| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
*
Department of Biochemistry, Fukushima Medical University School of Medicine, 1-Hikarigaoka, Fukushima, Japan; and
Department of Pediatrics, Chiba University School of Medicine, Inohana, Chyuo-ku, Chiba, Japan
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Several cases of human C1 deficiency have been reported 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18 in which the deficiency was due to the absence of the subcomponents C1q 11, 14, 15, 16, 17, 18 , C1r 8, 9, 13 , C1s 12 , or both C1r and C1s 10 . The molecular basis for hereditary C1q deficiencies are well defined in which homozygosity for nonsense mutations in both C1q A and B chain genes 11, 15, 17, 18 and deletion, nonsense, and missense mutations in the C1q C chain gene 15, 16 were demonstrated. However, there has been no report on the molecular basis for C1r and/or C1s deficiency, probably because of their low incidence 19, 20 and the lack of information needed for the analysis of patients’ genomes, especially with respect to the exon-intron structures of the C1r and C1s genes. In addition, it is difficult to obtain tissue specimens from the liver for mRNA analysis, the liver being a primary organ that expresses both genes 2, 6 .
Recently, we reported the full structure of exon organization of the human C1s gene 21 ; that study allowed us to assess the exon sequences throughout the coding region of this gene using genomic DNA. In this paper, we describe a case of selective C1s deficiency resulting from a compound heterozygosity for two different abnormalities in the C1s gene.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The pedigree of a Japanese family including one patient with the
C1s deficiency is shown in Fig. 1
A. The patient’s medical
history included a virus-associated hemophagocytic syndrome
(VAHS)3 (see Refs. 22 and 23)
at age 4. In 1996 at age 6, he was admitted to Chiba University Medical
Hospital (Chiba, Japan) because of a fever of unknown origin. A month
after the admission, he had a convulsive fit and lost consciousness.
Without recovering consciousness for 6 mo, he died at age 7 in 1997.
The family had no special medical history. CH50 in both serum and
plasma from the patient was undetectable despite normal levels of the
complement components C1q, C2–C9. By Ouchterlony immunodiffusion, no
C1s was detected in serum, whereas C1r was present. In seven live
members of the family, both C1r and C1s were detectable.
|
Serum C1s levels were estimated by rocket immunoelectrophoresis with goat anti-human C1s Ab (Calbiochem, La Jolla, CA) as described 24 .
Southern blot hybridization
Southern blot hybridization with genomic DNA, digested with several restriction enzymes such as EcoRI, BamHI, and HindIII, was performed as described 25 using full-length C1s cDNA as a probe. Genomic DNA was prepared from peripheral white blood cells as described 26 .
PCR amplification of each exon of the C1s gene
PCR was conducted to amplify each exon of the human C1s gene
using genomic DNA as a template and flanking intron sequences as
primers. The nucleotide sequences of the exon-specific primers used are
shown in Table I. The first exon, which
encodes the 5'-untranslated region of the C1s transcript, was excluded
from the present analysis. The longest exon, exon XII, which encodes
the entire protease domain of C1s, was amplified in three overlapping
sequences to achieve suitable sizes of the PCR products for
single-strand conformation polymorphism (SSCP) analysis. PCR was
performed in a 50 µl volume containing 1x buffer (cDNA reaction
buffer; Clontech, Palo Alto, CA), 2.5 mM of each dNTPs, 10 µM forward
and reverse primers, 200 ng genomic DNA, and 0.5 µl Advantage cDNA
polymerase mix (Clontech). The thermocycling protocol was as follows:
the first step at 94°C for 2 min; the second step consisting of 5
cycles at 94°C for 10 s, 65°C for 30 s, and 68°C for 2
min; the third step consisting of 5 cycles at 94°C for 10 s,
60°C for 30 s, and 68°C for 2 min; the fourth step consisting
of 20 cycles at 94°C for 10 s, 52–55°C for 30 s, and
68°C for 2 min; and the fifth step at 68°C for 7 min.
|
Nonradioisotopic SSCP analysis was performed as described 27 . PCR products of 1–2 µl were mixed with a 2-fold volume excess of formamide containing 2.5% bromophenol blue and 2.5% xylene cyanol, denatured at 80°C for 5 min, and then subjected to electrophoresis on 6–10% polyacrylamide gel in 25 mM Tris-glycine buffer (pH 8.3) under cooling condition. After electrophoresis, the gel was stained using a silver staining kit (Silver Stain Plus; Bio-Rad, Richmond, CA). In this conventional SSCP, a dsDNA was observed in addition to two ssDNAs that were generated by partial renaturation of the PCR product.
Cloning of PCR products
PCR products were ligated into a plasmid vector (pGEM-T easy; Promega, Madison, WI) using a ligation kit (Takara Shuzo, Kyoto, Japan), and then transfected into XL1-Blue (Stratagene, La Jolla, CA). After plating, at least 10 colonies were selected at random for nucleotide sequence analysis.
Nucleotide sequence analysis
DNA sequence was determined by the dideoxy chain termination method 28 using a Li-Cor DNA sequencer (model 4000; Li-Cor, Lincoln, NE). The labeling reaction was conducted using the SequiTherm Long-Read cycle sequencing kit (Epicentre Technologies, Madison, WI). Sequencing primers were synthesized by Nishinbo (Tokyo, Japan).
Western blot analysis
Western blot analysis was performed to detect the truncated forms of C1s in the patient’s serum. Briefly, up to 2 µl of serum was subjected to a 9% polyacrylamide gel electrophoresis under nonreducing condition. After blotting to a membrane filter, the filter was developed with rabbit anti-C1s Ab (Serotec, Oxford, U.K.) or goat anti-C1s Ab (Calbiochem) followed by peroxidase-conjugated anti-rabbit IgG (or anti-goat IgG) Ab (Dako, Glostrup, Denmark).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
A semiquantitative estimation of C1s by rocket
immunoelectrophoresis showed no detectable C1s in serum from the
patient (Fig. 1
B). The parents and one of two siblings
(corresponding to II-1, II-2, and III-1 in Fig. 1A) had half
of the level of C1s that was detectable in an unrelated healthy
subject, whereas the level in another sibling was as high as
that in the healthy subject. The levels in sera from the patient’s
grandfather and grandmothers were all lower than that in the healthy
subject. Among the three grandparents, the level in the paternal
grandmother was much lower, i.e., one-half the level of those in the
paternal grandfather and the maternal grandmother.
Southern blot hybridization analysis of the C1s gene
No differences in the Southern blot hybridization patterns were observed among the patient, the members of the family, and the unrelated healthy subject, indicating that there was not a gross deletion or rearrangement of the C1s gene in the patient (data not shown).
PCR-SSCP analysis of each exon of the C1s gene
The PCR products obtained by starting with genomic DNA were
subjected to SSCP analysis to screen for abnormality in the sequence.
When using the PCR products corresponding exon X, an abnormal pattern
was observed in the patient, his father, and his paternal grandmother
(Fig. 2A). In addition to
normal bands with the same mobilities as those in the healthy subject,
abnormal bands were observed in these three individuals, demonstrating
that they are all heterozygous for the abnormality. This result also
indicates that the mutation was located on the paternal allele in the
patient. PCR-SSCP patterns from the remaining five members, including
the patient’s mother, two siblings, paternal grandfather, and maternal
grandmother were all normal. No abnormality in PCR-SSCP for the other
exons was observed in any of the eight family members tested (data not
shown).
|
The PCR product corresponding exon X from the patient was cloned
in a plasmid. By sequencing the clones, we detected a deletion of 4 bp,
TTTG (Fig. 2B), which corresponds to cDNA sequences
1350–1353 2 or 1293–1296 3 . One-half species of the clones
selected at random have the same deletion, whereas the others have
normal sequences, indicating again that the patient was heterozygous
for the abnormal allele. In a PCR-SSCP starting with the cloned plasmid
as template DNA, we confirmed that the cloned abnormal sequence had the
same abnormal pattern as the uncloned genomic DNA (data not shown).
Because of a newly generated in-frame stop codon at about 90 bp
downstream from the 4-bp deletion, the abnormal gene encodes a
truncated form of C1s protein from the N terminus to the short
consensus repeat domain.
Abnormality in the nucleotide sequence of exon XII
To detect another abnormality in the C1s gene, all of the
SSCP-negative PCR products from the patient were cloned in a plasmid
and then sequenced. We identified a nonsense mutation at codon 608 in
exon XII that was caused by a nucleotide substitution from G
(GAA for glutamic acid) to T (TAA for stop
codon) (Fig. 3A). The abnormal
gene encodes a truncated form of C1s that lacks the C-terminal 80 amino
acids. By extensive sequencing of the PCR product clones from members
of the family, it was found that in addition to the patient, his mother
and one of his siblings had the same mutation, and that all three were
heterozygous for it (Fig. 3B). This result also indicates
that the nonsense mutation was located on the maternal allele in the
patient.
|
The truncated forms of C1s that were possibly generated from the two mutant alleles of the C1s gene in the patient were undetectable by Western blot analysis (data not shown). In a control experiment with serum from a healthy subject, the intact C1s was obtained in a band of about 8 kDa.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
A. The serum C1s levels were clearly dependent on the genotypes of the C1s gene at least in the second and third generations in the family in which the compound heterozygote had no detectable C1s and the single heterozygotes had roughly half of the normal level. The serum C1s levels in two grandparents, paternal grandfather and maternal grandmother, were lower than that in the healthy subject, despite their normal genotypes of the C1s gene, and the level in the paternal grandmother, who was heterozygote for the 4-bp deletion, was half of the level that was detected in the two former grandparents. Although there is no precise report concerning the relationship between serum C1s level and age, it is possible that the normal level of serum C1s in the aged generation may be lower than the normal level in the entire population.
We failed to identify the truncated C1s proteins in the patient’s serum by a Western blot analysis, suggesting that it was unstable in the circulation and/or was not antigenic. The putative truncated proteins lack the entire or part of the protease domain that includes the active-site Ser631 2, 3 , suggesting that they are functionally inactive. This finding implies that in the patient the classical pathway of complement activation did not function because of a failure in the formation of active C1 complex.
It has been reported that defects in the early components of
complement, C1–C4, are frequently associated with a systemic lupus
erythematosus- (SLE-) like syndrome or with recurrent viral and
bacterial infections or with both 14, 19, 20 . We observed no clear
SLE-like symptoms in our patient, except for weak titers of
autoantibodies to nuclear and cytoplasm Ags. Also he had no apparent
susceptibility to infectious disease. The patient’s medical history
included VAHS. His clinical features at the time of admission such as
persisting fever, swelling of lymphnodes, and elevated levels of serum
ferritin and glutamic-oxaloacetic transaminase/glutamic-pyrovic
transaminase 22, 23, 29 might indicate a second episode of
VAHS, although we failed to identify any pathogens in culture of his
blood and sputum. Risdall et al. 22 classified VAHS into two groups:
group I develops VAHS in the absence of any apparent underlying disease
and group II develops VAHS after the initiation of immunosuppression.
They also reported that one of the patients in the latter group had
SLE. VAHS is thought to be induced by the activation of macrophages or
T cells through an incomplete immune response against pathogens
followed by the overproduction of cytokines such as IFN-
, IL-1,
IL-2, IL-6, TNF, and soluble IL-2R 30, 31 . Taken together, it is
likely that in our patient dysfunction of the classical pathway due to
the C1s deficiency, i.e., a lack of host defense against pathogens,
finally resulted in VAHS. The present case seems to belong to group II
of the above classifications.
The clinical features of the present patient were quite different from those of another Japanese patient with C1s deficiency 12 who had a mild SLE-like syndrome. The differences in clinical features and severity among the cases with the C1 deficiencies have been reported 14, 19 . This might be explained by the differences in the genetic backgrounds among the patients. Based on experiments using C1q-deficient mice generated by gene targeting, Botto et al. 32 reported that the phenotype associated with C1q deficiency was modified by background gene(s).
The fact that members of the family heterozygous for either of the two mutations had no symptoms suggests that the disease is inherited in an autosomal recessive mode. Although the frequencies of the two abnormal alleles in the Japanese population are unknown, the fact that the origin of the 4-bp deletion in exon X was traced back at least to the first generation of the family suggests that the mutation is spread in the local area of Japan. Autosomal recessive inheritance has also been reported in hereditary C1q deficiency 9, 16 .
In conclusion, a Japanese patient with selective C1s deficiency was a compound heterozygote for two abnormalities in the C1s gene. The levels of serum C1s in the family were correlated with the genotypes of the C1s gene. The absence of functionally active C1s in the patient may have caused clinical features such as VAHS.
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Yuichi Endo, Dept. of Biochemistry, Fukushima Medical University School of Medicine, 1-Hikarigaoka, Fukushima 960-1295, Japan.
3 Abbreviations used in this paper: VAHS, virus-associated hemophagocytic syndrome; SSCP, single-strand conformation polymorphism; SLE, systemic lupus erythematosus.
Received for publication August 31, 1998. Accepted for publication November 9, 1998.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  M.-A. Dragon-Durey, P. Quartier, V. Fremeaux-Bacchi, J. Blouin, C. de Barace, A.-M. Prieur, L. Weiss, and W.-H. Fridman Molecular Basis of a Selective C1s Deficiency Associated with Early Onset Multiple Autoimmune Diseases J. Immunol., June 15, 2001; 166(12): 7612 - 7616. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
