| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Technical Briefs |
1 Service de Biochimie et de Génétique Moléculaire, Hôpital Henri Mondor, AP-HP, 94010 Créteil, France;
2
INSERM U468 Génétique Moléculaire et Physiopathologie, 94010 Créteil, France
aaddress correspondence to this author at: Service de Biochimie et de Génétique Moléculaire–INSERM U468, Hôpital Henri Mondor, 51, Avenue du Maréchal de Lattre de Tassigny, 94010 Créteil, France; fax 33-1-4981-2842, e-mail emmanuelle.girodon{at}im3.inserm.fr
Familial Mediterranean fever (FMF) is an autosomal-recessive disorder (MIM 249100) characterized by recurrent attacks of fever and serositis, affecting principally Sephardic Jewish, Armenian, Arab, and Turkish populations. Early diagnosis is important to initiate colchicine therapy, which prevents the occurrence of attacks and of renal amyloidosis, the major complication of the disease. The identification of MEFV (1)(2), the gene responsible for the disease, allowed the use of an early molecular test of diagnostic value for FMF patients (3), which negates the needs for unnecessary invasive investigations. The MEFV gene, located on chromosome 16p13.3, contains 10 exons and encodes the marenostrin/pyrin protein, a molecule acting as a regulator of the proinflammatory interleukin-1-dependent pathway and of the apoptosis mediated by the apoptosis-associated Speck-like protein containing a CARD (ASC) (4)(5), and which belongs to the death domain-fold family (6).
More than 40 different FMF-associated mutations have been described to date (7), the most frequent ones being located in exon 10. Indeed, the M694V, V726A, M680I, and M694I mutations account for 65–95% of FMF alleles depending on the ethnic origin of the patient (8). E148Q is a frequent sequence variation situated in exon 2, but its involvement in the development of the disease remains controversial (9). The molecular diagnosis of FMF is based on various methods, including tedious and time-consuming scanning techniques such as denaturing gradient gel electrophoresis (DGGE) (3) or direct sequencing (10). Restriction enzyme analysis enables the detection of known mutations but still requires multiple DNA amplifications. An in-house amplification refractory mutation system has been designed, but it detects only three mutations (11). Because the spectrum of the most frequent mutations in FMF has been characterized in all at-risk populations, a new method aimed at identifying a set of selected common mutations in a single step would be less time-consuming and would provide greater throughput than previous methods.
We investigated the practicality and the reliability of the FMF StripAssay (ViennaLab Labordiagnostika), the first commercially available assay that allows the detection of 12 MEFV gene mutations in 1 working day. Twelve mutations located in exons 2 (E148Q), 3 (P369S), 5 (F479L), and 10 [M680I (G>C), M680I (G>A), I692del, M694V, M694I, K695R, V726A, A744S, and R761H] can be simultaneously screened for by a reverse hybridization procedure.
The assay includes four successive steps for which reagents are provided: (a) DNA isolation from blood samples; (b) in vitro multiplex amplification reaction; (c) hybridization of amplification products to a test strip comprising a staining control, 8 wild-type-, and 12 mutant-specific immobilized oligonucleotide probes; and (d) detection of bound biotinylated sequences by streptavidin–alkaline phosphatase and color substrate. According to the protocol recommended by the manufacturer, DNA is extracted from 100 µL of whole blood (EDTA or citrate) to obtain 150 µL of a 35–80 ng/µL DNA solution, in a procedure that requires 
80 min. Five microliters of DNA template are amplified by a multiplex PCR in a single tube, requiring 2.5 h. The products from the 20-µL PCR reaction may be checked by visualizing four amplicons of 206, 236, 295, and 318 bp on a 3% agarose, ethidium bromide-stained electrophoresis gel. The hybridization can be performed either in a water bath with a shaking platform (50 rpm) and temperature adjustable to 45 °C, which requires 3 h, or in an automatic incubator, which requires 2.5 h.
After hybridization at 45 °C, stringent wash at 45 °C, and color development at room temperature, the results are interpreted with a coding table. In theory, the presence of only the wild-type signals corresponds to the absence of the 12 tested mutations, but it does not exclude the existence of one or two rare mutations. Heterozygous genotypes produce hybridization to mutant probe(s). One of the wild-type lines disappears in the case of a homozygous mutant or of a compound heterozygote for neighboring mutations, e.g., M694V and M964I (Fig. 1
) or M680I (G>C) and M680I (G>A). Staining intensities of positive lines may vary; however, this is of no significance for interpretation of the results.
|
We successively assessed the reverse hybridization procedure in terms of its specificity and sensitivity, the influence of temperature, the practicability of the assay if performed manually or with use of an automated incubator, and the efficacy of the DNA extraction protocol provided with the reagent set in comparison with our in-house phenol–chloroform extraction. Amplifications were conducted on an Applied Biosystems thermocycler 9700, using the protocol recommended by the manufacturer, and hybridizations were performed manually or in an AutoLIPATM automated incubator (Innogenetics), which can process up to 30 strips simultaneously.
We evaluated the specificity and the sensitivity of the reverse hybridization technique by testing 65 DNA samples from FMF patients (n = 32), carriers (n = 24), or control individuals who did not carry any of the 12 mutations (n = 9). Written informed consent for genetic studies was given by the patients and the healthy controls. DNA samples were obtained by a classic phenol-chloroform extraction and were successfully amplified. Reverse hybridization was performed according to an automated procedure. All samples had been previously characterized in our laboratory by the reference method, i.e., DGGE analysis of exons 10, 5, 3, and 2 and their intron-flanking regions, followed by restriction enzyme analysis or direct sequencing (3). The genotypes identified by both techniques are indicated in Table 1
of the Data Supplement that accompanies the online version of this Technical Brief at http://www.clinchem.org/content/vol49/issue11/. With regard to the mutations screened by the FMF StripAssay, concordance with the reference method was 100%.
|
To evaluate potential interferences leading to erroneous interpretation attributable to sequence variations located at primer or probe binding sequences, we tested DNA samples possessing mutations or polymorphisms situated in analyzed regions but differing from the 12 mutations detected by the assay (Table 1
in the online Data Supplement). As expected, we found no nonspecific hybridization to mutant probes with the following sequence variations situated in exons 2 (G138G, E148V, A165A, E167D, and R202Q), 3 (R354W and R408Q), 5 (E474E, Q476Q, and D510D), and 10 (S675N, S683S, A701A, S703S, P706P, F721F, and D723D). However, in the absence of compound heterozygous (such as E148Q/E148V) or homozygous samples containing those variants, we could not assess the affinity of these alleles for the corresponding wild-type oligonucleotides immobilized on the strips and the possible repercussion on the annealing of PCR primers. The occurrence of such pitfalls should be considered for any technique by screening laboratories.
We also evaluated the influence of the incubating temperature on the sensitivity of the method. A M694V/M694I PCR product (Fig. 1
) and a M694V/V726A PCR product were incubated at 43, 44, 45, 46, and 47 °C in the automatic incubator, 45 °C being the recommended temperature. The appearance of nonspecific hybridization to the M680I (G>A) probe at the two lower temperatures (44 and 43 °C) and to the A744S, R761H, and wild-type codon 692–695 probes at 43 °C highlighted the crucial role of temperature, especially when performing the manual procedure (data not shown). Hybridizations at higher temperatures (46 and 47 °C) produced a markedly reduced signal, which affected all of the probes.
We evaluated the convenience of using the manual protocol for five samples [M694V/M694V, M694V/V726A, M694V/N, V726A/N, and N/N (N/N indicates that none of the 12 mutations detected with the assay was present)]. The hybridization step was performed in 3 h (instead of 2.5 h with the automated procedure) and required continuous technical presence. We encountered nonspecific hybridization to the M680I (G>A) probe. The manual method required strict attention to the protocol, especially concerning hybridization temperature. The use of an automated incubator facilitated temperature management and yielded substantial saving of time, but it consumed more reagent.
We also assessed the DNA extraction protocol with the reagents included with the assay. Blood samples from 31 consecutive FMF patients and unaffected relatives were divided into two sets, and DNA was extracted in parallel and analyzed by two procedures: (a) phenol–chloroform extraction and DGGE analysis; and (b) DNA extraction and reverse hybridization analysis using the FMF StripAssay protocols. Tested genotypes were M694V/M694V (n = 3), M694V/V726A (n = 2), M694V/K695R (n = 2), M694I/V726A (n = 1), V726A/F479L (n = 1), M694V/N (n = 8), V726A/N (n = 4), M694I/N (n = 2), and N/N (n = 8). Using the extraction method included with the assay, we obtained DNA concentrations ranging from 33 to 72 µg/L (35–80 µg/L was the estimated yield from the manufacturer). The yield of the DNA extraction protocol included in the assay allows repeated tests from a limited blood amount (100 µL). We successfully amplified all of the DNA samples that had been extracted by a phenol–chloroform method and those obtained with the reagents provided with the assay. The results of the two procedures were fully concordant.
Because FMF lacks both specific clinical symptoms and biochemical abnormalities and is particularly frequent in some Mediterranean populations, the identification of two allelic MEFV mutations is of utmost diagnostic value (3). The availability of a test to detect the most frequent MEFV gene mutations simultaneously should prompt molecular genetics laboratories to investigate patients with recurrent fever on a routine basis. The FMF StripAssay is technically suitable for use in any diagnostic laboratory, allows the simple and rapid screening of up to 30 samples when an automated incubator is used, and can detect the most frequent MEFV mutated alleles in 1 working day. The assay gives reliable results as long as incubation recommendations are followed, especially for temperature, which can be controlled best by the use of an automated incubator. The reverse hybridization assay does not detect rare mutations; nevertheless, considering the mutated alleles characterized in our laboratory over the past 3 years, the use of the reverse-hybridization method would have detected 99.3% of the MEFV mutations characterized by the DGGE strategy for exons 2, 3, 5, and 10 (Table 1
). Identification of polymorphisms by DGGE still appears useful for segregation studies in families in which the patients present with a FMF or FMF-like phenotype, possibly allowing ruling-out of involvement of the MEFV locus in the occurrence of the disease. The high cost of the FMF StripAssay may be a limitation (€850–950 for 20 tests) especially when making use of an automatic incubator, which further raises its cost, but its use provides savings in time.
Acknowledgments
We wish to thank Laurent Skopinski and Danielle Huot for valuable technical assistance, Dr. Catherine Dodé and Christophe Pêcheux (Service de Biochemiè-Génétique, Hôpital Cochin, Paris, France) for providing the E148Q-I692del/E148Q DNA sample, Ingen (Rungis, France) and AmpliTech (Compiegne, France) for providing reagent sets for our study, and Sean Crosson for critical reading of the manuscript.
Footnotes
* the two first authors contributed equally to this work; 
References
The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  Q. Yao and D. E. Furst Autoinflammatory diseases: an update of clinical and genetic aspects Rheumatology, July 1, 2008; 47(7): 946 - 951. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Hultin, M. Kaller, A. Ahmadian, and J. Lundeberg Competitive enzymatic reaction to control allele-specific extensions Nucleic Acids Res., March 14, 2005; 33(5): e48 - e48. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 I. Touitou Standardized Testing for Mutations in Familial Mediterranean Fever Clin. Chem., November 1, 2003; 49(11): 1781 - 1782. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
