Hereditary Sensory Neuropathy Type I: Haplotype Analysis Shows Founders in Southern England and Europe
2001; Elsevier BV; Volume: 69; Issue: 3 Linguagem: Inglês
10.1086/323252
ISSN1537-6605
AutoresGarth A. Nicholson, Jennifer L. Dawkins, Ian P. Blair, Michaela Auer‐Grumbach, Sonal Brahmbhatt, D J Hulme,
Tópico(s)Cellular transport and secretion
ResumoHereditary sensory neuropathy type I (HSN1) is the most common dominantly inherited degenerative disorder of sensory neurons. The gene mutation was mapped to chromosome 9 in a large Australian family, descended from an ancestor from southern England who was a convict. Dawkins et al. recently reported gene mutations in the SPTLC1 gene, in this and other families. The first description of hereditary sensory neuropathy, by Hicks, was in a family from London and Exeter. To determine if the families in the present study that have SPTLC1 mutations are related to English families with HSN1 and, possibly, to the family studied by Hicks, we performed haplotype analysis of four Australian families of English extraction, four English families, and one Austrian family. Three Australian families of English extraction and three English families (two of whom have been described elsewhere) had the 399T→G SPTLC1 mutation, the same chromosome 9 haplotype, and the same phenotype. The Australian and English families may therefore have a common founder who, on the basis of historical information, has been determined to have lived in southern England prior to 1800. The sensorimotor neuropathy phenotype caused by the 399T→G SPTLC1 mutation is the same as that reported by Campbell and Hoffman and, possibly, the same as that originally described by Hicks. Hereditary sensory neuropathy type I (HSN1) is the most common dominantly inherited degenerative disorder of sensory neurons. The gene mutation was mapped to chromosome 9 in a large Australian family, descended from an ancestor from southern England who was a convict. Dawkins et al. recently reported gene mutations in the SPTLC1 gene, in this and other families. The first description of hereditary sensory neuropathy, by Hicks, was in a family from London and Exeter. To determine if the families in the present study that have SPTLC1 mutations are related to English families with HSN1 and, possibly, to the family studied by Hicks, we performed haplotype analysis of four Australian families of English extraction, four English families, and one Austrian family. Three Australian families of English extraction and three English families (two of whom have been described elsewhere) had the 399T→G SPTLC1 mutation, the same chromosome 9 haplotype, and the same phenotype. The Australian and English families may therefore have a common founder who, on the basis of historical information, has been determined to have lived in southern England prior to 1800. The sensorimotor neuropathy phenotype caused by the 399T→G SPTLC1 mutation is the same as that reported by Campbell and Hoffman and, possibly, the same as that originally described by Hicks. Hereditary sensory neuropathy type I (HSN1 [MIM 162400]) is a dominantly inherited sensorimotor axonal neuropathy accompanied by variable sensorineural deafness (for review, see Thomas Thomas, 1993Thomas PK Hereditary sensory neuropathies.Brain Pathol. 1993; 3: 157-163Crossref PubMed Scopus (38) Google Scholar). The presence of severe sensory loss leads to painless injuries, chronic skin ulcers, and distal amputations, distinguishing the disorder from other dominantly inherited sensorimotor neuropathies—such as Charcot-Marie-Tooth (CMT) type 2 syndromes—although, occasionally, patients with CMT type 1 or 2 develop skin ulcers. In past literature, there have been many terms for this disease, including "trophoneurosis," "lumbosacral syringomyelia," and "ulcero-mutilating acropathy." This literature has been reviewed by Dyck et al. (Dyck et al., 1993Dyck PJ Chance P Lebo R Carney JA Neuronal atrophy and degeneration predominantly affecting peripheral sensory and autonomic neurones.in: Dyck PJ Thomas PK Peripheral neuropathy. Vol 2. WB Saunders, Philadelphia1993: 1094-1136Google Scholar), who classified the disease as hereditary sensory and autonomic neuropathy (Dyck et al. Dyck et al., 1993Dyck PJ Chance P Lebo R Carney JA Neuronal atrophy and degeneration predominantly affecting peripheral sensory and autonomic neurones.in: Dyck PJ Thomas PK Peripheral neuropathy. Vol 2. WB Saunders, Philadelphia1993: 1094-1136Google Scholar); however, frank autonomic symptoms and signs are rare in the families in the present study, but motor signs are present in all but early cases. An English family with a phenotype similar to that of the large Australian family described elsewhere and herein (family 1 in this report) was reported, by Hicks (Hicks, 1922Hicks EP Hereditary perforating ulcer of the foot.Lancet. 1922; 1: 319-321Abstract Scopus (60) Google Scholar), as having hereditary perforating ulcer of the foot. The same family was later reported, by Denny-Brown (Denny-Brown, 1951Denny-Brown D Hereditary sensory radicular neuropathy.J Neurol Neurosurg Psychiatry. 1951; 14: 237-252Crossref PubMed Scopus (150) Google Scholar), as having hereditary sensory radicular neuropathy. One year later, the same family was reported as having peroneal muscular atrophy with severe sensory changes (England and Denny-Brown England and Denny-Brown, 1952England AC Denny-Brown D Severe sensory changes, and trophic disorder, in peroneal muscular atrophy (Charcot-Marie-Tooth type).Arch Neurol Psychiatry. 1952; 67: 1-22Crossref Scopus (32) Google Scholar), a description that recognized motor involvement. These reports underlie the overlap between HSN1 and the other dominantly inherited hereditary sensorimotor neuropathies (e.g., CMT neuropathies). This problem was recognized by Dyck et al. (Dyck et al., 1993Dyck PJ Chance P Lebo R Carney JA Neuronal atrophy and degeneration predominantly affecting peripheral sensory and autonomic neurones.in: Dyck PJ Thomas PK Peripheral neuropathy. Vol 2. WB Saunders, Philadelphia1993: 1094-1136Google Scholar), who felt that an improved classification would depend on identification of the gene abnormality. Foot ulcers are common in an axonal form of CMT type 2 neuropathy linked to chromosome 3 (Kwon et al. Kwon et al., 1995Kwon JM Elliott JL Yee WC Ivanovich J Scavarda NJ Moolsintong PJ Goodfellow PJ Assignment of a second Charcot-Marie-Tooth type II locus to chromosome 3q.Am J Hum Genet. 1995; 57: 853-858PubMed Google Scholar; De Jonghe et al. De Jonghe et al., 1997De Jonghe P Timmerman V FitzPatrick D Spoelders P Martin JJ Van Broeckhoven C Mutilating neuropathic ulcerations in a chromosome 3q13-q22 linked Charcot-Marie-Tooth disease type 2B family.J Neurol Neurosurg Psychiatry. 1997; 62: 570-573Crossref PubMed Scopus (63) Google Scholar). This form of CMT could also be classified as hereditary sensory neuropathy (Vance et al. Vance et al., 1996Vance JM Speer MC Stajich JM West S Wolpert C Gaskell P Lennon F Tim RM Rozear M Othmane KB Pericak-Vance MA Misclassification and linkage of hereditary sensory and autonomic neuropathy type 1 as Charcot-Marie-Tooth disease, type 2B.Am J Hum Genet. 1996; 59: 258-262PubMed Google Scholar). We have used the term "HSN1" to describe this form of dominantly inherited sensorimotor axonal neuropathy with variable sensorineural deafness (Thomas Thomas, 1993Thomas PK Hereditary sensory neuropathies.Brain Pathol. 1993; 3: 157-163Crossref PubMed Scopus (38) Google Scholar). We previously mapped the HSN1 locus to chromosome 9 (Nicholson et al. Nicholson et al., 1996Nicholson GA Dawkins JL Blair IP Kennerson ML Gordon MJ Cherryson AK Nash J Bananis T The gene for hereditary sensory neuropathy type I (HSN-I) maps to chromosome 9q22.1-q22.3.Nat Genet. 1996; 13: 101-104Crossref PubMed Scopus (108) Google Scholar), in the large Australian family (family 1) with clinical features similar to those in the London/Exeter family studied by Hicks (Hicks, 1922Hicks EP Hereditary perforating ulcer of the foot.Lancet. 1922; 1: 319-321Abstract Scopus (60) Google Scholar) and Denny-Brown (Denny-Brown, 1951Denny-Brown D Hereditary sensory radicular neuropathy.J Neurol Neurosurg Psychiatry. 1951; 14: 237-252Crossref PubMed Scopus (150) Google Scholar), and recently found mutations in the SPTLC1 gene (Dawkins et al. Dawkins et al., 2001Dawkins JL Hulme DJ Brahmbhatt SB Auer-Grumbach M Nicholson GA Mutations in SPTLC1, encoding serine palmitoyltransferase, long chain base subunit-1, cause hereditary sensory neuropathy type I.Nat Genet. 2001; 27: 309-312Crossref PubMed Scopus (341) Google Scholar), both in family 1 and in 10 other families with HSN1. The Australian family was originally described as having familial lumbosacral syringomyelia (Jackson Jackson, 1949Jackson M Familial lumbosacral syringomyelia and the significance of developmental errors of the spinal cord and column.Med J Aust. 1949; 1: 434-439Google Scholar), and, later, the same family was reported as having hereditary sensory radicular neuropathy (Wallace Wallace, 1969Wallace DC A study of an hereditary neuropathy. University of Sydney, Sydney1969Google Scholar, Wallace, 1970Wallace DC Hereditary sensory radicular neuropathy. Archdall Medical Monograph 8. Australasian Medical, Sydney1970Google Scholar). HSN1 is genetically heterogeneous, with at least three loci: chromosome 9 (Nicholson et al. Nicholson et al., 1996Nicholson GA Dawkins JL Blair IP Kennerson ML Gordon MJ Cherryson AK Nash J Bananis T The gene for hereditary sensory neuropathy type I (HSN-I) maps to chromosome 9q22.1-q22.3.Nat Genet. 1996; 13: 101-104Crossref PubMed Scopus (108) Google Scholar; Bejaoui et al. Bejaoui et al., 1999Bejaoui K McKenna-Yasek D Hosler BA Burns-Deater E Deater LM O'Neill G Haines JL Brown Jr, RH Confirmation of linkage of type 1 hereditary sensory neuropathy to human chromosome 9q22.Neurology. 1999; 52: 510-515Crossref PubMed Google Scholar); a chromosome 3 locus (Kwon et al. Kwon et al., 1995Kwon JM Elliott JL Yee WC Ivanovich J Scavarda NJ Moolsintong PJ Goodfellow PJ Assignment of a second Charcot-Marie-Tooth type II locus to chromosome 3q.Am J Hum Genet. 1995; 57: 853-858PubMed Google Scholar), for a form of CMT type 2 suggested to be similar to HSN1 (Vance et al. Vance et al., 1996Vance JM Speer MC Stajich JM West S Wolpert C Gaskell P Lennon F Tim RM Rozear M Othmane KB Pericak-Vance MA Misclassification and linkage of hereditary sensory and autonomic neuropathy type 1 as Charcot-Marie-Tooth disease, type 2B.Am J Hum Genet. 1996; 59: 258-262PubMed Google Scholar); and at least one other locus, for families without linkage to either of these two loci (Auer-Grumbach et al. Auer-Grumbach et al., 2000Auer-Grumbach M Wagner K Timmerman V De Jonghe P Hartung HP Ulcero-mutilating neuropathy in an Austrian kinship without linkage to hereditary motor and sensory neuropathy IIB and hereditary sensory neuropathy I loci.Neurology. 2000; 54: 45-52Crossref PubMed Google Scholar). To determine whether the Australian and English families have a common founder and whether the phenotype described by us is likely to be the same as that originally described by Hicks (Hicks, 1922Hicks EP Hereditary perforating ulcer of the foot.Lancet. 1922; 1: 319-321Abstract Scopus (60) Google Scholar), we generated additional microsatellite markers and constructed a disease haplotype for each family. Multigenerational families with a diagnosis of HSN1 and with mutations in SPTLC1 were selected. The clinical diagnosis of HSN1 was based on (1) a history of either chronic skin ulcers or painless injuries, (2) gross sensory loss in all modalities, and (3) electrophysiological evidence of a chronic axonal neuropathy. Severe, brief shooting pains were characteristic of HSN1 but were not a selection criterion. Sensorineural deafness was present in some members of some families. Australian families 1, 2, 4, and 6 have been described elsewhere (but identified as families 1, 3, 2, and 4, respectively; Nicholson et al. Nicholson et al., 1996Nicholson GA Dawkins JL Blair IP Kennerson ML Gordon MJ Cherryson AK Nash J Bananis T The gene for hereditary sensory neuropathy type I (HSN-I) maps to chromosome 9q22.1-q22.3.Nat Genet. 1996; 13: 101-104Crossref PubMed Scopus (108) Google Scholar). One of the English families (family 24) has previously been characterized, by Campbell and Hoffman (Campbell and Hoffman Campbell and Hoffman, 1964Campbell AMG Hoffman HL Sensory radicular neuropathy associated with muscle wasting in two cases.Brain. 1964; 87: 67-74Crossref PubMed Scopus (14) Google Scholar), as family X. Fourteen microsatellite markers on chromosome 9 were analyzed in each of the nine families with HSN1, as well as in the control individuals. These markers included three new polymorphic microsatellites (71J15, 20L24, and 20L24TA) that we isolated from cosmids within the HSN1 critical region. All 14 microsatellite markers were localized to our 8-Mb yeast-artificial-chromosome (YAC)–based physical map of the HSN1 critical interval (Blair et al. Blair et al., 1998Blair IP Hulme D Dawkins JL Nicholson GA A YAC-based transcript map of human chromosome 9q22.1-q22.3 encompassing the loci for hereditary sensory neuropathy type I and multiple self-healing squamous epithelioma.Genomics. 1998; 51: 277-281Crossref PubMed Scopus (19) Google Scholar). Figure 1 shows the locations of the 3 new markers and of 13 markers described elsewhere, in relation to the HSN1 critical interval and our YAC-based physical map. The order of markers within individual expressed-sequence-tag (EST) bins was determined, when possible, by use of information from published linkage maps (Genome Database), as well as by use of sequence data and bacterial-artificial-chromosome (BAC) contig maps of the region that, as part of their project to sequence chromosome 9, were generated by the Sanger Centre. For each of the 14 markers, alleles associated with HSN1 were identified in each family with HSN1, and a disease haplotype was constructed. Haplotypes spanning these markers were also determined in 50 controls (100 chromosomes), consisting of those married-in spouses and individuals of known English extraction for whom samples were available from three generations. Statistical analysis, to compare the frequency of disease-carrying haplotypes to that of controls, was performed by the χ2 test, with 1 df and the Yates continuity correction for a 2×2 contingency table. Only two families with HSN1 were large enough to show definite linkage (LOD score >3) to chromosome 9. Maximum LOD scores for each family are shown in table 1. Linkage to chromosome 9 was not excluded in any family. Most families were too small to yield a significant LOD score, but, in each of these small families, we were able to identify a disease haplotype that was shared by all affected family members. No recombination events within the HSN1 critical interval were observed in any individuals in these families.Table 1Haplotype Analysis of Families with HSN1Allele at MarkeraFor details of markers and assignment, see text. The common haplotype, in families 1, 2, 6, 22, 23, and 24, and a group of common alleles, in families 4 and 16, are indicated by the larger and smaller boxed regions, respectively. D9S1781 and 71J15 flank the region containing the SPTLC1 gene.FamilyD9S1836D9S1796D9S255D9S1815D9S1841D9S178171J1520L2420L24TAD9S151D9S12D9S12IID9S1803D9S197Maximum LOD ScoreOriginbA/E = Australian, of English extraction; W = Wiltshire, in southern England; L = London; Au = Austria.Mutation1253336741572367.45A/E399T→G2253336741734251.18A/E399T→G625333674157236.76A/E399T→G22253336741572361.12W399T→G23253336741572364.30W399T→G2425333674157236.98W399T→G4343545419572521.14A/E341T→A1634/5/7?cDisease allele unknown, because marker was uninformative in the family.331541987235.58L399T→G201/2?cDisease allele unknown, because marker was uninformative in the family.13351743534121.79Au398G→Aa For details of markers and assignment, see text. The common haplotype, in families 1, 2, 6, 22, 23, and 24, and a group of common alleles, in families 4 and 16, are indicated by the larger and smaller boxed regions, respectively. D9S1781 and 71J15 flank the region containing the SPTLC1 gene.b A/E = Australian, of English extraction; W = Wiltshire, in southern England; L = London; Au = Austria.c Disease allele unknown, because marker was uninformative in the family. Open table in a new tab Table 1 shows HSN1 chromosome haplotypes. An extended haplotype spanning the nine markers—D9S1836, D9S1796, D9S255, D9S1815, D9S1841, D9S1781, 71J15, 20L24, and 20L24TA—was present in 6 of the 9 disease chromosomes studied but in 0 of the 100 control chromosomes studied (χ2=58.32; P<.0001). These results provide strong evidence for a common founder for these families. A group of common alleles was found, with markers 71J15, 20L24, and 20L24TA, in 2 of the 9 disease chromosomes (in families 4 and 16), compared to 28 of the 100 control chromosomes, but this smaller haplotype was not statistically significant (χ2=0.00; P=.986). Families 4 and 16 exhibit different SPTLC1 mutations, providing further evidence against a common founder for these two families. Five of the six families with HSN1 who share the 9-marker haplotype, from D9S1836 to 20L24TA, have a larger haplotype—spanning 14 markers, from D9S1836 to D9S197 (table 1); family 2 shares the common 9-marker haplotype, to marker 20L24TA, but the haplotype diverges at the next marker, D9S151 (see table 1), indicating a historical recombination between D9S151 and the gene mutation. In this study, we have identified, in the largest of the families that we studied (family 1), an HSN1 haplotype that is common to families 2, 6, 22, 23, and 24. Given the strong statistical evidence (P<.0001), it is highly unlikely that the common haplotype occurs by chance. No control families of English extraction had the same haplotype, indicating that the common haplotype was present in <1/50 families. These results suggest a common founder for the three Australian and the three Wiltshire families with HSN1. One of the Wiltshire families with the common haplotype had come from London and therefore may be related to the London/Exeter family studied by Hicks (Hicks, 1922Hicks EP Hereditary perforating ulcer of the foot.Lancet. 1922; 1: 319-321Abstract Scopus (60) Google Scholar), but no records for the London/Exeter family can be found. The clinical phenotype in the families with the common chromosome 9 haplotype is similar to that in the family studied by Hicks (Hicks, 1922Hicks EP Hereditary perforating ulcer of the foot.Lancet. 1922; 1: 319-321Abstract Scopus (60) Google Scholar), with frequent but not invariable deafness, severe shooting or lancinating pains in some affected individuals, and late-teenage– or adult-onset CMT type 2. This lancinating pain/deafness HSN1 phenotype was recognized by Dyck et al. (Dyck et al., 1993Dyck PJ Chance P Lebo R Carney JA Neuronal atrophy and degeneration predominantly affecting peripheral sensory and autonomic neurones.in: Dyck PJ Thomas PK Peripheral neuropathy. Vol 2. WB Saunders, Philadelphia1993: 1094-1136Google Scholar). Family 20, an Austrian family, has this phenotype as well as a different SPTLC1 mutation with a different haplotype, suggesting a separate founder. Shooting pains and deafness were not reported in either the families with chromosome 3 linkage (De Jonghe et al. De Jonghe et al., 1997De Jonghe P Timmerman V FitzPatrick D Spoelders P Martin JJ Van Broeckhoven C Mutilating neuropathic ulcerations in a chromosome 3q13-q22 linked Charcot-Marie-Tooth disease type 2B family.J Neurol Neurosurg Psychiatry. 1997; 62: 570-573Crossref PubMed Scopus (63) Google Scholar; Elliott et al. Elliott et al., 1997Elliott JL Kwon JM Goodfellow PJ Yee WC Hereditary motor and sensory neuropathy IIB: clinical and electrodiagnostic characteristics.Neurology. 1997; 48: 23-28Crossref PubMed Scopus (34) Google Scholar) or the family in which loci of chromosomes 3 and 9 were excluded (Auer-Grumbach et al. Auer-Grumbach et al., 2000Auer-Grumbach M Wagner K Timmerman V De Jonghe P Hartung HP Ulcero-mutilating neuropathy in an Austrian kinship without linkage to hereditary motor and sensory neuropathy IIB and hereditary sensory neuropathy I loci.Neurology. 2000; 54: 45-52Crossref PubMed Google Scholar). These associated features may therefore help to distinguish between families with HSN1 and families with other forms of hereditary sensory neuropathy and CMT neuropathy. The haplotype information suggests that a number of Australian and English families are descended from a common southern-English founder; this is supported by our finding the same phenotype and mutation (399T→G) in such families. Historical information shows that the founder lived in southern England prior to 1800, the date of deportation, from Wiltshire, of the convict ancestor of family 1 (Wallace Wallace, 1969Wallace DC A study of an hereditary neuropathy. University of Sydney, Sydney1969Google Scholar). The neuropathy phenotype caused by the 399T→G SPTLC1 mutation is the same as that reported by Campbell and Hoffman (Campbell and Hoffman, 1964Campbell AMG Hoffman HL Sensory radicular neuropathy associated with muscle wasting in two cases.Brain. 1964; 87: 67-74Crossref PubMed Scopus (14) Google Scholar) and, possibly, the same as that originally described by Hicks (Hicks, 1922Hicks EP Hereditary perforating ulcer of the foot.Lancet. 1922; 1: 319-321Abstract Scopus (60) Google Scholar). It is interesting that the family with the historical recombination (family 2) migrated to Australia from Trowbridge, a town in Wiltshire, near where three of the English families with HSN1 live. We wish to thank Prof. P. K. Thomas (Royal Free Hospital, London) and Dr. M. Campbell (Department of Neurology, Fenchay Hospital, Bristol), for providing access to families in their care. Dr. D. Ellison (Melksham) kindly assisted in the relocation of a family (family 24) originally described in his University of Edinburgh M.D. thesis. Ms. M. Williams (Southmead Hospital, Bristol) kindly assisted with blood collections and DNA preparations. Dr. N. Wood (Institute of Neurology, London) contributed DNA from the London family (family 16). This work was made possible by the willing assistance of many Australian and English family members. Mr. R. Thorburn (Australian Genealogical Education Center) found the links between family 1 and the convict from Wiltshire. Financial support was provided through grants from the National Health and Medical Research Council (Australia) and from the Muscular Dystrophy Association (U.S.A.).
Referência(s)