Genes and Immunity (2014) 15, 190–194 & 2014 Macmillan Publishers Limited All rights reserved 1466-4879/14 www.nature.com/gene
SHORT COMMUNICATION
SPAG7 is a candidate gene for the periodic fever, aphthous stomatitis, pharyngitis and adenopathy (PFAPA) syndrome S Bens1, T Zichner2, AM Stu¨tz2, A Caliebe1, R Wagener1, K Hoff1,3, JO Korbel2, P von Bismarck4 and R Siebert1 Periodic fever, aphthous stomatitis, pharyngitis and adenopathy (PFAPA) syndrome is an auto-inflammatory disease for which a genetic basis has been postulated. Nevertheless, in contrast to the other periodic fever syndromes, no candidate genes have yet been identified. By cloning, following long insert size paired-end sequencing, of a de novo chromosomal translocation t(10;17)(q11.2;p13) in a patient with typical PFAPA syndrome lacking mutations in genes associated with other periodic fever syndromes we identified SPAG7 as a candidate gene for PFAPA. SPAG7 protein is expressed in tissues affected by PFAPA and has been functionally linked to antiviral and inflammatory responses. Haploinsufficiency of SPAG7 due to a microdeletion at the translocation breakpoint leading to loss of exons 2–7 from one allele was associated with PFAPA in the index. Sequence analyses of SPAG7 in additional patients with PFAPA point to genetic heterogeneity or alternative mechanisms of SPAG7 deregulation, such as somatic or epigenetic changes. Genes and Immunity (2014) 15, 190–194; doi:10.1038/gene.2013.73; published online 23 January 2014 Keywords: SPAG7; PFAPA; periodic fever; chromosomal translocation
INTRODUCTION Periodic fever, aphthous stomatitis, pharyngitis and adenopathy (PFAPA) syndrome is an auto-inflammatory disorder first described by Marshall et al. in 1987.1 It is characterised by fever episodes lasting 3–6 days with a recurrence every 3–8 weeks, exudative tonsillitis, cervical adenopathy and malaise in all and aphthae in the majority of patients.2–5 Headache, abdominal pain and arthralgia are also recurrently observed. Owing to the recurrent fever episodes PFAPA has been assigned to the clinical spectrum of (periodic) fever syndromes. These also include familial Mediterranean fever (FMF MIM #134610/#249100, associated gene MEFV*608107), TNF receptor-associated periodic fever syndrome (TRAPS #142680, gene TNFRSF1A*191190), hyperimmunoglobulinaemia D (HIDS #260920, gene MVK*251170), cryopyrin-associated periodic syndrome (CAPS) including familial cold auto-inflammatory syndrome (FCAS1 #120100, gene NLRP3*606416, and FCAS2 #611762, NLRP12*609648), Muckle– Wells syndrome (MWS #191900, gene NLRP3*606416) and neonatal onset multisystem inflammatory disease (NOMID/CINCA #607115, gene NLRP3*606416).6,7 Remarkably, all of these fever syndromes are monogenetically inherited disorders for which the responsible genes have been identified. Interestingly, this does not hold true yet for PFAPA syndrome. Despite the current lack of a candidate gene for PFAPA syndrome, various studies point to the fact that this autoinflammatory disease is not a sporadic disorder but based on a genetic predisposition.8 Major epidemiologic evidence for this assumption comes from the observations by Cochard et al.9 These authors showed that a significant percentage of PFAPA patients presents a positive family history of recurrent fever in general and of PFAPA in particular.9 Indeed, in this study, 10 of 84 (12%) patients had a relative affected with PFAPA syndrome. Even in 38
of 84 (45%) patients there was a family history of recurrent fever.9 The high frequency of a recurrent fever other than PFAPA in the families of the patients is in line with the observation that patients with PFAPA syndrome show an increased probability to be carriers of heterozygous mutations in one of the genes associated with the other familial periodic fever syndromes.10 In particular, heterozygous mutations in the MEFV gene associated with FMF are frequent in PFAPA patients. Such MEFV mutations might modify the clinical course of the disease.11 Overall, molecular data, along with the absence of a clear monogenetic trait despite significant familial clustering, might point to a polygenetic or multifactorial origin of PFAPA rather than a monogenetic predisposition. Part of this predisposition might be conferred by sequence variants in inflammasome-associated genes like NLRP3. These have been described in B20% of patients with PFAPA.10 Such variants might increase the probability of dysregulated interleukin 1b production by monocytes during febrile episodes typical for PFAPA syndrome.10 Overall, the hitherto published functional studies point to a pathophysiologic model of PFAPA syndrome in which microbial triggers activate a cascade that begins with the innate immune system and ultimately recruits activated T cells to the periphery.12 Thus, PFAPA syndrome could be due to an abnormal adaptive immune response to viral or other infectious agents, likely at the level of lymphoid organs, able to induce a rapid activation of cells of innate immunity.8,13 This infectious-origin hypothesis at the level of lymphoid organs would be well in line with the histology of tonsils of patients with PFAPA syndrome, which shows non-specific chronic inflammation, characterised by lymphoid and follicular immunoblastic hyperplasia, focal histiocytic clusters, hyalinising fibrosis, crypt abscesses and keratinising debris.14
1 Institute of Human Genetics, Christian-Albrechts-University Kiel & University Hospital Schleswig-Holstein, Kiel, Germany; 2European Molecular Biology Laboratory (EMBL), Genome Biology Research Unit, Heidelberg, Germany; 3Department of Congenital Heart Disease and Pediatric Cardiology, Christian-Albrechts-University Kiel & University Hospital Schleswig-Holstein, Kiel, Germany and 4Department of Pediatrics, Christian-Albrechts-University Kiel & University Hospital Schleswig-Holstein, Kiel, Germany. Correspondence: Dr S Bens, Institute of Human Genetics, Christian-Albrechts-University Kiel & University Hospital Schleswig-Holstein, Campus Kiel, Arnold-Heller Str. 3 (Haus 10), D-24105 Kiel, Germany. E-mail: [email protected] Received 9 November 2013; revised 15 December 2013; accepted 16 December 2013; published online 23 January 2014
SPAG7 in PFAPA syndrome S Bens et al
191 It has recently been shown that cloning of breakpoints of de novo balanced chromosomal translocations in patients with phenotypic abnormalities of unknown origin can lead to the identification of disease-causing genes.15,16 We have adapted this strategy here to PFAPA and characterised the breakpoints of a de novo chromosomal translocation in a patient with PFAPA syndrome without family history of recurrent fevers and lacking mutations in known genes associated with recurrent fever syndromes, with the use of massively parallel DNA sequencing. By this approach we have identified herein SPAG7 as a novel candidate gene for PFAPA syndrome. The finding of de novo haploinsufficiency of the SPAG7 gene in the patient, the expression of the SPAG7 protein in tonsils and lymph nodes, as well as the evidence for its involvement in antiviral and inflammatory responses are in full agreement with the aboveoutlined models of PFAPA pathogenesis. RESULTS AND DISCUSSION Aiming at identifying potential genetic causes of PFAPA syndrome we investigated whether a chromosomal aberration had been diagnosed in any of the patients presented at our institution with this auto-inflammatory disorder. Using this approach, we identified a girl with typical clinical features of PFAPA syndrome in whom prior to this diagnosis a chromosome analysis on peripheral blood lymphocytes had been performed due to growth retardation, facial dysmorphisms and dysplastic brachymesophalangy V (for a detailed clinical description see the clinical synopsis in Materials and methods). G-banded karyotyping at 400 bphs revealed a cytogenetically balanced translocation between the long arm of one chromosome 10 and the short arm of one chromosome 17 in all 20 analysed metaphases (Figure 1). The karyotype was described as: 46,XX,t(10;17)(q11.2;p13). Remarkably,
chromosome analyses on the peripheral blood of the parents revealed normal karyotypes in both mother and father. Thus, the t(10;17) was a de novo event in the patient. Given that both parents are generally healthy and particularly have neither signs of PFAPA syndrome nor any other reported autoimmune or autoinflammatory disease, the de novo appearance of both PFAPA syndrome and the translocation might point to a causal relationship. Indeed, it has been shown that carriers of a de novo balanced chromosomal rearrangement have a two- to threefold increased risk for congenital anomalies.15,17 Consequently, cloning the breakpoints of such de novo balanced chromosomal rearrangements in patients with phenotypic abnormalities of unknown origin has been successfully applied to identify diseasecausing genes.15,16 In order to exclude a submicroscopic chromosomal imbalance related or unrelated to the chromosomal translocation we performed array-comparative genomic hybridisation with a functional resolution of 0.1 Mb. Besides well-known copy number variations, no imbalances were detected at the resolution applied. In particular, we did not detect any imbalances around the cytogenetically assigned breakpoints. Moreover, pathogenic sequence variants in genes associated with periodic fever syndromes or auto-inflammatory diseases, that is, MEFV, NLRP3, NLRP12, ELA2, TNFRSF1A, MVK and CARD15, as a cause or modifier of the phenotype were excluded. Given the cytogenetically and molecular cytogenetically balanced-appearing change and the lack of mutations in genes associated with periodic fever syndromes, we wondered whether the translocation might lead to disruption or otherwise deregulation of a gene potentially associated with PFAPA. Thus, we aimed at cloning the fusions between chromosomes 10 and 17 on both derivative chromosomes, der(10) and der(17). To this end, we generated a long insert size paired-end18 (mate-pair) library with a
Figure 1. Cytogenetic and molecular characterisation of the translocation t(10;17)(q11.2;p13)dn. (a) Partial karyotype showing chromosomes 10 and 17 of the index patient with PFAPA and t(10;17). The normal chromosomes 10 and 17 are shown on the left side of each pair. The derivative chromosomes 10 and 17, respectively, of the t(10;17)(q11.2;p13) are shown on the right side of each pair. (b) Schematic representation of the results of sequence analysis of the translocation. Top and bottom ideograms of normal chromosomes 10 (grey) and 17 (violet), respectively. The middle panel shows the molecular junctions with the der(10) junction on the left and the der(17) junction on the right side. Deletions on both chromosomes at the breakpoint regions (middle part) lead to loss of most of the SPAG7 gene from chromosome 17. Numbers give breakpoint mapping coordinates in base pairs. Genes (blue) are indicated in the respective exon–intron structure, with arrows indicating transcriptional direction. (c) Confirmation of the breakpoint junctions by Sanger sequencing. The middle part shows the derivative chromosomes 10 and 17 with the annotated breakpoints. Top and bottom panels show electropherograms from Sanger sequencing on both derivative chromosomes with the derivation of the sequences and the breakpoints indicated. & 2014 Macmillan Publishers Limited
Genes and Immunity (2014) 190 – 194
SPAG7 in PFAPA syndrome S Bens et al
192 median insert size of 4600 bp, followed by sequencing the library using 36 bp reads. After mapping and quality filtering, 24.1 million non-redundant read pairs remained, resulting in a physical genome coverage of 38.7x. High-confidence structural variation calls for inter-chromosomal variations were identified and filtered against germline variants identified in samples of the 1000 genomes project19 and an additional whole-genome sequencing study.20 By this approach, high-confidence balanced fusions of chromosomes 10 and 17 near the cytogenetically assigned breakpoints were inferred for both derivative chromosomes. The junctions were amplified by PCR using different primer combinations and validated by Sanger sequencing (Figure 1). This resulted in the exact definition of the fusion breakpoints: on the der(10) chromosome, chr10:43,844,710 was fused to chr17:4,861,900. On the der(17), chr17:4,867,779 was fused to chr10:43,850,287 (all hg19, see Figure 1). At both junctions there was one base identical from both chromosomes. Remarkably, on the molecular level the translocation is not balanced but rather leads to loss of genetic material on both affected chromosomes, that is, a deletion of 5577 bp from chromosome 10 and of 5879 bp from chromosome 17. These losses could also be identified in read-depth analyses of the short read data (Figure 2), which indicates that the respective genomic material is not integrated elsewhere in the genome. Whereas the breakpoints and the associated deletion on chromosome 10 affect no gene, the deletion in chromosome 17 results in the loss of exons 2–7 and, thus, most of the coding region of the SPAG7 gene (Figure 1). In order to investigate whether the second allele of SPAG7 was also altered in the patient, we sequenced all exons and exon–intron boundaries of this gene in the patient. Besides known polymorphisms no sequence
variants were detected. Thus, the haploinsufficiency of SPAG7 might be associated with PFAPA in the proposita. SPAG7 encodes the sperm-associated antigen 7 (aliases ACRP, FSA-1, MGC20134; MIM*610056). The gene has been identified by sequencing cDNAs from CD34 þ hematopoietic stem cells as a homologue of fox sperm acrosomal protein-1.21,22 It is highly conserved and encodes a protein with 227 amino acids (NP_004881), which is widely expressed. Given the typical presentation of PFAPA syndrome, nuclear expression in cells of lymph nodes and tonsils is notable (http://www.proteinatlas.org). The SPAG7 protein contains two nuclear localisation signals (AA35– 51 and 122–139) and a R3H domain (AA46–109) named from the characteristic spacing of the most conserved arginine and histidine residues.23 This domain is predicted to bind ssDNA or ssRNA in a sequence-specific manner.23 This might indicate a potential role of the protein in response to viruses or free (cellular) DNAs or RNAs. In this context it is remarkable that SPAG7 was shown by array- and qPCR-based analyses to be significantly overexpressed in RNA from peripheral blood mononuclear cells from 57 parvovirus B19seropositive as compared with 13 parvovirus B19-seronegative donors, pointing to a role in reaction to viral infections.24 Given the identification of haploinsufficiency of the SPAG7 gene in the index patient along with expression of the encoded protein in lymphatic tissues affected in PFAPA syndrome and functional links of SPAG7 to antiviral and inflammatory responses, we consider SPAG7 a reasonable candidate for PFAPA syndrome. Thus, we investigated whether other patients with this syndrome might also show alterations of the protein. To this end, we sequenced all exons and exon–intron junctions of the SPAG7 gene in four additional patients fulfilling the typical clinical criteria for PFAPA syndrome (Table 1); one was of Turkish (ID 2) and three of German (ID 3–5) descent. Moreover, three patients had a positive family history for febrile episodes during early childhood (Table 1). The Turkish patient lacked a family history but was found to be heterozygous for the mutation c.2080A4G (p.Met694Val) in the MEFV gene associated with FMF. In this small cohort of four patients we did not find any potentially pathogenic mutations in the SPAG7 gene. In summary, through cloning of a molecularly unbalanced de novo chromosomal translocation we identified SPAG7 as a candidate gene for PFAPA. Nevertheless, our analyses on additional patients failed to identify recurrent haploinsufficiency though the cohort was limited in size. We cannot exclude somatic alterations of SPAG7 in the affected tissues or epigenetic changes leading to SPAG7 deregulation in these patients. Alternatively, and in line with previous reports, the syndrome might be genetically heterogeneous and associated with a polygenetic predisposition. In this context, it is notable that the SPAG7 protein has been shown to physically interact with the ABHD16A (abhydrolase domain containing 16A) protein by Y2H screening.25 ABHD16A is encoded by the BAT5 gene located in the HLA gene cluster. Remarkably, the BAT5 locus has also been linked to autoinflammatory diseases like Kawasaki syndrome.26 This might point to a group of interacting proteins involved in the pathogenesis of auto-inflammatory disorders.
MATERIALS AND METHODS Patient cohort Figure 2. Read-depth plot showing the deletions at the breakpoint junctions in chromosomes 10 and 17. The dark-grey bars represent the number of sequencing reads mapping between the breakpoints on chromosome 10 (5577 bp region) and chromosome 17 (5879 bp region). The light-grey bars represent the number of sequencing reads mapping into adjacent regions of identical size up- and downstream of the breakpoints. The lower number of reads between the breakpoints compared with the surrounding regions supports a heterozygous deletion on both chromosomes. Genes and Immunity (2014) 190 – 194
The index patient presented with a ‘typical PFAPA’ syndrome.27 The girl (Table 1 ID 1) was 5 years old when she experienced first febrile episodes with intermittent body temperatures between 39 and 40 1C. During the first two years the febrile temperatures lasted between 3 and 5 days without any concomitant symptoms and recurred with an interval of 8–12 weeks. The girl was asymptomatic in between the fever episodes. An infectious disease, cyclic neutropenia and malignancies or a monogenic hereditary fever syndrome were excluded and a symptomatic therapy with antipyretic drugs was established. In the third year of the disease the girl & 2014 Macmillan Publishers Limited
SPAG7 in PFAPA syndrome S Bens et al
193 Table 1. ID
Clinical description of patients with PFAPA analysed in this study Sex
Age at onset (years)
Length of febrile episodes (days)
Interval between episodes (weeks)
PFAPA symptoms
1 index
F
5
4–5
4–8
PF, A, P, LA
2
F
2,5
4–6
3–6
3 4
F F
3 9
3–5 4–5
8–10 2–4
PF, A, P, LA PF, A, LA PF, P, LA
5
M
1,5
4–6
3–4
PF, A, P, LA
Additional symptoms during episodes
Resolution of episodes to therapy with:
Positive family history for recurrent febrile episodes
Genetic Analysis Hereditary periodic fever syndrome
Cimetidine
normal
Abdominal pain
tonsillectomy
(a)
Abdominal pain, occ. vomiting
Colchicine Cimetidine
þ (father) þ (father)
(b) (b, c)
Colchicine
þ (father)
not done
Abbreviations: A, aphthous stomatitis; F, female; LA, cervical lymphadenitis; M, male; occ, occasionally; P, pharyngitis; PF, periodic fever. Clinical characteristics of all five patients diagnosed with PFAPA syndrome. Age at onset corresponds to the start of febrile episodes. Hereditary periodic fever syndrome (HPF) was excluded by absence of typical clinical signs for HPF (skin rash, arthritis, serositis, splenomegaly, fever episodes longer than 7 days, hearing loss and symptoms secondary to cold exposure). Patient 2 had severe abdominal pain and an ethnic origin from Turkey, which led to a molecular testing for FMF. (a) This patient had a heterozygous mutation in the MEFV gene c.2080A4G (p.Met694Val). (b) Molecular testing of TNFRSF1A exons 2, 3, 4 and 6 including the corresponding exon–intron junctions, with normal results. (c) Molecular testing of MVK exons 9 and 11 including the corresponding exon–intron junctions, with normal results.
developed 2–3 days prior to febrile episodes abdominal pain, nausea and, with the febrile temperature, an aphthous stomatitis. In the later course she suffered additionally from cervical lymphadenitis and pharyngitis during the days with fever. Oral prednisolone caused immediate fever reduction within 2 h but shortened the asymptomatic interval (2–3 weeks). The diagnosis of PFAPA syndrome was made 28 months after the commencement of primary symptoms. A therapy with Cimetidine (20 mg per kg per day) was started, which led to a complete cessation of fever episodes. The medication was tapered after one and a half year, which led to recurrence of febrile episodes. Prior to this auto-inflammatory disease the girl was presented to our hospital with signs of mild growth retardation. Her parents were Caucasian, overall healthy and not related. The girl had during her first two years of life a proportional growth rate shortly below the third percentile. A hypothyreosis and a metabolic disorder were excluded. Further mild stigmata like an ocular hypertelorism and short bones as well as dysplastic brachymesophalangy V of the hand led to karyotyping. In addition, four further children who fulfilled the typical clinical criteria for PFAPA syndrome were selected for molecular genetic testing. All except one had recurrent febrile episodes from an early age (p5 years). None of the patients had signs of an infection during the episodes but presented at least one or two of the following symptoms: aphthous stomatitis, cervical lymphadenitis or pharyngitis. The children were additionally completely asymptomatic between the febrile episodes. Cyclic neutropenia or malignancies (leukemia or neuroblastoma) were excluded. Three patients had a positive family history (one of the parents) for febrile episodes during early childhood. In one patient (Table 1 ID: 2) we found a heterozygous mutation in the MEFV gene (c.2080A4G[p.Met694Val]). This patient presented, except for recurrent fever, no further signs of FMF (rash, arthritis, serositis). All children were initially treated with Cimetidine; second-line therapy was Colchicine. One child who did not respond to the medical therapy received a tonsillectomy. The clinical characteristics of all patients with PFAPA syndrome are summarised in Table 1. All patients agreed to the study and the parents gave written informed consent. The procedures of the study were approved by the Ethics Commission of the Medical Faculty of the Christian-Albrechts-University in Kiel (AZ D401/08 and amendment 08 December 2011).
Karyotying and molecular karyotyping Karyotyping was performed on phytohemagglutinin-stimulated cultures of peripheral blood using G-banding analyses according to standard techniques. Metaphases were captured using the IKAROS software (MetaSystems, Altlussheim, Germany). Karyotypes were described according to ISCN2013. DNA was extracted from the peripheral blood of all patients using standard methods. Array-comparative genomic hybridisation using the Human Genome CGH microarray 244A platform (Agilent Technologies, Santa Clara, CA, USA) was performed according to & 2014 Macmillan Publishers Limited
the manufacturer’s instructions. Pooled peripheral blood DNA from 10 healthy donors with a female karyotype served as hybridisation control. Analysis was performed using the CGH Analytics Software (Agilent Technologies). Regions that were determined as significantly deviated by the software and that included more than 10 single oligonucleotides were regarded as aberrant, thus reaching a functional resolution of around 0.1 Mb. The differentiation between pathologic and benign copy number variations was performed by using a database integrated into the software.
Mate-pair structural rearrangement analysis Mate-pair (MP, or long insert size) DNA library preparation was carried out using the v2 protocol from Illumina Gmbh (Munich, Germany). In brief, 10 mg of genomic DNA was fragmented using a Covaris S2 instrument and red miniTUBE (LGC Genomics Ltd., Hoddesdon, UK) with the manufacturer’s recommended settings to B5 kb, followed by size selection through agarose gel excision. Deep sequencing was carried out with an Illumina GAIIx (2 36 bp paired-end mode) instrument to reach an average physical coverage of at least 30 . Mate-pair reads were aligned to the hg19 assembly of the human reference genome using the Illuminaprovided alignment software (ELAND, version 2). Rearrangements based on mate-pair data were identified using the DELLY program.28 Structural rearrangement calls that were also inferred from 1000 genomes project19 genome data or germline data of additional whole-genome sequencing samples20 were removed to exclude polymorphic germline structural variants as well as rearrangement calls caused by mapping artefacts. Furthermore, rearrangement calls with less than four supporting read pairs as well as supporting pairs with an average mapping quality o20 were excluded. To assess whether the genomic regions between the translocation breakpoints on chromosome 10 and 17 are lost as a consequence of the rearrangement, the number of reads that map within these regions was counted and compared with neighbouring regions of identical size (5577 bp on chromosome 10; 5879 bp on chromosome 17).
PCR and Sanger sequencing For PCR amplification of the translocation breakpoint a primer walking strategy was applied. Primers leading to amplification of the breakpoint and respective sequencing primers are listed in Supplementary Table 1. For sequencing of the SPAG7 gene, primers flanking all exon–intron junctions of the gene were designed as outlined in Supplementary Table 1. PCR was performed according to standard protocols. Direct sequencing of the PCR products using BigDye Terminator technology (BigDye Terminator v1.1 Cycle Sequencing Kit, Applied Biosystems, Foster City, CA, USA) was performed using the same primers as for PCR if not otherwise indicated (Supplementary Table 1) on an ABI 3100 Genetic Analyzer (Applied Biosystems). Alternatively, sequencing at GATC Biotech AG (Constance, Germany) was performed. Genes and Immunity (2014) 190 – 194
SPAG7 in PFAPA syndrome S Bens et al
194 CONFLICT OF INTEREST The authors declare no conflict of interest.
ACKNOWLEDGEMENTS We thank the technical staff of the cytogenetic and molecular genetic laboratories of the Institute of Human Genetics for expert assistance, as well as all clinicians involved in the care of the patients for support and all individuals for their participation. JOK was supported by an Emmy Noether Fellowship (KO 4037/1-1) from the German Research Foundation and RW is recipient of a stipend from the KinderKrebsInitiative Buchholz/Holm-Seppensen. RS is a member of the Excellence Cluster ‘Inflammation at Interfaces’.
REFERENCES 1 Marshall GS, Edwards KM, Butler J, Lawton AR. Syndrome of periodic fever, pharyngitis, and aphthous stomatitis. J Pediatr 1987; 110: 43–46. 2 Padeh S, Brezniak N, Zemer D, Pras E, Livneh A, Langevitz P et al. Periodic fever, aphtous stomatitis, pharyngitis, and adenopathy syndrome: clinical characteristics and outcome. J Pediatr 1999; 135: 98–101. 3 Tasher D, Somekh E, Dalal I. PFAPA syndrome: new clinical aspects disclosed. Arch Dis Child 2006; 91: 981–984. 4 Vigo G, Zulian F. Periodic fevers with aphthous stomatitis, and adenitis (PFAPA). Autoimmun Rev 2012; 12: 52–55. 5 Feder HM, Salazar JC. A clinical review of 105 patients with PFAPA (a periodic fever syndrome). Acta Paediatr 2010; 99: 178–184. 6 Marcuzzi A, Piscianz E, Kleiner G, Tommasini A, Severini GM, Monasta L et al. Clinical genetic testing of periodic fever syndromes. Biomed Res Int 2013; 147: 155–174. 7 Savic S, Dickie LJ, Battelino M, McDermott MF. Familial Mediterranean fever and related periodic fever syndromes/autoinflammatory diseases. Curr Opin Rheumatol 2012; 24: 103–112. 8 Akelma AZ, Cizmeci MN, Kanburoglu MK, Mete E, Bozkaya D, Tufan N et al. Is PFAPA syndrome really a sporadic disorder or is it genetic? Med Hypotheses 2013; 81: 279–281. 9 Cochard M, Clet J, Le L, Pillet P, Onrubia X, Gue´ron T et al. PFAPA syndrome is not a sporadic disease. Rheumatology (Oxford) 2010; 49: 1984–1987. 10 Kolly L, Busso N, von Scheven-Gete A, Bagnoud N, Moix I, Holzinger D et al. Periodic fever, aphthous stomatitis, pharyngitis, cervical adenitis syndrome is linked to dysregulated monocyte IL-1b production. J Allergy Clin Immunol 2013; 131: 1635–1643. 11 Berkun Y, Levy R, Hurwitz A, Meir-Harel M, Lidar M, Livneh A et al. The familial Mediterranean fever gene as a modifier of periodic fever, aphthous stomatitis, pharyngitis, and adenopathy syndrome. Semin Arthritis Rheum 2011; 40: 467–472. 12 Stojanov S, Lapidus S, Chitkara P, Feder H, Salazar JC, Fleisher TA et al. Periodic fever, aphthous stomatitis, pharyngitis, and adenitis (PFAPA) is a disorder of innate immunity and Th1 activation responsive to IL-1 blockade. Proc Natl Acad Sci USA 2011; 108: 7148–7153.
13 Caorsi R, Pelagatti MA, Federici S, Finetti M, Martini A, Gattorno M. Periodic fever, aphthous stomatitis, pharyngitis and adenitis syndrome. Curr Opin Rheumatol 2010; 22: 579–584. 14 Peridis S, Pilgrim G, Koudoumnakis E, Athanasopoulos I, Houlakis M, Parpounas K. PFAPA syndrome in children: a meta-analysis on surgical versus medical treatment. Int J Pediatr Otorhinolaryngol 2010; 74: 1203–1208. 15 Higgins AW, Alkuraya FS, Bosco AF, Brown KK, Bruns GA, Donovan DJ et al. Characterization of apparently balanced chromosomal rearrangements from the developmental genome anatomy project. Am J Hum Genet 2008; 82: 712–722. 16 Cacciagli P, Haddad MR, Mignon-Ravix C, El-Waly B, Moncla A, Missirian C et al. Disruption of the ATP8A2 gene in a patient with a t(10;13) de novo balanced translocation and a severe neurological phenotype. Eur J Hum Genet 2010; 18: 1360–1363. 17 Warburton D. De novo balanced chromosome rearrangements and extra marker chromosomes identified at prenatal diagnosis: clinical significance and distribution of breakpoints. Am J Hum Genet 1991; 49: 995–1013. 18 Korbel JO, Urban AE, Affourtit JP, Godwin B, Grubert F, Simons JF et al. Paired-end mapping reveals extensive structural variation in the human genome. Science 2007; 318: 420–426. 19 1000 Genomes Project Consortium, Abecasis GR, Auton A, Brooks LD, DePristo MA, Durbin RM et al. An integrated map of genetic variation from 1,092 human genomes. Nature 2012; 491: 56–65. 20 Jones DT, Ja¨ger N, Kool M, Zichner T, Hutter B, Sultan M et al. Dissecting the genomic complexity underlying medulloblastoma. Nature 2012; 488: 100–105. 21 Mao M, Fu G, Wu JS, Zhang QH, Zhou J, Kan LX et al. Identification of genes expressed in human CD34( þ ) hematopoietic stem/progenitor cells by expressed sequence tags and efficient full-length cDNA cloning. Proc Natl Acad Sci USA 1998; 95: 8175–8180. 22 Beaton S, Cleary A, ten Have J, Bradley MP. Cloning and characterization of a fox sperm protein FSA-1. Reprod Fertil Dev 1994; 6: 761–770. 23 Grishin NV. The R3H motif: a domain that binds single-stranded nucleic acids. Trends Biochem Sci 1998; 23: 329–330. 24 Kerr JR. Pathogenesis of parvovirus B19 infection: host gene variability, and possible means and effects of virus persistence. J Vet Med B Infect Dis Vet Public Health 2005; 52: 335–339. 25 Lehner B, Semple JI, Brown SE, Counsell D, Campbell RD, Sanderson CM. Analysis of a high-throughput yeast two-hybrid system and its use to predict the function of intracellular proteins encoded within the human MHC class III region. Genomics 2004; 83: 153–167. 26 Hsieh YY, Lin YJ, Chang CC, Chen CY, Hsu CM, Wang YK et al. Human lymphocyte antigen B-associated transcript 2,3, and 5 polymorphisms and haplotypes are associated with susceptibility of Kawasaki disease and coronary artery aneurysm. J Clin Lab Anal 2010; 24: 262–268. 27 Thomas KT, Feder HMJ, Lawton AR, Edwards KM. Periodic fever syndrome in children. J Pediatr 1999; 135: 15–21. 28 Rausch T, Zichner T, Schlattl A, Stu¨tz AM, Benes V, Korbel JO. DELLY: structural variant discovery by integrated paired-end and split-read analysis. Bioinformatics 2012; 28: i333–i339.
Supplementary Information accompanies this paper on Genes and Immunity website (http://www.nature.com/gene)
Genes and Immunity (2014) 190 – 194
& 2014 Macmillan Publishers Limited
SHORT COMMUNICATION
SPAG7 is a candidate gene for the periodic fever, aphthous stomatitis, pharyngitis and adenopathy (PFAPA) syndrome S Bens1, T Zichner2, AM Stu¨tz2, A Caliebe1, R Wagener1, K Hoff1,3, JO Korbel2, P von Bismarck4 and R Siebert1 Periodic fever, aphthous stomatitis, pharyngitis and adenopathy (PFAPA) syndrome is an auto-inflammatory disease for which a genetic basis has been postulated. Nevertheless, in contrast to the other periodic fever syndromes, no candidate genes have yet been identified. By cloning, following long insert size paired-end sequencing, of a de novo chromosomal translocation t(10;17)(q11.2;p13) in a patient with typical PFAPA syndrome lacking mutations in genes associated with other periodic fever syndromes we identified SPAG7 as a candidate gene for PFAPA. SPAG7 protein is expressed in tissues affected by PFAPA and has been functionally linked to antiviral and inflammatory responses. Haploinsufficiency of SPAG7 due to a microdeletion at the translocation breakpoint leading to loss of exons 2–7 from one allele was associated with PFAPA in the index. Sequence analyses of SPAG7 in additional patients with PFAPA point to genetic heterogeneity or alternative mechanisms of SPAG7 deregulation, such as somatic or epigenetic changes. Genes and Immunity (2014) 15, 190–194; doi:10.1038/gene.2013.73; published online 23 January 2014 Keywords: SPAG7; PFAPA; periodic fever; chromosomal translocation
INTRODUCTION Periodic fever, aphthous stomatitis, pharyngitis and adenopathy (PFAPA) syndrome is an auto-inflammatory disorder first described by Marshall et al. in 1987.1 It is characterised by fever episodes lasting 3–6 days with a recurrence every 3–8 weeks, exudative tonsillitis, cervical adenopathy and malaise in all and aphthae in the majority of patients.2–5 Headache, abdominal pain and arthralgia are also recurrently observed. Owing to the recurrent fever episodes PFAPA has been assigned to the clinical spectrum of (periodic) fever syndromes. These also include familial Mediterranean fever (FMF MIM #134610/#249100, associated gene MEFV*608107), TNF receptor-associated periodic fever syndrome (TRAPS #142680, gene TNFRSF1A*191190), hyperimmunoglobulinaemia D (HIDS #260920, gene MVK*251170), cryopyrin-associated periodic syndrome (CAPS) including familial cold auto-inflammatory syndrome (FCAS1 #120100, gene NLRP3*606416, and FCAS2 #611762, NLRP12*609648), Muckle– Wells syndrome (MWS #191900, gene NLRP3*606416) and neonatal onset multisystem inflammatory disease (NOMID/CINCA #607115, gene NLRP3*606416).6,7 Remarkably, all of these fever syndromes are monogenetically inherited disorders for which the responsible genes have been identified. Interestingly, this does not hold true yet for PFAPA syndrome. Despite the current lack of a candidate gene for PFAPA syndrome, various studies point to the fact that this autoinflammatory disease is not a sporadic disorder but based on a genetic predisposition.8 Major epidemiologic evidence for this assumption comes from the observations by Cochard et al.9 These authors showed that a significant percentage of PFAPA patients presents a positive family history of recurrent fever in general and of PFAPA in particular.9 Indeed, in this study, 10 of 84 (12%) patients had a relative affected with PFAPA syndrome. Even in 38
of 84 (45%) patients there was a family history of recurrent fever.9 The high frequency of a recurrent fever other than PFAPA in the families of the patients is in line with the observation that patients with PFAPA syndrome show an increased probability to be carriers of heterozygous mutations in one of the genes associated with the other familial periodic fever syndromes.10 In particular, heterozygous mutations in the MEFV gene associated with FMF are frequent in PFAPA patients. Such MEFV mutations might modify the clinical course of the disease.11 Overall, molecular data, along with the absence of a clear monogenetic trait despite significant familial clustering, might point to a polygenetic or multifactorial origin of PFAPA rather than a monogenetic predisposition. Part of this predisposition might be conferred by sequence variants in inflammasome-associated genes like NLRP3. These have been described in B20% of patients with PFAPA.10 Such variants might increase the probability of dysregulated interleukin 1b production by monocytes during febrile episodes typical for PFAPA syndrome.10 Overall, the hitherto published functional studies point to a pathophysiologic model of PFAPA syndrome in which microbial triggers activate a cascade that begins with the innate immune system and ultimately recruits activated T cells to the periphery.12 Thus, PFAPA syndrome could be due to an abnormal adaptive immune response to viral or other infectious agents, likely at the level of lymphoid organs, able to induce a rapid activation of cells of innate immunity.8,13 This infectious-origin hypothesis at the level of lymphoid organs would be well in line with the histology of tonsils of patients with PFAPA syndrome, which shows non-specific chronic inflammation, characterised by lymphoid and follicular immunoblastic hyperplasia, focal histiocytic clusters, hyalinising fibrosis, crypt abscesses and keratinising debris.14
1 Institute of Human Genetics, Christian-Albrechts-University Kiel & University Hospital Schleswig-Holstein, Kiel, Germany; 2European Molecular Biology Laboratory (EMBL), Genome Biology Research Unit, Heidelberg, Germany; 3Department of Congenital Heart Disease and Pediatric Cardiology, Christian-Albrechts-University Kiel & University Hospital Schleswig-Holstein, Kiel, Germany and 4Department of Pediatrics, Christian-Albrechts-University Kiel & University Hospital Schleswig-Holstein, Kiel, Germany. Correspondence: Dr S Bens, Institute of Human Genetics, Christian-Albrechts-University Kiel & University Hospital Schleswig-Holstein, Campus Kiel, Arnold-Heller Str. 3 (Haus 10), D-24105 Kiel, Germany. E-mail: [email protected] Received 9 November 2013; revised 15 December 2013; accepted 16 December 2013; published online 23 January 2014
SPAG7 in PFAPA syndrome S Bens et al
191 It has recently been shown that cloning of breakpoints of de novo balanced chromosomal translocations in patients with phenotypic abnormalities of unknown origin can lead to the identification of disease-causing genes.15,16 We have adapted this strategy here to PFAPA and characterised the breakpoints of a de novo chromosomal translocation in a patient with PFAPA syndrome without family history of recurrent fevers and lacking mutations in known genes associated with recurrent fever syndromes, with the use of massively parallel DNA sequencing. By this approach we have identified herein SPAG7 as a novel candidate gene for PFAPA syndrome. The finding of de novo haploinsufficiency of the SPAG7 gene in the patient, the expression of the SPAG7 protein in tonsils and lymph nodes, as well as the evidence for its involvement in antiviral and inflammatory responses are in full agreement with the aboveoutlined models of PFAPA pathogenesis. RESULTS AND DISCUSSION Aiming at identifying potential genetic causes of PFAPA syndrome we investigated whether a chromosomal aberration had been diagnosed in any of the patients presented at our institution with this auto-inflammatory disorder. Using this approach, we identified a girl with typical clinical features of PFAPA syndrome in whom prior to this diagnosis a chromosome analysis on peripheral blood lymphocytes had been performed due to growth retardation, facial dysmorphisms and dysplastic brachymesophalangy V (for a detailed clinical description see the clinical synopsis in Materials and methods). G-banded karyotyping at 400 bphs revealed a cytogenetically balanced translocation between the long arm of one chromosome 10 and the short arm of one chromosome 17 in all 20 analysed metaphases (Figure 1). The karyotype was described as: 46,XX,t(10;17)(q11.2;p13). Remarkably,
chromosome analyses on the peripheral blood of the parents revealed normal karyotypes in both mother and father. Thus, the t(10;17) was a de novo event in the patient. Given that both parents are generally healthy and particularly have neither signs of PFAPA syndrome nor any other reported autoimmune or autoinflammatory disease, the de novo appearance of both PFAPA syndrome and the translocation might point to a causal relationship. Indeed, it has been shown that carriers of a de novo balanced chromosomal rearrangement have a two- to threefold increased risk for congenital anomalies.15,17 Consequently, cloning the breakpoints of such de novo balanced chromosomal rearrangements in patients with phenotypic abnormalities of unknown origin has been successfully applied to identify diseasecausing genes.15,16 In order to exclude a submicroscopic chromosomal imbalance related or unrelated to the chromosomal translocation we performed array-comparative genomic hybridisation with a functional resolution of 0.1 Mb. Besides well-known copy number variations, no imbalances were detected at the resolution applied. In particular, we did not detect any imbalances around the cytogenetically assigned breakpoints. Moreover, pathogenic sequence variants in genes associated with periodic fever syndromes or auto-inflammatory diseases, that is, MEFV, NLRP3, NLRP12, ELA2, TNFRSF1A, MVK and CARD15, as a cause or modifier of the phenotype were excluded. Given the cytogenetically and molecular cytogenetically balanced-appearing change and the lack of mutations in genes associated with periodic fever syndromes, we wondered whether the translocation might lead to disruption or otherwise deregulation of a gene potentially associated with PFAPA. Thus, we aimed at cloning the fusions between chromosomes 10 and 17 on both derivative chromosomes, der(10) and der(17). To this end, we generated a long insert size paired-end18 (mate-pair) library with a
Figure 1. Cytogenetic and molecular characterisation of the translocation t(10;17)(q11.2;p13)dn. (a) Partial karyotype showing chromosomes 10 and 17 of the index patient with PFAPA and t(10;17). The normal chromosomes 10 and 17 are shown on the left side of each pair. The derivative chromosomes 10 and 17, respectively, of the t(10;17)(q11.2;p13) are shown on the right side of each pair. (b) Schematic representation of the results of sequence analysis of the translocation. Top and bottom ideograms of normal chromosomes 10 (grey) and 17 (violet), respectively. The middle panel shows the molecular junctions with the der(10) junction on the left and the der(17) junction on the right side. Deletions on both chromosomes at the breakpoint regions (middle part) lead to loss of most of the SPAG7 gene from chromosome 17. Numbers give breakpoint mapping coordinates in base pairs. Genes (blue) are indicated in the respective exon–intron structure, with arrows indicating transcriptional direction. (c) Confirmation of the breakpoint junctions by Sanger sequencing. The middle part shows the derivative chromosomes 10 and 17 with the annotated breakpoints. Top and bottom panels show electropherograms from Sanger sequencing on both derivative chromosomes with the derivation of the sequences and the breakpoints indicated. & 2014 Macmillan Publishers Limited
Genes and Immunity (2014) 190 – 194
SPAG7 in PFAPA syndrome S Bens et al
192 median insert size of 4600 bp, followed by sequencing the library using 36 bp reads. After mapping and quality filtering, 24.1 million non-redundant read pairs remained, resulting in a physical genome coverage of 38.7x. High-confidence structural variation calls for inter-chromosomal variations were identified and filtered against germline variants identified in samples of the 1000 genomes project19 and an additional whole-genome sequencing study.20 By this approach, high-confidence balanced fusions of chromosomes 10 and 17 near the cytogenetically assigned breakpoints were inferred for both derivative chromosomes. The junctions were amplified by PCR using different primer combinations and validated by Sanger sequencing (Figure 1). This resulted in the exact definition of the fusion breakpoints: on the der(10) chromosome, chr10:43,844,710 was fused to chr17:4,861,900. On the der(17), chr17:4,867,779 was fused to chr10:43,850,287 (all hg19, see Figure 1). At both junctions there was one base identical from both chromosomes. Remarkably, on the molecular level the translocation is not balanced but rather leads to loss of genetic material on both affected chromosomes, that is, a deletion of 5577 bp from chromosome 10 and of 5879 bp from chromosome 17. These losses could also be identified in read-depth analyses of the short read data (Figure 2), which indicates that the respective genomic material is not integrated elsewhere in the genome. Whereas the breakpoints and the associated deletion on chromosome 10 affect no gene, the deletion in chromosome 17 results in the loss of exons 2–7 and, thus, most of the coding region of the SPAG7 gene (Figure 1). In order to investigate whether the second allele of SPAG7 was also altered in the patient, we sequenced all exons and exon–intron boundaries of this gene in the patient. Besides known polymorphisms no sequence
variants were detected. Thus, the haploinsufficiency of SPAG7 might be associated with PFAPA in the proposita. SPAG7 encodes the sperm-associated antigen 7 (aliases ACRP, FSA-1, MGC20134; MIM*610056). The gene has been identified by sequencing cDNAs from CD34 þ hematopoietic stem cells as a homologue of fox sperm acrosomal protein-1.21,22 It is highly conserved and encodes a protein with 227 amino acids (NP_004881), which is widely expressed. Given the typical presentation of PFAPA syndrome, nuclear expression in cells of lymph nodes and tonsils is notable (http://www.proteinatlas.org). The SPAG7 protein contains two nuclear localisation signals (AA35– 51 and 122–139) and a R3H domain (AA46–109) named from the characteristic spacing of the most conserved arginine and histidine residues.23 This domain is predicted to bind ssDNA or ssRNA in a sequence-specific manner.23 This might indicate a potential role of the protein in response to viruses or free (cellular) DNAs or RNAs. In this context it is remarkable that SPAG7 was shown by array- and qPCR-based analyses to be significantly overexpressed in RNA from peripheral blood mononuclear cells from 57 parvovirus B19seropositive as compared with 13 parvovirus B19-seronegative donors, pointing to a role in reaction to viral infections.24 Given the identification of haploinsufficiency of the SPAG7 gene in the index patient along with expression of the encoded protein in lymphatic tissues affected in PFAPA syndrome and functional links of SPAG7 to antiviral and inflammatory responses, we consider SPAG7 a reasonable candidate for PFAPA syndrome. Thus, we investigated whether other patients with this syndrome might also show alterations of the protein. To this end, we sequenced all exons and exon–intron junctions of the SPAG7 gene in four additional patients fulfilling the typical clinical criteria for PFAPA syndrome (Table 1); one was of Turkish (ID 2) and three of German (ID 3–5) descent. Moreover, three patients had a positive family history for febrile episodes during early childhood (Table 1). The Turkish patient lacked a family history but was found to be heterozygous for the mutation c.2080A4G (p.Met694Val) in the MEFV gene associated with FMF. In this small cohort of four patients we did not find any potentially pathogenic mutations in the SPAG7 gene. In summary, through cloning of a molecularly unbalanced de novo chromosomal translocation we identified SPAG7 as a candidate gene for PFAPA. Nevertheless, our analyses on additional patients failed to identify recurrent haploinsufficiency though the cohort was limited in size. We cannot exclude somatic alterations of SPAG7 in the affected tissues or epigenetic changes leading to SPAG7 deregulation in these patients. Alternatively, and in line with previous reports, the syndrome might be genetically heterogeneous and associated with a polygenetic predisposition. In this context, it is notable that the SPAG7 protein has been shown to physically interact with the ABHD16A (abhydrolase domain containing 16A) protein by Y2H screening.25 ABHD16A is encoded by the BAT5 gene located in the HLA gene cluster. Remarkably, the BAT5 locus has also been linked to autoinflammatory diseases like Kawasaki syndrome.26 This might point to a group of interacting proteins involved in the pathogenesis of auto-inflammatory disorders.
MATERIALS AND METHODS Patient cohort Figure 2. Read-depth plot showing the deletions at the breakpoint junctions in chromosomes 10 and 17. The dark-grey bars represent the number of sequencing reads mapping between the breakpoints on chromosome 10 (5577 bp region) and chromosome 17 (5879 bp region). The light-grey bars represent the number of sequencing reads mapping into adjacent regions of identical size up- and downstream of the breakpoints. The lower number of reads between the breakpoints compared with the surrounding regions supports a heterozygous deletion on both chromosomes. Genes and Immunity (2014) 190 – 194
The index patient presented with a ‘typical PFAPA’ syndrome.27 The girl (Table 1 ID 1) was 5 years old when she experienced first febrile episodes with intermittent body temperatures between 39 and 40 1C. During the first two years the febrile temperatures lasted between 3 and 5 days without any concomitant symptoms and recurred with an interval of 8–12 weeks. The girl was asymptomatic in between the fever episodes. An infectious disease, cyclic neutropenia and malignancies or a monogenic hereditary fever syndrome were excluded and a symptomatic therapy with antipyretic drugs was established. In the third year of the disease the girl & 2014 Macmillan Publishers Limited
SPAG7 in PFAPA syndrome S Bens et al
193 Table 1. ID
Clinical description of patients with PFAPA analysed in this study Sex
Age at onset (years)
Length of febrile episodes (days)
Interval between episodes (weeks)
PFAPA symptoms
1 index
F
5
4–5
4–8
PF, A, P, LA
2
F
2,5
4–6
3–6
3 4
F F
3 9
3–5 4–5
8–10 2–4
PF, A, P, LA PF, A, LA PF, P, LA
5
M
1,5
4–6
3–4
PF, A, P, LA
Additional symptoms during episodes
Resolution of episodes to therapy with:
Positive family history for recurrent febrile episodes
Genetic Analysis Hereditary periodic fever syndrome
Cimetidine
normal
Abdominal pain
tonsillectomy
(a)
Abdominal pain, occ. vomiting
Colchicine Cimetidine
þ (father) þ (father)
(b) (b, c)
Colchicine
þ (father)
not done
Abbreviations: A, aphthous stomatitis; F, female; LA, cervical lymphadenitis; M, male; occ, occasionally; P, pharyngitis; PF, periodic fever. Clinical characteristics of all five patients diagnosed with PFAPA syndrome. Age at onset corresponds to the start of febrile episodes. Hereditary periodic fever syndrome (HPF) was excluded by absence of typical clinical signs for HPF (skin rash, arthritis, serositis, splenomegaly, fever episodes longer than 7 days, hearing loss and symptoms secondary to cold exposure). Patient 2 had severe abdominal pain and an ethnic origin from Turkey, which led to a molecular testing for FMF. (a) This patient had a heterozygous mutation in the MEFV gene c.2080A4G (p.Met694Val). (b) Molecular testing of TNFRSF1A exons 2, 3, 4 and 6 including the corresponding exon–intron junctions, with normal results. (c) Molecular testing of MVK exons 9 and 11 including the corresponding exon–intron junctions, with normal results.
developed 2–3 days prior to febrile episodes abdominal pain, nausea and, with the febrile temperature, an aphthous stomatitis. In the later course she suffered additionally from cervical lymphadenitis and pharyngitis during the days with fever. Oral prednisolone caused immediate fever reduction within 2 h but shortened the asymptomatic interval (2–3 weeks). The diagnosis of PFAPA syndrome was made 28 months after the commencement of primary symptoms. A therapy with Cimetidine (20 mg per kg per day) was started, which led to a complete cessation of fever episodes. The medication was tapered after one and a half year, which led to recurrence of febrile episodes. Prior to this auto-inflammatory disease the girl was presented to our hospital with signs of mild growth retardation. Her parents were Caucasian, overall healthy and not related. The girl had during her first two years of life a proportional growth rate shortly below the third percentile. A hypothyreosis and a metabolic disorder were excluded. Further mild stigmata like an ocular hypertelorism and short bones as well as dysplastic brachymesophalangy V of the hand led to karyotyping. In addition, four further children who fulfilled the typical clinical criteria for PFAPA syndrome were selected for molecular genetic testing. All except one had recurrent febrile episodes from an early age (p5 years). None of the patients had signs of an infection during the episodes but presented at least one or two of the following symptoms: aphthous stomatitis, cervical lymphadenitis or pharyngitis. The children were additionally completely asymptomatic between the febrile episodes. Cyclic neutropenia or malignancies (leukemia or neuroblastoma) were excluded. Three patients had a positive family history (one of the parents) for febrile episodes during early childhood. In one patient (Table 1 ID: 2) we found a heterozygous mutation in the MEFV gene (c.2080A4G[p.Met694Val]). This patient presented, except for recurrent fever, no further signs of FMF (rash, arthritis, serositis). All children were initially treated with Cimetidine; second-line therapy was Colchicine. One child who did not respond to the medical therapy received a tonsillectomy. The clinical characteristics of all patients with PFAPA syndrome are summarised in Table 1. All patients agreed to the study and the parents gave written informed consent. The procedures of the study were approved by the Ethics Commission of the Medical Faculty of the Christian-Albrechts-University in Kiel (AZ D401/08 and amendment 08 December 2011).
Karyotying and molecular karyotyping Karyotyping was performed on phytohemagglutinin-stimulated cultures of peripheral blood using G-banding analyses according to standard techniques. Metaphases were captured using the IKAROS software (MetaSystems, Altlussheim, Germany). Karyotypes were described according to ISCN2013. DNA was extracted from the peripheral blood of all patients using standard methods. Array-comparative genomic hybridisation using the Human Genome CGH microarray 244A platform (Agilent Technologies, Santa Clara, CA, USA) was performed according to & 2014 Macmillan Publishers Limited
the manufacturer’s instructions. Pooled peripheral blood DNA from 10 healthy donors with a female karyotype served as hybridisation control. Analysis was performed using the CGH Analytics Software (Agilent Technologies). Regions that were determined as significantly deviated by the software and that included more than 10 single oligonucleotides were regarded as aberrant, thus reaching a functional resolution of around 0.1 Mb. The differentiation between pathologic and benign copy number variations was performed by using a database integrated into the software.
Mate-pair structural rearrangement analysis Mate-pair (MP, or long insert size) DNA library preparation was carried out using the v2 protocol from Illumina Gmbh (Munich, Germany). In brief, 10 mg of genomic DNA was fragmented using a Covaris S2 instrument and red miniTUBE (LGC Genomics Ltd., Hoddesdon, UK) with the manufacturer’s recommended settings to B5 kb, followed by size selection through agarose gel excision. Deep sequencing was carried out with an Illumina GAIIx (2 36 bp paired-end mode) instrument to reach an average physical coverage of at least 30 . Mate-pair reads were aligned to the hg19 assembly of the human reference genome using the Illuminaprovided alignment software (ELAND, version 2). Rearrangements based on mate-pair data were identified using the DELLY program.28 Structural rearrangement calls that were also inferred from 1000 genomes project19 genome data or germline data of additional whole-genome sequencing samples20 were removed to exclude polymorphic germline structural variants as well as rearrangement calls caused by mapping artefacts. Furthermore, rearrangement calls with less than four supporting read pairs as well as supporting pairs with an average mapping quality o20 were excluded. To assess whether the genomic regions between the translocation breakpoints on chromosome 10 and 17 are lost as a consequence of the rearrangement, the number of reads that map within these regions was counted and compared with neighbouring regions of identical size (5577 bp on chromosome 10; 5879 bp on chromosome 17).
PCR and Sanger sequencing For PCR amplification of the translocation breakpoint a primer walking strategy was applied. Primers leading to amplification of the breakpoint and respective sequencing primers are listed in Supplementary Table 1. For sequencing of the SPAG7 gene, primers flanking all exon–intron junctions of the gene were designed as outlined in Supplementary Table 1. PCR was performed according to standard protocols. Direct sequencing of the PCR products using BigDye Terminator technology (BigDye Terminator v1.1 Cycle Sequencing Kit, Applied Biosystems, Foster City, CA, USA) was performed using the same primers as for PCR if not otherwise indicated (Supplementary Table 1) on an ABI 3100 Genetic Analyzer (Applied Biosystems). Alternatively, sequencing at GATC Biotech AG (Constance, Germany) was performed. Genes and Immunity (2014) 190 – 194
SPAG7 in PFAPA syndrome S Bens et al
194 CONFLICT OF INTEREST The authors declare no conflict of interest.
ACKNOWLEDGEMENTS We thank the technical staff of the cytogenetic and molecular genetic laboratories of the Institute of Human Genetics for expert assistance, as well as all clinicians involved in the care of the patients for support and all individuals for their participation. JOK was supported by an Emmy Noether Fellowship (KO 4037/1-1) from the German Research Foundation and RW is recipient of a stipend from the KinderKrebsInitiative Buchholz/Holm-Seppensen. RS is a member of the Excellence Cluster ‘Inflammation at Interfaces’.
REFERENCES 1 Marshall GS, Edwards KM, Butler J, Lawton AR. Syndrome of periodic fever, pharyngitis, and aphthous stomatitis. J Pediatr 1987; 110: 43–46. 2 Padeh S, Brezniak N, Zemer D, Pras E, Livneh A, Langevitz P et al. Periodic fever, aphtous stomatitis, pharyngitis, and adenopathy syndrome: clinical characteristics and outcome. J Pediatr 1999; 135: 98–101. 3 Tasher D, Somekh E, Dalal I. PFAPA syndrome: new clinical aspects disclosed. Arch Dis Child 2006; 91: 981–984. 4 Vigo G, Zulian F. Periodic fevers with aphthous stomatitis, and adenitis (PFAPA). Autoimmun Rev 2012; 12: 52–55. 5 Feder HM, Salazar JC. A clinical review of 105 patients with PFAPA (a periodic fever syndrome). Acta Paediatr 2010; 99: 178–184. 6 Marcuzzi A, Piscianz E, Kleiner G, Tommasini A, Severini GM, Monasta L et al. Clinical genetic testing of periodic fever syndromes. Biomed Res Int 2013; 147: 155–174. 7 Savic S, Dickie LJ, Battelino M, McDermott MF. Familial Mediterranean fever and related periodic fever syndromes/autoinflammatory diseases. Curr Opin Rheumatol 2012; 24: 103–112. 8 Akelma AZ, Cizmeci MN, Kanburoglu MK, Mete E, Bozkaya D, Tufan N et al. Is PFAPA syndrome really a sporadic disorder or is it genetic? Med Hypotheses 2013; 81: 279–281. 9 Cochard M, Clet J, Le L, Pillet P, Onrubia X, Gue´ron T et al. PFAPA syndrome is not a sporadic disease. Rheumatology (Oxford) 2010; 49: 1984–1987. 10 Kolly L, Busso N, von Scheven-Gete A, Bagnoud N, Moix I, Holzinger D et al. Periodic fever, aphthous stomatitis, pharyngitis, cervical adenitis syndrome is linked to dysregulated monocyte IL-1b production. J Allergy Clin Immunol 2013; 131: 1635–1643. 11 Berkun Y, Levy R, Hurwitz A, Meir-Harel M, Lidar M, Livneh A et al. The familial Mediterranean fever gene as a modifier of periodic fever, aphthous stomatitis, pharyngitis, and adenopathy syndrome. Semin Arthritis Rheum 2011; 40: 467–472. 12 Stojanov S, Lapidus S, Chitkara P, Feder H, Salazar JC, Fleisher TA et al. Periodic fever, aphthous stomatitis, pharyngitis, and adenitis (PFAPA) is a disorder of innate immunity and Th1 activation responsive to IL-1 blockade. Proc Natl Acad Sci USA 2011; 108: 7148–7153.
13 Caorsi R, Pelagatti MA, Federici S, Finetti M, Martini A, Gattorno M. Periodic fever, aphthous stomatitis, pharyngitis and adenitis syndrome. Curr Opin Rheumatol 2010; 22: 579–584. 14 Peridis S, Pilgrim G, Koudoumnakis E, Athanasopoulos I, Houlakis M, Parpounas K. PFAPA syndrome in children: a meta-analysis on surgical versus medical treatment. Int J Pediatr Otorhinolaryngol 2010; 74: 1203–1208. 15 Higgins AW, Alkuraya FS, Bosco AF, Brown KK, Bruns GA, Donovan DJ et al. Characterization of apparently balanced chromosomal rearrangements from the developmental genome anatomy project. Am J Hum Genet 2008; 82: 712–722. 16 Cacciagli P, Haddad MR, Mignon-Ravix C, El-Waly B, Moncla A, Missirian C et al. Disruption of the ATP8A2 gene in a patient with a t(10;13) de novo balanced translocation and a severe neurological phenotype. Eur J Hum Genet 2010; 18: 1360–1363. 17 Warburton D. De novo balanced chromosome rearrangements and extra marker chromosomes identified at prenatal diagnosis: clinical significance and distribution of breakpoints. Am J Hum Genet 1991; 49: 995–1013. 18 Korbel JO, Urban AE, Affourtit JP, Godwin B, Grubert F, Simons JF et al. Paired-end mapping reveals extensive structural variation in the human genome. Science 2007; 318: 420–426. 19 1000 Genomes Project Consortium, Abecasis GR, Auton A, Brooks LD, DePristo MA, Durbin RM et al. An integrated map of genetic variation from 1,092 human genomes. Nature 2012; 491: 56–65. 20 Jones DT, Ja¨ger N, Kool M, Zichner T, Hutter B, Sultan M et al. Dissecting the genomic complexity underlying medulloblastoma. Nature 2012; 488: 100–105. 21 Mao M, Fu G, Wu JS, Zhang QH, Zhou J, Kan LX et al. Identification of genes expressed in human CD34( þ ) hematopoietic stem/progenitor cells by expressed sequence tags and efficient full-length cDNA cloning. Proc Natl Acad Sci USA 1998; 95: 8175–8180. 22 Beaton S, Cleary A, ten Have J, Bradley MP. Cloning and characterization of a fox sperm protein FSA-1. Reprod Fertil Dev 1994; 6: 761–770. 23 Grishin NV. The R3H motif: a domain that binds single-stranded nucleic acids. Trends Biochem Sci 1998; 23: 329–330. 24 Kerr JR. Pathogenesis of parvovirus B19 infection: host gene variability, and possible means and effects of virus persistence. J Vet Med B Infect Dis Vet Public Health 2005; 52: 335–339. 25 Lehner B, Semple JI, Brown SE, Counsell D, Campbell RD, Sanderson CM. Analysis of a high-throughput yeast two-hybrid system and its use to predict the function of intracellular proteins encoded within the human MHC class III region. Genomics 2004; 83: 153–167. 26 Hsieh YY, Lin YJ, Chang CC, Chen CY, Hsu CM, Wang YK et al. Human lymphocyte antigen B-associated transcript 2,3, and 5 polymorphisms and haplotypes are associated with susceptibility of Kawasaki disease and coronary artery aneurysm. J Clin Lab Anal 2010; 24: 262–268. 27 Thomas KT, Feder HMJ, Lawton AR, Edwards KM. Periodic fever syndrome in children. J Pediatr 1999; 135: 15–21. 28 Rausch T, Zichner T, Schlattl A, Stu¨tz AM, Benes V, Korbel JO. DELLY: structural variant discovery by integrated paired-end and split-read analysis. Bioinformatics 2012; 28: i333–i339.
Supplementary Information accompanies this paper on Genes and Immunity website (http://www.nature.com/gene)
Genes and Immunity (2014) 190 – 194
& 2014 Macmillan Publishers Limited
