Identification of a C→ T Mutation in the Reactive-Site CodingRegion of the C1-Inhibitor Gene and its Detection by an ImprovedMutation-Specific Polymerase Chain Reaction Method

E. W. NIELSEN*†, H. FURE†, P. WINGE‡ & T. E. MOLLNES†

Departments of*Anesthesiology and†Immunology and Transfusion Medicine, Nordland Central Hospital, Bodø and University of Tromsø; and‡UNIGEN, University of Trondheim, Norway

(Received 15 October 1997; Accepted in revised form 18 November 1997)

Nielsen EW, Fure H, Winge P, Mollnes TE. Identification of a C→ T Mutation in the Reactive-Site CodingRegion of the C1-Inhibitor Gene and its Detection by an Improved Mutation-Specific Polymerase ChainReaction Method. Scand J Immunol 1998;47:273–276

Mutations in the C1-inhibitor (C1-INH) gene, leading to low functional levels of C1-inhibitor protein, causehereditary angioedema (HAE). The disease is characterized by episodic edema in a number of organs.Typically, swellings occur in extremities and face, often accompanied by crampy abdominal pain. Laryngealedema may lead to suffocation. Type II HAE patients have low functional C1-INH values stemming fromonly one normal allele. Antigenic C1-INH values, however, are normal or increased owing to the presenceof a dysfunctional protein from the mutated allele. The mutations are usually found in exon 8 coding forthe amino acids near the reactive centre (P1). Previously, no mutations in the C1-INH gene had beenpublished from the Scandinavian countries. In this work, exon 8 of the C1-inhibitor gene was sequenced inmembers of two different kindreds, from western and northern Norway, who were suffering from HAE type II.A common point mutation was found within the bait region encoding the reactive centre. The codon CGC wasconverted to TGC at position 17 970, corresponding to an Arg→ Cys replacement which reportedly is thesecond most frequent type II HAE mutation. This information was utilized to develop a mutation-specificpolymerase chain reaction (PCR) for the identification of affected family members. The antisense 17-merprimer (50-AAGACCAGCAGGGTGCA-30) was successfully applied and AmpliTaq Gold was used inthe PCR.

Erik Waage Nielsen, Department of Anaesthesiology, Nordland Central Hospital, 8017 Bodø, Norway

INTRODUCTION

Hereditary angioedema (HAE) manifests as acute intermittentswellings of nearly every organ of the body. Edema of the skinand the gastrointestinal tract is encountered most frequently,although a wide variation in severity and frequency is typical,between patients and also within the affected individual. Laryn-geal edema may cause suffocation. The disease is caused by amutation in one of the two alleles coding for the C1-inhibitor(C1-INH), an important inhibitor of the complement andkallikrein-kinin systems. As a consequence of low levels offunctional C1-INH, vasoactive peptides are liberated. In type IHAE the mutated allele either produces no protein or onewhich is not detected antigenically [1, 2]. Both functional andantigenic C1-INH assays show low values in type I HAE, as onlyC1-INH from the intact allele is measured. In patients with type

II HAE the mutated allele produces a dysfunctional C1-INHprotein with preserved antigenic properties. The antigenicC1-INH value may be high, as this test detects both the wild-type and the mutated protein. The functional test still measuresonly the wild-type protein and shows low values. An earlydiagnosis of HAE is important as symptoms from the larynxand gastrointestinal tract can evolve in infancy [3, 4]. Eitheralone or together with complement analyses [5], the detectionof the mutation in a child from a kindred suffering from HAEis of pivotal clinical importance. We have therefore investi-gated a type II HAE mutation shared by two kindreds fromwestern and northern Norway, and tailored a direct mutation-specific polymerase chain reaction (PCR) method for the easydetection of all affected individuals with this genetic defect.This is the first HAE mutation of any kind to be publishedfrom Scandinavia.

Scand. J. Immunol.47, 273–276, 1998

q 1998 Blackwell Science Ltd

MATERIALS AND METHODS

Samples and preparation of genomic DNA. Four patients with type IIHAE from one family in northern Norway and one patient with type IIHAE from a family in western Norway were studied. Four healthy blooddonors were used as negative controls. Human DNA was prepared fromanticoagulated whole blood by denaturation and precipitation withtrimethyl-ammonium bromide salts according to the method describedby Gustincichet al. [6].

Amplification of exon 8 of the C1-INH gene. Primers flanking the exon8 of the C1-inhibitor gene were chosen by using the computer programOLIGO 4.1 [7]. The primer sequences were: 50-ATGCTGGCTTCT-GACTCTGT-30 (HAE Primer1) and the antisense primer 50-ACTGA-GAGCTGAGGCTGGAG-30 (HAE Primer2). The primers used weresynthesized by Genosys Biotechnologies Inc. (Cambridge, UK). Exon 8of the C1-INH gene was amplified by PCR, using a thermal cycler (GeneAmp PCR System 2400; Perkin Elmer, Norwalk, CT), giving a productlength of 349 bp. PCR was carried out in a 50ml reaction mixture usingapproximately 0.5mg genomic DNA, 50 pmol each primer, 200mM

dNTPs (Pharmacia, Uppsala, Sweden), 50 mM KCl, 1.5 mM MgCl2,10 mM Tris-HCl (pH 8.3), 0.001% (w/v) gelatin and 1.25 U AmpliTaqGold (Perkin Elmer). After the first denaturation step at 948C for12 min, samples were subjected to 25 cycles of PCR (948C for 30 s,608C for 30 s and 728C for 30 s). The amplified samples were electro-phoresed on a 2% low-melting agarose gel at 200 V for 20 min, andstained with 0.5mg/ml ethidium bromide. Amplified fragments werepurified after electrophoresis in low-melting agarose gels [8].

DNA sequencing. Amplified DNA fragments were sequenced by amodification of the dideoxynucleotide chain-terminating method [9].The sequencing reaction was performed with one or the other primerused for genomic DNA amplification. The radioactive label was incor-porated into the reaction products by using four [a-33P] dideoxynucleo-tide (ddNTP) terminators (G, A, T and C) and Thermo Sequenase (bothreagents from Amersham Life Science, Cleveland, OH) [10]. The cyclesequencing conditions were similar to the amplification of exon 8, exceptfor the initial denaturation step which was reduced to 1 min. Themixtures were denatured and resolved on an 8M Urea/6% acrylamidegel in a buffer containing 0.09M Tris, 0.03M Taurine and 0.5 mM EDTA.The gel was fixed, dried and exposed overnight to X-ray film (HyperfilmMP; Amersham Life Science).

Mutation-specific PCR. Mutation-specific PCR was performed usinga combination of HAE Primer1 and the antisense 17-mer oligonucleotide50-AAGACCAGCAGGGTGCA-30, with the mutation specific A in the30-position (here displayed in bold), corresponding to the region betweenpositions 17 970 and 17 986. The mutation-specific PCR was performedon 2ml of the HAE Primer1–Primer2 product. The reaction conditionswere similar to the initial PCR amplification described above, but thedNTP concentration was reduced to 20mM. After denaturation at 948Cfor 12 min, 15 cycles were performed as follows: 30 s at 948C, 30 s at488C and 30 s at 728C, with a final step at 728C for an additional 7 min.Ten microlitres of the above reaction mixture were electrophoresed on a2% agarose gel. Bands were identified following ethidium bromidestaining. The identification of a 203-bp band was regarded as a C→ Tmutation at position 17 970 of the C1-inhibitor gene. Molecular markerswere supplied by Sigma (St. Louis, MO).

RESULTS

Sequence flanking the C→ T mutation is shown for the sensestrands in Fig. 1, as well as the corresponding sequence from a

control. Of the four patients from the family of northern Norway,all were heterozygous for the CGC to TGC mutations, as shownby rectangles in Fig. 1. The same mutation was also found in thepatient from western Norway, and antisense strands in bothfamilies confirmed this mutation. The detection by mutation-specific PCR of the same mutation in both families and the use ofAmpliTaq Gold is shown in Fig. 2.

DISCUSSION

Most type II HAE mutations replace the Arg444 at the reactive

274 E. W. Nielsen et al.

q 1998 Blackwell Science Ltd,Scandinavian Journal of Immunology, 47, 273–276

Fig. 1. DNA sequence from exon 8 of the C1-inhibitor gene. Apoint mutation at position 17 970 of the C1-inhibitor gene is present(rectangle), which is within the bait region encoding the reactivecentre. Codon CGC is converted to TGC, corresponding to an Argto Cys replacement.

Fig. 2. Mutation-specific PCR. The presence of a 203-bp banddemonstrates a C→ T point mutation at position 17 970 in theC1-inhibitor gene. The presence of a 349-bp band is the result of apre-PCR amplifying the exon 8 of the C1-inhibitor gene. Lanes 1–4,four members from a family in northern Norway; lane 5, one memberof a kindred in western Norway; lanes 6–10, four healthy blooddonors. Lane 10 displays base-pair markers.

centre or within the hinge region, some 10 amino acids N-terminal to the reactive site [11], although substitutions as faras 193 amino acids from the N-terminus have been describedrecently [12]. Some amino acid replacements near the reactivecentre may even split the substrate specificity of C1-INH. Theinhibitory capacity against the different proteases may thereforevary according to the type of mutation [13]. Thus, an Ala443 toVal substitution in a kindred without angioedema led to activa-tion of the classical complement pathway with low C4 valuesowing to a defect of the mutated C1-INH in complexing with C1rand C1s, while the capacity for binding to kallikrein and FactorXIIa was maintained [14]. An ordinary test battery of functionaland antigenic C1-INH and C4 would falsely diagnose thosepersons as having type II HAE. This is so because functionalC1-INH tests are often based on the ability of C1-INH in thesample to inhibit exogenously added C1s from splitting achromogenic substrate. A DNA-based detection is thereforevaluable.

Both antigenic and functional C1-INH reference values changeconsiderably from the newborn to the toddler, making theinterpretation of test values more difficult [5]. Once the mutationof the kindred is known, the mutation-specific PCR method canalso be a valuable diagnostic tool in early age.

CGC is a codon for arginine. The cytosine in CG dinucleotidesis a hot-spot for DNA methylation, often followed by spontaneous5-methylcytosine deamination. This yields thymine instead, andthe new triplet TGC codes for cysteine. If the methylation occursin the antisense triplet first, converting GCG to GTG, thecorresponding sense triplet will subsequently be transformedinto CAC, a codon for histidine. Other mechanisms may alsobe operative, as CG also appears to be hypermutable, irrespectiveof methylation-mediated deamination [15].

The Arg→ Cys exchange, however, accounts for the secondmost frequent HAE type II mutation. In a total of 37 HAE typeII kindreds this amino acid switch was found in 11 differentfamilies [11], which broadens the applicability of the mutation-specific test we have presented. Furthermore, the cysteinereplacement allows the C1-INH molecule to bind covalentlyto albumin, a property utilized as a diagnostic tool by some[16].

The selective amplification of 30-terminal C and T mismatcheshas in some cases shown to be difficult, as both the mutatedstrand and the wild type have been amplified equally effectively[17]. By contrast, other reports have shown that by using suitablePCR primers and good reaction conditions most mismatchedbase pairs at template–primer 30-termini are refractive to ampli-fication byTaqpolymerase [18]. One of the reasons that primerswith 30 mismatches can be extended by DNA polymerase is thatrare tautomeric forms of the purines and pyrimidines can producethe stable bonds between the mispairs. Particularly stable bondsare formed between the mispairs C:A and T:G [19]. The mutation-specific primer we have applied creates a potential C:A mispairwith the normal allele. The amplification of specific alleles seemsto depend on a number of variables: the type of mismatchbetween PCR primer and template, the annealing temperature,

the dNTP concentration and also the type of DNA polymeraseused.

We present here a method that allows for selective amplifica-tion of a mutated HAE allele. Our selective amplification failedwhen using the ordinary AmpliTaq. The exchange of the con-ventional DNA polymerase for the new AmpliTaq Gold, how-ever, reproducibly amplified the two different strands. Althoughthis paper does not provide an explanation for this phenomenon,a participating factor might be that while using AmpliTaq Gold,heat activation of the enzyme might not be 100% in the initialcycles. The reduced enzyme activity in the first cycles maytherefore increase the specificity of the amplification.

To our knowledge, this is the first published mutation-specificPCR for an HAE mutation.

ACKNOWLEDGMENT

This work was supported by grants from the Odd Berg Founda-tion and the Norwegian Council on Cardiovascular Disease.

REFERENCES

1 Verpy E, Couture Tosi E, Tosi M. C1 inhibitor mutations whichaffect intracellular transport and secretion in type I hereditaryangioedema. Behring Inst Mitt 1993;93:120–4.

2 Verpy E, Couture Tosi E, Eldering Eet al. Crucial residues in thecarboxy-terminal end of C1 inhibitor revealed by pathogenicmutants impaired in secretion or function. J Clin Invest 1995;95:350–9.

3 Watson IM. Angioneurotic edema. Report of a case with necropsyfindings. J Am Med Assoc 1926;86:1332–3.

4 Donaldson VH, Rosen FS. Hereditary angioneurotic edema: Aclinical survey. Pediatrics 1966;37:1017–27.

5 Nielsen EW, Johansen HT, Holt J, Mollnes TE. C1 inhibitor anddiagnosis of hereditary angioedema in newborns. Pediatr Res 1994;35:184–7.

6 Gustincich S, Manfioletti G, Del Sal G, Schneider C, Carninci P. Afast method for high-quality genomic DNA extraction from wholehuman blood. BioTechniques 1991;11:298–300.

7 Rychlik W, Rhoads RE. A computer program for choosing optimaloligonucleotides for filter hybridization, sequencing and in vitroamplification of DNA. Nucl Acids Res 1989;17:8543–51.

8 Sambrook JF, Fritsch EF, Maniatis T. Molecular Cloning. ALaboratory Manual. 2nd edn. New York: Cold Spring HarborLaboratory Press, 1989.

9 Sanger F, Nicklen S, Coulson AR. DNA sequencing with chain-terminating inhibitors. Proc Natl Acad Sci USA 1977;74:5463–7.

10 Tabor S, Richardson CC. A single residue in DNA polymerases ofthe Escherichia coli DNA polymerase I family is critical fordistinguishing between deoxy- and dideoxyribonucleotides. ProcNatl Acad Sci USA 1995;92:6339–43.

11 Davis AE, Bissler JJ, Cicardi M. Mutations in the C1 inhibitor genethat result in hereditary angioneurotic edema. Behring Inst Mitt1993;93:313–20.

12 Zahedi R, Aulak KS, Eldering E, Davis AE. Characterization of C1inhibitor-Ta – a dysfunctional C1INH with deletion of lysine 251. JBiol Chem 1996;271:24307–12.

13 Davis AE, Aulak KS, Parad RBet al. C1 inhibitor hinge region

A C1-INH Mutation Detected by Specific PCR275

q 1998 Blackwell Science Ltd,Scandinavian Journal of Immunology, 47, 273–276

mutations produce dysfunction by different mechanisms. NatureGenet 1992;1:354–8.

14 Zahedi R, Bissler JJ, Davis AE, Andreadis C, Wisnieski JJ. UniqueC1 inhibitor dysfunction in a kindred without angioedema. II.Identification of an Ala(443)→Val substitution and functionalanalysis of the recombinant mutant protein. J Clin Invest 1995;95:1299–305.

15 Cooper DN, Krawczak M. The mutational spectrum of single base-pair substitutions causing human genetic disease: patterns andpredictions. Human Genet 1990;85:55–74.

16 Starzia Z, Hegyi E, Kuklinek P, Cap J, Stefanovic J, Lokaj J.Hereditary angioedema in Czechoslovakia. Clinical, immunological,

genetic and therapeutic studies in 16 families. Allerg Immunol1988;34:35–42.

17 Kwok S, Kellogg DE, McKinney Net al. Effects of primer–templatemismatches on the polymerase chain reaction: human immunodefi-ciency virus type 1 model studies. Nucl Acids Res 1990;18:999–1005.

18 Huang MM, Arnheim N, Goodman MF. Extension of base mispairsby Taq DNA polymerase: implications for single nucleotide dis-crimination in PCR. Nucl Acids Res 1992;20:4567–73.

19 Mendelman LV, Petruska J, Goodman MF. Base mispair extensionkinetics. Comparison of DNA polymerase alpha and reverse tran-scriptase. J Biol Chem 1990;265:2338–46.

276 E. W. Nielsen et al.

q 1998 Blackwell Science Ltd,Scandinavian Journal of Immunology, 47, 273–276