methods. The A-, B-, and C-chain genes of Clq
were amplified by polymerase chain reaction, cloned,
and sequenced. The DNA sequence was checked for
mutations.
Results. Patient 1 had a homozygous G-to-A
change at codon 6 of the C chain, causing an amino acid
change from Gly to Arg. Patient 2 had a homozygous
deletion of a C nucleotide at codon 43 of the C-chain,
causing a frame shift, leading to a premature stop codon
at codon 108. Patient 3 had a homozygous C-to-T
mutation at amino acid position 41 of the C chain,
resulting in a premature stop codon.
Conclusion. In the homozygous state, the mutations are sufficient to cause complete deficiency of Clq.
The mutation in patient 1 has been previously reported
in a patient of different ethnic origin. A survey of a
series of 158 DNA samples from patients with systemic
lupus erythematosus showed no other examples of this
mutant allele.
Homozygous C l q deficiency has been described in 30 people, 28 of whom had systemic lupus
Supported by grants from the Arthritis and Rheumatism
Council. Mr. Slingsby's work was supported by a PhD Studentship
sponsored by the Arthritis and Rheumatism Council for Research.
Jason H. Slingsby, BA, Peter Korsworthy, BSc, Glen
Pearce, Akshay K. Vaishnaw, MRCP, Bernard 3. Morley, BSc,
DPhil, Mark J. Walport. PhD, MRCP: Hammersmith Hospital,
London, UK; Helen Issler: Brook General Hospital, London, UK.
Address reprint requests to Prof. Mark J. Walport, Rheumatology Unit, Department of Medicine, RPMS, Hammersmith
Hospital, Du Cane Road, London, W12 ONN, UK.
Submitted for publication August 7, 1995; accepted in
revised form November 15, 1995.
erythematosus (SLE), 1 of whom had discoid lupus,
and 1 of whom was a healthy 38-year-old (1). Inherited
C l q deficiency represents the most powerful disease
susceptibility gene so far identified for the development of SLE in humans, although it is an extremely
rare cause of the disease as a whole. The autoantibody
profile found in patients with homozygous C l q deficiency closely resembles that seen in patients with
S L E without homozygous complement deficiency.
Seventeen of 23 patients with C l q deficiency in whom
a result was reported had antinuclear antibodies
(ANA), and autoantibodies to extractable nuclear antigen (ENA) were identified in 12 (1).
C l q is encoded by 3 genes, ClqA, ClqB, and
ClqC, each comprising 2 exons and organized tandemly on chromosome 1 (2,3). The molecular basis for
C l q deficiency has been reported in 3 families, and in
1 was due to a point mutation that led to a premature
stop codon at residue 150 in the B-chain of C l q (4).In
the second family, there was a CAG-to-TAG change at
codon 186 in the A-chain, which also produced a
premature stop codon (5). The third family showed a
glycine-to-arginine substitution caused by a change
from a G to an A nucleotide in the first base position of
the sixth codon in the C-chain (5).
We describe here a further family with homozygous C l q deficiency in whom 2 siblings developed
cutaneous lupus erythematosus, accompanied by ANA
and anti-Ro antibodies. The molecular basis for C l q
deficiency was identified in this family and in 2 patients
whose cases have been reported previously (1,6).
PATIENTS AND METHODS
Antibodies. Polyclonal anticomplement reagents.
Sheep antisera to Clr, Cls, C4, C2, and C3 were obtained
from The Binding Site (Birmingham, UK).
664
SLINGSBY ET AL
Table 1. Oligonucleotides used during the study
Oligonucleotide
ClQAI5'
ClQA23'
K3
K4
ClQCl5'
ClQC23'
CIQC13'
JOK-1
JOK-2
JOK-3
JOK-4
JOK-5
JOK-6
Annealing
temp. ("C)
Oligonucleotide sequence
(5' to 3')
Oligo. position
codon number
B
55
55
55
55
C
C
58
58
CCGTGTCTCCACAGAGGCATC
GCATGGAAGCCAGAGAGGTG
ATTGTCTTCTAGACAGGAGGCGTCT
TGGATGTGAATTCGCCCACAGGTCA
CCTCCCAGTTCCTTCTCCGG
GTGGGTCTCGAGCGCATCTG
GAGCCAGGTGAGTCTGCTGG
GGTACGGTTCTTCCTGG'ITG
CTGAAGCACCATGCCCCCTG
ATGTTGGTGATCACGTGGTC
GTCACAGAAGGTGACCACC
GGGTGCAGGGGGCTGGTGG
GATTGGTMTGGACGTGTGG
-29 to -22
240 to 233
-38 to -31
235 to 227
-21 to -27
227 to 221
37 to 31
123 to 115
186 to 180
122 to 116
172 to 166
106 to loo
172 to 166
Chain
A
A
B
c
A
A
H
B
c
<:
55
55
55
55
55
62
62
* The annealing temp. is the temperature used under most circumstances. The location of the
oligonucleotide (oligo. position) in the coding sequence is given according to the codon numbering
system as described elsewhere (2.3).
Monoclonal anti-Clq reagent. IgG mouse monoclonal anti-Clq (1A4) was a gift from Dr. Philip Shephard
(UDMS, Guy's Hospital, London, UK) (7).
Complement assays. CH50 and AP50. Classical pathway hemolytic complement activity was measured by plate
assay, using antibody-sensitized sheep erythrocytes, as previously described (8). Alternative pathway hemolytic complement activity was measured by plate assay, using guinea
pig erythrocytes in magnesium EGTA buffer, as previously
described (8).
C l q . Plasma samples were initially screened by
double immunodiffusion assay using sheep polyclonal
anti-Clq (The Binding Site). Antigenic Clq levels were
measured by enzyme-linked immunosorbent assay (ELISA),
as previously described (1).
C l q repletion assay. Purified Clq (Sigma, Poole,
UK) over a concentration range of 0.625-20 p g h l was
added to patients' sera, and CH5O was estimated as described above. Control samples containing equivalent volumes of added buffer were assayed in parallel.
Functional Clq assay. A modification of the Clq
repletion assay was used to measure functional Clq activity
in serum samples. Using the Clq antigenic concentration in
the sera measured by ELISA, sera were diluted to concentrations of antigenic Clq that vaned over a concentration
range of 0.625-20 &ml and were added to homozygous Clq
deficient sera. The resulting lysis of antibody-coated erythrocytes in a plate assay was then compared with purified Clq
over the same concentration range.
Other component assays. Clr, CIS, C4, C2, and C3
levels were measured by single radial immunodiffusion or
rocket immunoelectrophoresis in 1.2% CFD/agarose, using
specific polyclonal antisera (The Binding Site).
Anti-Clq antibodies. Antibodies to Clq were measured by a modification of the method described by Siegert
and colleagues (9) as previously described (1).
DNA samples from patients with Clq deficiency and
SLE. Samples of genomic DNA were prepared (10) from
peripheral blood obtained from the propositus (patient 1) in
the family described here and from both his mother and his
father. To screen for the mutation detected in patient 1,
genomic DNA samples were used from 136 Caucasoid and
24 Afro-Caribbean patients who met the American College
of Rheumatology 1982 classification criteria for SLE (11).
Stored samples of genomic DNA were available from 2
deceased patients described previously, and in each case
from a single relative: patient 2, (described in ref. 1) and her
brother; and patient 3 (described in ref. 6) and his father.
Amplification of A-, B-, and C-chain genes of Clq by
polymerase chain reaction (PCR). PCR was used to amplify
the complete A, R , and C chains of Clq. This PCR product
contained 2 coding exons and the single intron. PCK amplification was camed out in a 100-/6 volume containing I X
Taq Extender Buffer (Stratagene, La Jolla, CA), 2.0 mM
MgCl,, dATP, dCTP, dGTP, and dTTP each at 100 pU, 150
ng of forward and reverse oligonucleotide primers, 200 ng of
genomic DNA, 10 units of Taq DNA polymerase (Gibco
BRL, Paisley, Scotland), and 10 units of Tuq Extender PCR
additive (Stratagene). We used the oligonucleotide primers
ClQAl5' and ClQA23' to amplify the A-chain, K3 and K4 to
amplify the B-chain, and ClQCl5' and CIQC23' to amplify
the C-chain (see Table 1). The thermocycling protocol was
as follows: 30 cycles at 94°C for 45 seconds, annealing
temperature for 45 seconds, and 72°C for 2 minutes 30
seconds, with a final cycle of 72°C for 10 minutes.
The amplification products were quantified on a 0.8%
agarose gel in 1x TBE (90mM Tris HCl, 90 mM boric acid,
2.5 mM disodium EDTA, pH 8.3). The PCR products were
either cloned prior to sequence analysis or were purified
using Wizard DNA Clean-Up System (Promega, Madison,
WI). The cleaned PCR products were used as template
DNA for sequence analysis using the fmol DNA sequencing
system, according to the manufacturer's instructions
(Promega).
Cloning of PCR products for sequence analysis. PCR
products were cloned into the pCR I1 vector using the
Original TA Cloning Kit, according to the manufacturer's
instructions (Invitrogen, San Diego, CA). Transformed bacteria were plated onto 135-mm LB agar plates (50 M m l of
ampicillin and 220 pl of a 2% solution of X-Gal [5-bromo-4-
665
HEREDITARY C l q DEFICIENCY IN SLE
chloro-3-indolyl-~~-galactoside]),
and after 18 hours of incubation at 3TC, 12 white colonies were selected and
plasmid DNA was extracted using a standard protocol (10).
Presence of the insert was confirmed by restriction enzyme
digestion of plasmid DNA.
Positive-clone DNA was purified for sequencing using Wizard DNA Clean-Up System. A combination of
cloned DNA and purified PCR products from amplified
genomic DNA was used as template DNA for sequence
analysis, using the fmoi DNA sequencing system, according
to the manufacturer’s instructions. The sequence was
checked for mutations against the published sequences.
Restriction enzyme analysis of exon 1 of the C-chain.
We amplified exon 1 of the C-chain by PCR, using the
oligonucleotide primers ClQCl5‘ and ClQC13‘ (see Table
11, from the genomic DNA of patient 1 and both his parents,
along with a panel of 158 genomic DNA samples from a
series of patients with SLE. PCR amplification was carried
out in a 2 0 - 4 volume containing 20 mM Tris HCl (pH 8.4),
50 mM KCI, 1.5 mM MgCI,, dATP, dCTP, dGTP, and d l T P
30 ng of forward and reverse oligonucleotide
each at 100
primers, 50 ng of genomic DNA, and 1 unit of Taq DNA
polymerase (Gibco BRL). The thermocycling protocol was
as follows: 30 cycles at 94°C for 30 seconds, 58°C for 1
minute, and 72°C for 45 seconds. The PCR products were
digested at 37°C for 1 hour in a 2 0 - 4 volume as follows: 1x
NEBuffer 4 (New England Biolabs, Beverly, MA), 5 pl of
PCR product, 0.1 mg/ml of bovine serum albumin, and 2
units of Sfc I (New England Biolabs). The fragments were
separated on a 3% Metaphor gel (Flowgen, Sittingbourne,
Kent, UK) in 1 x TBE buffer, by electrophoresis at 5OV for 1
hour. The fragments were visualized by staining with ethidium
bromide.
Case reports. Patient 1 and his sister. The patients
were 2 children, a male born in 1976 and a female born in
1988. Their parents were of Indian ethnic origin and were
consanguineously related. The male propositus presented 10
days after birth with umbilical sepsis, which was treated with
gentamicin and penicillin. At the age of 19 months, an
erythematous, scaly rash developed on his cheeks, forehead,
hands, buttocks, and knees. There was inflammation of the
nailbeds of his fingers and toes.
A biopsy of the affected skin showed hyperkeratosis,
with an increase in the stratum granulosum and basal layer
degeneration. Immunofluorescence analysis showed linear
deposition of IgM and fibrin at the dermal-epidemal junction. Blood tests revealed ANA. Treatment was started with
oral prednisolone, 7.5 mg daily.
At the age of 2, a further rash developed, comprising
erythema of both soles, palmar erythema, and telangiectasia
of the nailfolds of his fingers and toes. He was treated with
mepacrine. At the age of 6, systemic therapy was stopped,
although a rash persisted. At the age of 7, right parotid
swelling developed, and this became bilateral at the age of
12, when sialectasis was documented. The patient is currently well, aged 19, though has a persisting rash for which
he uses topical corticosteroids.
Blood tests have shown strongly positive ANA;
anti-Ro antibodies were first noted when the patient was age
13 and have remained present. Anticardiolipin antibodies
(aCL) were first measured when the patient was age 19 and
a,
Table 2. Complement levels in patient 1 and his parents
Complement protein*
Normal
range Father Mother Propositus
~
CH50 (% NHS)
50-125
97
AP50 (% NHS)
60-103
97
47-2 1 1
193
Clq (ELISA; mg/ml)
Clr (rocket IEP; % NHS)
74-1 13
133
103
CIS (rocket IEP; % NHS)
54-116
C2 (rocket IEP; % NHS)
64-108
82
C4 (Mancini; g d i t e r )
0.160.38 0.33
C3 (Mancini; gdliter)
0.60-1.35 I .39
Clinh (rocket IEP; % NHS) 60-135
101
91
I93
133
100
94
0.56
1.75
-
0
94
0
10s
110
111
0.49
0.92
135
* NHS = normal human serum; ELISA = enzymel-inked immunosorbent assay; IEP = immunoelectrophoresis;Clinh = C1 inhibitor.
were moderately positive for IgG aCL (22.6 arbitrary ELISA
units, normal <9.0 AEU); IgM aCL were negative (5.8
AEU, normal <8.0 AEU). Autoantibodies to C l q were also
first measured when the patient was age 19 and were
negative (33 AEU, normal <78 AEU). Antibodies to doublestranded DNA (dsDNA), La, Sm, RNP, and Jo-1 have been
persistently negative.
The patient’s sister presented at the age of 4 with
facial discoid lupus and photosensitivity. Investigations
showed positive ANA (titer 1:80) and weakly reactive ENA
(by counterimmunoelectrophoresis; specificity unidentified).
She is currently well, age 7, but has a persisting rash.
The results of assays of the complement system are
shown in Table 2. No C l q was detected by functional assay
or by ELISA using 1A4 (mouse monoclonal anti-Clq) as the
capture reagent. A band was detected by Ouchterlony
analysis using 2 polyclonal anti-Clq antisera (from The
Binding Site, and donated by Prof. Ken Reid, MRC Immunochemistry Unit, Oxford, UK), which showed complete antigenic identity with Clq from normal serum using the
antiserum from The Binding Site and partial antigenic identity using the antiserum from Prof. Ken Reid, corresponding
to the presence of dysfunctional C l q (Figure 1).
Serum from the propositus was reconstituted with
purified Clq. A CH5O of 54% of a pool of normal human
serum was achieved by addition of purified C l q at a final
concentration of 5 d m l . Antigenic concentrations of C l q
measured in both heterozygous parents of the 2 patients
were at the upper end of the normal range measured by
ELISA (Table 2), and this had the same level of functional
activity as C l q derived from 3 normal sera when measured in
a functional Clq reconstitution assay (data not shown).
RESULTS
The complete DNA sequence of the A-, B-, and
C-chain genes of Clq, including the splice donor and
splice acceptor sites, was determined for all 3 patients,
using a combination of cloned DNA and purified PCR
products of directly amplified genomic DNA as template DNA.
Findings in patient 1. A homozygous change
was detected from a G to an A nucleotide in the first
666
Figure 1. Double immunodiffusion analysis of the dysfunctional
Clq present in the serum of patient 1. A, The central well contains
polyclonal anti-CIq antiserum from Prof. Ken Reid (MRC Immunochemistry Unit, Oxford, UK). A line of partial identity is seen
between the Clq protein from the patient and Clq in a normal
control serum. B, The central well contains polyclonal anti-Clq
antiserum from The Binding Site (Birmingham, UK). A line of
complete identity is seen between the dysfunctional protein and the
antiserum.
base position of the sixth codon in the C chain (GGG
to AGG). This causes a change in the amino acid at
position 6 from a glycine to an arginine (Figure 2).
Both parents were demonstrated to be heterozygous
for this mutation by sequence analysis. This mutation
causes the gain of a recognition site for the restriction
enzyme Sfc I (C/TPuPyAG) (Figure 3). No other
coding polymorphism was found in the rest of the 3
chains.
Figure 2. DNA sequence of C-chain polymerase chain reaction
product from genomic DNA of a normal individual, the father of
patient 1, and patient 1, using oligonucleotide ClQCl5'. The change
of a G to an A nucleotide in the first base position of the sixth codon
(GGG to AGG) can be seen in patient I in the homozygous state.
The mutant A nucleotide is marked with an asterisk. The presence
of the mutant and normal alleles is seen in the heterozygote. The gel
was electrophoresed for 2 hours at 1. I kV; the autoradiogram was
exposed for 24 hours.
SLINGSBY ET AL
Figure 3. Digestion of Clq C-chain exon I polymerase chain reaction (PCR) product with the restriction enzyme Sfc I. The fragments
were separated on a 3% Metaphor agarose gel in 1x TBE, electrophoresed at 4OV for 90 minutes, and stained with ethidium bromide.
Lane 1 contains 200 ng of bX174 DNA digested with Hue 111. Lanes
2-5 contain PCR product digested with Sfc I: lane 2, normal
individual; lane 3, father of patient 1; lane 4, mother of patient 1;
lane 5 , patient 1. Lane 6 contains uncut PCR product from patient 1.
Presence of the mutant allele is indicated by 97 bp and 117 bp
fragments. Patient I was shown to be homozygous for this mutation
by sequence analysis, showing that the PCR product in lane 5 was
not digested to completion with the enzyme Sfc I. Complete
digestion could not be obtained by varying the digestion conditions.
Because of this, we were unable to determine whether an individual
is homozygous or heterozygous for this mutation using this PCRbased approach, but could identify the presence of this mutant allele
in either heterozygous or homozygous form.
This mutation is the same as that previously
reported in a Caucasoid Sudeten-German patient with
Clq deficiency (5). The finding of the same mutation
responsible for cases of homozygous Clq deficiency in
patients of 2 different ethnic groups (Indian and
Sudeten-German) suggests that the mutation might be
of some antiquity. This prompted us to investigate
whether this mutation might be found in other patients
with SLE and whether it could conceivably cause
increased disease susceptibility when expressed in
heterozygous form.
We screened genomic DNA samples from 158
patients with SLE by Sfc 1 digestion of the C-chain
exon 1 PCR (data not shown). Samples from the
homozygous patient and from both heterozygous parents were used as positive controls. As shown in
Figure 3, the enzyme Sfc I did not digest PCR product
from patient 1 (homozygous for this mutation) to
completion. None of the 158 samples produced the
117-bp and 97-bp products indicative of the presence
of the Sfc I restriction enzyme recognition site. It is
therefore unlikely that the mutation identified in patient l is a significant disease susceptibility gene in the
population of SLE patients as a whole.
Findings in patient 2. In this patient (described
in ref. l), a homozygous deletion of a C nucleotide at
codon 43 in the C-chain was detected. This single base
667
HEREDITARY C l q DEFlCIENCY IN SLE
Figure 4. DNA sequence C-chain polymerase chain reaction product of a normal individual,
the heterozygous brother of patient 2, and patient 2, using oligonucleotide JOK-5. The deletion
of a C nucleotide is shown in patient 1. The presence of both alleles is clearly seen in the
heterozygote. The gel was electrophoresed for 5 hours at 1.1 kV; the autoradiogram was
exposed for 24 hours.
deletion causes a shift in reading frame in the remainder of the C-chain until a premature stop codon is
reached in-frame at codon 108 (see Figure 4). Therefore, the C-chain polypeptide produced by this mutant
allele will be 109 amino acids shorter than wild-type
C-chain polypeptide and contain 65 amino acids translated with a shifted reading frame. The healthy brother
of patient 2 was shown to be heterozygous for the
mutation by sequence analysis of the C-chain PCR
product. Neither of the parents of patient 2 was
available for study.
Findings in patient 3. The molecular lesion in
this patient (described in ref. 6) was found to be a
homozygous C-to-T transition at position 41 of the C-
chain. This alters a codon for arginine to a stop codon
(see Figure 5). This will prematurely terminate the
C-chain polypeptide, such that 176 amino acids are
deleted. This mutation causes the loss of a restriction
site for the restriction enzyme Bum HI (GGATCC
wild-type to GGATCT in patient 3). The father of
patient 3 was shown to be heterozygous for this
mutation, both by sequence analysis and digestion
with Barn HI. Patient 3 was also confirmed as homozygous for this mutation by digestion of the C-chain PCR
product with Bum HI (see Figure 6).
Analysis of the genomic structure of the Clq C
chain intron. The sizes of the A-, B-, and C-chain
genes have been published previously (2,3). In all 3
Figure 5. DNA sequence of C-chain polymerase chain reaction product of a normal individual,
the heterozygous father of patient 3, and patient 3, using oligonucleotide JOK-5. The C-to-T
nucleotide change is present in patient 3, and the father is a C/T heterozygote at this position.
The gel was electrophoresed for 2.5 hours at 1.1 kV; the autoradiogram was exposed for 24
hours.
SLINGSBY ET AL
668
Figure 6. Digestion of C-chain polymerase chain reaction (PCR) product with restriction
enzyme Burn HI. ClQCl5' and ClQC23' oligonucleotides were used in the amplification of the
C-chain gene. The fragments were separated on a 0.8% agarose gel in 1 x TBE, electrophoresed
at 40V for 90 minutes, and stained with ethidium bromide. Lanes 1 and 6 contain 300 ng of A
DNA digestcd with Hind I11 and 200 ng 4x174 DNA digested with Hue Ill. The sizes of the
marker bands were 4.4 kb, 2.3 kb, I .4 kb, I . 1 kb, 870 bp, 600 bp, and 560 bp. Lane 2 is uncut
C-chain PCR product from a normal individual. Lanes 3-5 are C-chain PCR digested with Ram
HI: lane 3, normal individual; lane 4, father of patient 3; lane 5 , patient 3. The patient shows loss
of a Barn HI site, producing a 2.3-kb fragment, and the father is confirmed as a heterozygote for
this mutation. The 120-bp fragment is barely visible at the bottom of the gel.
families, the A- and B-chain PCR products were 1.8
kb, as predicted (2,3). However, the C-chain PCR
product in all 3 families was 3.8 kb, as opposed to the
predicted size of 2.5 kb. The C-chain intron was
previously reported to be 1.7 kb, with 2 Barn HI sites
flanking the intron (3). Digestion with Barn HI of the
C-chain PCR product, and digestion of the product
cloned into the pCR 11 vector gave bands of 1.7 kb and
1.4 kb and 580 bp in a normal individual, with the
products separated on a 0.8% 1 x TBE gel (Figure 6).
According to Sellar et al (3), this digestion should
produce 3 bands of 120 bp, 580 bp, and 1.7 kb. The
limits of resolution of the gel prevented us from
observing clearly the 120 bp fragment. We identified
an additional band of 1.4 kb, which corresponds to an
exon 1
205 bp
1 70 bp
DISCUSSION
The present cases bring to 32 the number of
persons described with homozygous Clq deficiency.
Of these, 30 had SLE, 1 had discoid lupus, and 1
subject is healthy, age 38 (Dr. Georges Hauptmann:
personal communication). We have recently reviewed
the clinical features of SLE accompanied by Clq
deficiency (1). The clinical features of the 2 sibling
intron
exon 2
580 bp
3.0 kb
Barn HI
I
-+-1.4
additional Bum HI site located within the C-chain
intron. This revised genomic structure is consistent
with the C-chain PCR product size being 3.8 kb, as
opposed to the 2.5 kb predicted on the basis of the
previously published C-chain gene structure (Figure 7).
Barn HI
kb +f- 1.7 kb -4-580
I
-
C1 qC23'
bp -+
Figure 7. Diagram of revised genomic structure for Clq C chain. The oligonucleotides used in
the amplification of the C-chain gene were ClQClS' and ClQC23'. The 2 exons are indicated
and the Barn H I sites shown; * denotes the Burn HI site lost in patient 3. The restriction digest
bands of 1.7 kb, 1.4 kb, and 580 bp in a normal individual become 2.3 kb and 1.4 kb with the
loss of this Barn HI site. This enables us to orient the internal Barn HI site. The polymerase
chain reaction product covering the 2 coding exons is therefore 3.8 kb.
HEREDITARY C l q DEFICIENCY IN SLE
patients reported here fall within the spectrum of SLE
associated with homozygous C lq deficiency, with
early-onset disease (ages 19 months and 4 years,
respectively, compared with a median onset age of 7
years in previously reported cases) and prominent rash
(previously reported in 25 cases).
The results of tests for autoantibodies to extractable nuclear antigens were previously reported in
17 subjects (1). Five patients were negative and 12
patients had one or more positive test result (anti-RNP
in 6; anti-Sm in 7, including 2 reported as having
high-titer RNase-resistant ENA; and anti-Ro in 5 ) . Of
the 2 new patients described here, 1 had anti-Ro
antibodies and the other, a weak and unidentified ENA
specificity. Neither had anti-dsDNA antibodies, which
have been reported in only 4 subjects with Clq deficiency.
We found that the Clq-deficient patients in each
of the 3 kindreds reported here were homozygous for
different mutations. We confirmed the identity of each
mutation by finding at least 1 other family member
(parent or sibling) who was heterozygous for the
mutation found in the patient. There have been 2
previous reports of the molecular basis of Clq deficiency in 3 kindreds (43). The first molecular lesion
identified was a homozygous G-to-A transition in the B
chain at residue 150, changing an arginine codon to a
premature stop codon (4). This point mutation also
caused the loss of a Tuq I site. The authors reported
that no material corresponding to the truncated B
chain or the full-length A and C chains could be
detected by antigenic analysis of the patient’s serum.
The second kindred showed a CAG to TAG change at
codon 186 in the A chain, which also produced a
premature stop codon (5).
The third kindred, a family of Sudeten-German
origin (ethnic origin ascertained by personal communication with Dr. Franz Petry), had a homozygous
change from a G to an A nucleotide in the first base
position of the sixth codon in the C chain’ ( 5 ) . This is
the only missense mutation identified, causing a
glycine residue to be replaced by an arginine. This
mutation is identical to that found in patient 1 of the
present report.
This change involves the introduction of a
larger amino acid, carrying a positive charge, into the
extreme 5’ portion of the C-chain polypeptide. As with
the other chains of Clq, the 5 ‘ portion of the C-chain
consists of a repeating structural motif of GlycineX-Y. This motif enables the Clq molecule to form a
collagen-like triple-helical stalk, topped with a globu-
669
lar domain. This amino acid substitution will disrupt
the second of 27 Glycine-X-Y motifs due to the larger
size of the arginine compared to glycine and by the
introduction of a positive charge. It is surprising that
such a mutation in the extreme 5’ end of the polypeptide would be sufficient to cause the total loss of Clq
function in the homozygous state.
In both patient 1 described in this report and the
Sudeten-German patient described elsewhere, there is
a dysfunctional Clq molecule present in serum. The
Clq protein from the German patient has been previously analyzed (12). The defective Clq molecule was
hemolytically inactive, did not bind immune complexes, and was not recognized by the monocyte Clq
receptor. The Clq molecule was present in the serum
in a 150-kd form, corresponding to an (A-B),C2 structural subunit. Therefore, this mutation would appear
to inhibit the association of 3 such structural subunits
into a single biologically active Clq molecule.
It is surprising to note that Clq-deficient patients from 2 racially distinct populations (Punjabi
from India and Sudeten-German) carry an identical
molecular lesion, suggesting that the mutation may be
of considerable antiquity. We therefore examined the
possibility that this mutation might be overrepresented
in lupus patients in whom no complement deficiency
had been identified. However, we found no S’c I sites
(corresponding to the mutated sequence) in exon 1 of
158 genomic DNA samples derived from patients with
SLE. The alternative explanation is that this mutation
occurred independently on 2 occasions. These 2 possibilities (1 or 2 founder mutations) cannot be distinguished at present.
The mutation in patient 2 is the first description
of a single-base deletion in a Clq-deficient patient.
This causes a predicted shift in the reading frame in the
C-chain messenger RNA (mRNA), with 42 correct
amino acids followed by 65 incorrect amino acids.
Normal levels of Clq (71 mg/ml) were identified in the
heterozygous brother, which demonstrates that this
mutation does not have a dominant-negative effect on
Clq expression. Such dominant mutations in collagenlike triple-helical regions have been described in
mannose-binding protein (13,14). There are a number
of possible reasons why we did not observe a dominant effect in patient 2. The molecular lesion may have
caused the altered mRNA to be unstable with respect
to the wild-type mRNA (15), such that little or no
mutant polypeptide was produced, or the molecular
lesion could have prevented secretion of the mutant
polypeptide species. Both these mechanisms would
SLINGSBY ET AL
670
effectively produce a null allele. Alternatively, the
mutant C-chain might have been unable to dimerize or
if a C2 dimer was formed, it might not have been able
to form a stable (A-B),C2 subunit. This patient is
deceased and no material is available from other family
members, so these possibilities cannot be explored.
The mutation in patient 3 resulted in a premature stop codon and caused the deletion of 177 amino
acids from the mature C-chain polypeptide, including
the whole of the carboxy-terminal globular domain.
The predicted polypeptide produced would have contained 12 of 27 Gly-X-Y motifs. Once again, this
abrogated polypeptide chain did not have a dominantnegative effect and did not appear to disrupt the
assembly of biologically active Clq molecules in heterozygous family members.
This study increases the number of reports of
the molecular basis of Clq deficiency from 3 to 6
kindreds. Mutations have now been identified in all 3
chains of Clq. The mutations identified are all single
basepair mutations with 3 premature stop codons, 2
kindreds of different ethnic origin with the same
glycine-to-arginine amino acid change, and 1 single
basepair deletion. As yet, no dominant or spontaneous
Clq mutations have been described.
ACKNOWLEDGMENTS
We are grateful for the gift of oligos K3 and K4 from
Prof. K. B. M. Reid. We are grateful to Prof. Ron Thompson
for drawing our attention to the family described in this
paper and for helpful discussions.
REFERENCES
1. Bowness P, Davies KA, Norsworthy PJ, Athanassiou P,
Taylor-Wiedman J, Borysiewicz LK, Meyer PAR, Walport MJ:
Hereditary Clq deficiency and SLE: case report and review of
the literature. Q J Med 87:455464, 1994
2. Reid KBM: Molecular cloning and characterization of the
complementary DNA and gene coding for the B-chain of subcomponent Clq of the human complement system. Biochem J
231:729-735, 1985
3. Sellar GC, Blake DJ, Reid KBM: Characterization and organization of the genes encoding the A-, B-, and C-chains of human
complement subcomponent Clq. Biochem J 274:481490, 1991
4. McAdam RA, Goundis D, Reid KBM: A homozygous point
mutation results in a stop codon in the C l q B-chain of a
Clq-deficient individual. Immunogenetics 27:259-264, 1988
5 . Petry F, Le DT, Loos M: Sequence analysis of two cases of C l q
deficiency gives an insight into the structural requirements of
the assembly of the functionally active Clq molecule (abstract).
Clin Exp Immunol 97 (suppl 2):39, 1994
6. Mikuska AM, Stefanovic Z, Miletic V, Oxelius V-A, Sjoholm
AG: Systemic lupus erythematosus-like disease in two siblings
with complete C l q deficiency. Periodicum Hiologorum 85:271273, 1983
7. Samuel DJ, Amlot PL, Shepherd P, Lachmann PJ: An efficient
one-step method for isolating immune complexes from whole
serum using a monoclonal anti-C3g affinity immunosorbent. Clin
Exp Immunol65:458-464, 1986
8. Harrison RA, Lachmann PJ: Complement technology. In,
Handbook of Experimental Immunology. Fourth edition. Edited by DM Weir, Oxford, Blackwell Scientific Publications,
1986
9. Siegert CE, Dahd MR, Halma C, van der Voort EA, Breedveld
FC: IgG and IgA autoantibodies to Clq in systemic and renal
diseases. Clin Exp Rheumatol 10:19-23, 1992
10. Sambrook J, Fritsch EF, Maniatis T: Molecular Cloning: A
Laboratory Manual. Second edition. Cold Spring Harbor, NY,
Cold Spring Harbor Laboratory Press, 1989
11. Tan EM, Cohen AS, Fries JF, Masi AT, McShane DJ, Rothfield
NF, Schaller JG, Tala1 N, Winchester RJ: The 1982 revised
criteria for the classification of systemic lupus erythematosus.
Arthritis Rheum 25: 1271-1277, 1982
12. Kirschfink M, Petry F, Khirwadkar K, Wigand R, Kaltwassser
JP, Loos M: Complete functional Clq deficiency associated
with systemic lupus erythematosus (SLE). Clin Exp Immunol
94~267-272, 1993
13. Sumiya M, Super M, Tabona P, Levinsky RJ, Arai T, Turner
MW, Summefield JA: Molecular basis of opsonic defect in
immunodeficient children. Lancet 337: 1569-1570, 1991
14. Summefield JA, Ryder S, Sumiya M, Thursz M, Gorchein A,
Monteil MA, Turner MW: Mannose binding protein gene mutations associated with unusual and severe infections in adults.
Lancet 345:886489, 1995
15. Sachs A: Messenger RNA degradation in eukaryotes. Cell
74:413421, 1993
Документ
Категория
Без категории
Просмотров
2
Размер файла
739 Кб
Теги
molecular, family, systemic, erythematosus, hereditary, three, new, familie, lupus, basic, deficiency, homozygous, c1q
1/--страниц
Пожаловаться на содержимое документа