Editor—Hereditary spastic paraplegia, spastic paraplegia, or familial spastic paraplegia (HSP, SPG, or FSP) are a heterogeneous group of syndromes characterised by degeneration of corticospinal tracts. Currently, two loci for the X linked recessive type are well established, at Xq28 (SPG1, MIM 312900)1 and at Xq22 (SPG2, MIM 312920).2 3 Six loci for the autosomal dominant type have been reported, at 14q12-q23 (SPG3, MIM 182600),4 at 2p21-p24 (SPG4, MIM 182601),5 6 at 15q11.1 (SPG6, MIM 600363),7 at 8q23-q24 (SPG8, MIM 603563),8at 10q23.3-q24.1 (SPG9, MIM 601162),9 and at 12q13 (SPG10, MIM 604187).10 Two loci for the autosomal recessive type were mapped to 8p12-q13 (SPG5A, MIM 270800)11 and to 16q24.3 (SPG7, MIM 602783),12 13 respectively.

HSPs are clinically subdivided into pure and complicated forms. The autosomal dominant form can be complicated by dementia and epilepsy.14 The X linked form can be more severe if combined with ataxia, absent extensor pollicis longus, and involvement of cerebral cortex and optic nerves (SPG1). The complicated autosomal recessive type is rare and is described as microcephaly with spastic quadriplegia (MIM 251280).15-19 The fact that familial cases were reported, where no linkage to any of the previously mentioned loci could be found, suggests that additional gene loci for hereditary spastic paraplegia exist.20 21 The genetic heterogeneity supports the concept of a multitude of different genes responsible for spastic paraplegia. Since, however, a marked clinical similarity is found within HSP families with positive linkage to each of the reported loci, it was proposed that gene products from HSP loci may participate in a common biochemical cascade which, if disturbed, results in axonal degeneration that is most pronounced at the ends of the longest CNS axons.22

Complex chromosomal rearrangements (CCRs) are defined as any structural rearrangement involving more than two chromosome breaks with exchange of segments between at least two chromosomes.23Rearrangements with up to seven derivative chromosomes24and 10 breakpoints25 have been reported. Batanianet al 26 reviewed 100 published CCR cases. The detection and interpretation of CCRs is most efficiently achieved by a combination of classical cytogenetic methods, including high resolution banding techniques and FISH using both whole chromosome painting and band specific probes.27 Recently, multicolour FISH assays have been developed which allow the identification of cryptic translocations, marker chromosomes, and delineation of complex chromosomal aberrations in a single experiment. Based on cross species colour banding technology, RxFISH combines traditional banding capability with colour classification and allows also the detection of intrachromosomal rearrangements, such as duplications, deletions, insertions, and inversions.28

The strategy for a detailed characterisation of the chromosomal breakpoints of the CCR involving chromosomes 2, 3, and 10 was based on RxFISH and detailed mapping of single site specific YAC clones selected from the CEPH “MegaYAC” library. We analysed the chromosomal breakpoints in order to narrow down the region for putative candidate genes involved in the phenotype of the patient.

The proband, a 2½ year old boy with severe psychomotor retardation, is the first child of non-consanguineous parents. At the time of birth, the mother was 28 and the father 31 years old. He was born at 41 weeks of gestation following an uneventful pregnancy. Birth weight was 2480 g and Apgar scores were 8/9/10. During the neonatal period general muscular hypotonia combined with hyperreflexia of the lower limbs and the presence of bilateral Babinski signs was noticed. In the first months of life, two episodes of complicated febrile convulsions (EEG normal) occurred. At 12 months, psychomotor development was classified as markedly delayed on a clinical examination. At this time, microcephaly with a head circumference of 44 cm (−2.4 SD)29 with normal cranial MRI was recorded. Additional clinical abnormalities, such as epicanthus, a broad nasal root, severe myopia, a strabismus convergens alternans, slightly dysplastic ears with a preauricular fistula on the right side, a sacralporus, and hypopigmented skin were noticed (fig 1). Cardiac ultrasound and ECG were normal. All standard laboratory parameters were within the normal range as were results of amino acid analysis and of very long chain fatty acid screening. At the age of 16 months, spasticity with lower limb predominance became evident and developed to quadriplegia with preserved but considerably reduced motor function of the left upper extremity. At 26 months, his weight is 12.2 kg (−0.42 SD) and head circumference 47 cm (−1.8 SD). The Babinski sign is present bilaterally. He can sit independently but is still unable to walk. He has good eye contact but his social interactions are poor. He does not show any speech development so far, but apparently understands simple commands. Despite extensive testing, no exogenous causes for the psychomotor retardation and the quadriplegia could be found.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

The proband aged 12 months.

Chromosomes of the proband and his parents were investigated using BrdU synchronised cultures in order to obtain preparations suitable for high resolution analysis.30 A lymphoblast cell line of the patient was established from whole blood according to standard procedures.

YAC clones and information on positive STS hits were taken from Whitehead YAC contigs.31 DNA from CEPH “MegaYAC” clones were isolated by pulsed field gel electrophoresis and amplified by using degenerate oligonucleotide primed (DOP)-PCR to generate probes for FISH analysis.32 RxFISH (Applied Imaging International Ltd), and FISH with a subtelomeric probe from 2q HBSP(Research Genetics) and the human pantelomeric probe (ID labs Inc) were performed on standard metaphase spreads according to the recommendations of the suppliers.

Images were recorded using a Zeiss axiophot microscope equipped with a cooled CCD camera (Photometrics). Digitised images were captured and processed on a CytoVision Ultra workstation (Applied Imaging International Ltd).

Conventional cytogenetic examinations on G banded metaphases of the proband showed a CCR of chromosomes 2, 3, and 10. Using high resolution chromosome banding, the translocation breakpoints were mapped to 2q37.3, 3p22.3, and 10q25.2. Therefore, the karyotype is 46,XY,t(2;3;10)(2pter→2q37.3::10q25.2→10qter;3qter→ 3p22.3::?2pter;10pter→10q25.2::3p22. 3→3pter)de novo. A partial ideogram and high resolution karyotype is shown in fig 2. Analysis of the parental chromosomes did not show any abnormalities.

• Download figure
• Open in new tab
• Download powerpoint

Figure 2

Partial ideogram and high resolution karyotype (G and R bands) of the proband. The breakpoints are indicated by arrows.

RxFISH (Applied Imaging International, UK), performed on metaphase spreads of the proband, was used to analyse the whole genome for further inter- and intrachromosomal rearrangements in a single assay. Cytogenetic findings were confirmed by this technique and no further rearrangements have been detected (data not shown).

For a detailed mapping of the chromosomal breakpoints, FISH studies were performed on metaphase preparations using numerous YAC clones from the chromosomal regions 3p22.1-p22.3 and 10q24.3-q25.3, which were selected by searching the genome database (GDB). Results of our FISH analysis are summarised in table1.

Five of the chromosome 3 YAC clones (720_d_5, 793_g_8, 807_d_1, 938_g_7, 938_h_11) are assigned distal to the breakpoint. However, the more proximal YAC clones 750_d_3 and 802_g_1 show signals on the normal chromosome 3, the der(10), and the der(2) indicating a cryptic insertion of chromosome 3 material into the telomeric region of the long arm of chromosome 2 (fig 3, above). Unfortunately, the second breakpoint on chromosome 3 is, according to chromosomal 3 linkage data from GDB, about 3 cM proximal to the other breakpoint mentioned above and apparently maps to a gap between the Whitehead contigs 3.6 and 3.7. The YAC clones 712_a_7 and 758_g_3 gave hybridisation signals proximal to both chromosome 3 breakpoints.

• Download figure
• Open in new tab
• Download powerpoint

Figure 3

(Above) FISH mapping of YAC 750_d_3 to metaphase chromosomes of the patient shows hybridisation signals on the normal chromosome 3 (arrow), the der(2) (arrowhead), and the der(10) (open arrowhead), indicating that this YAC spans the distal translocation breakpoint on chromosome 3. (Below) FISH results obtained with YAC 806_h_8 which is spanning the translocation breakpoint on chromosome 10. Specific signals are localised on the normal chromosome 10 (arrow), the der(2) (open arrowhead), and the der(10) (arrowhead). Since 806_h_8 is a chimeric YAC, additional signals on chromosomes 5 and 11 can be seen. Rhodamine antidigoxigenin antibody was used to detect hybridisation sites on DAPI stained chromosomes.

To refine the localisation of the translocation breakpoint on the long arm of chromosome 10, six YAC clones specific for 10q24.3-10q25.3 were used for FISH experiments. Among those clones YAC 806_h_8 was found to span the chromosomal breakpoint since it exhibits signals on chromosome 10, the der(10), as well as the der(2) (fig 3, below). Although additional signals on chromosomes 5 and 11 indicate that this YAC is chimeric, this does not affect the usefulness of this clone. The signal of a subtelomeric probe from 2q (HBSP) is proximal to the inserted chromosome 3 material. However, by using a pantelomeric probe, no interstitial signal on the long arm of the derivative chromosome 2 could be obtained. As expected, a terminal signal of the pantelomeric probe is present on the short arm of the derivative chromosome 3 (data not shown). The karyotype according to ISCN (1995) nomenclature is 46,XY,t(2;3;10)(2pter→2q37.3::10q25.2→ 10qter;3qter→3p22.3::?2pter;10pter→10q25.2::3p22.3→ 3pter).ish t(2;3;10)(2pter→2q37.3::3p22.3::10q25.2→10 qter; 3qter→3p22.3::?2pter;10pter→10q25.2::3p2 2.3→3 pter)(2qter+, 750_d_3+, 802_g_1+,806_h_8+;712_a_7+, pantel+;806_h_8+,750_d_3+,80 2_g_1+,938_h_11+).

In our investigation we used a de novo complex chromosomal translocation which evolved as a consequence of two chromosomal breakage events on chromosome 3 and one on each of chromosomes 10 and 2. The most striking clinical features of the proband, besides the spastic quadriplegia, are microcephaly, psychomotor retardation, and distinct facial dysmorphic signs. Spasticity without detectable brain malformation is a very uncommon finding in patients with chromosomal aberrations. Even if we consider the fact that temporary spasticity is a common finding in microcephalic patients with psychomotor retardation, the clinical similarity, especially to microcephaly with spastic quadriplegia (MIM 251280), strongly suggests that at least one critical gene is affected by one of the breakpoints of this CCR. An extensive database search for genes already mapped to the chromosomal subbands 2q37.3 and 3p22.3 has shown no potential candidate gene so far, which is likely to be associated with spasticity and related motor neurone syndromes. On the other hand, the breakpoint on chromosome 10 is located near the recently published region for SPG9 (10q23.3-q24.2) and within a chromosomal region (10q23.3-q25.2) which is associated with different neurological disorders, including partial epilepsy at 10q24 (MIM 600512),33 infantile onset spinocerebellar ataxia with sensory neuropathy between D10S192 and D10S1265 (MIM 271245),34 and progressive external ophthalmoplegia between D10S198 and D10S562 (MIM 157640).35

CCRs diagnosed by standard cytogenetic analysis are frequently combined with further cryptic rearrangements,36 which can only be resolved by a detailed molecular cytogenetic analysis using more sophisticated FISH methods. New techniques such as RxFISH are capable, at a higher level of resolution, of verifying or ruling out major additional rearrangements down to several megabases in a single FISH assay. However, it has to be emphasised that even rearrangements or structural abnormalities occurring between different chromosomes but within bands labelled with the identical colour will not be detected by RxFISH. In practice, only detailed mapping with single site specific YAC clones was sufficiently sensitive to show that a tiny fragment of chromosome 3 was translocated to chromosome 2 as well. The interstitial representation of the subtelomeric 2qter probe on the derivative chromosome 2 and the fact that the pantelomeric probe showed no signal in this position but one terminal signal on the short arm of derivative chromosome 3 leads to the conclusion that either the breakpoint at 2q37.3 is precisely located between the subtelomeric and telomeric regions or telomeric repeats have been added by other mechanisms.37 By using YACs, which were previously assigned to contigs on chromosomes 3 and 10, we were also able to identify clones which cover two and flank one additional region of the chromosomal breakpoints of this CCR. Since one of the four breakpoints is located at the very end of the long arm of chromosome 2, distal even to the subtelomeric FISH probe we used, it is likely that this genomic region is of no relevance to the phenotype of the proband.

All the clinical findings could be the result of a loss or gain of function mutation of a single gene rearranged by one of the translocation events. Alternatively, an as yet unidentified submicroscopic deletion or duplication with partial aneusomy of a single or even several rearranged genes could be responsible for the phenotype. A loss or gain of chromosomal material was not detectable in our FISH analysis although an extremely small submicroscopic deletion, insertion, or duplication cannot be ruled out completely. In previous studies, it was convincingly shown that apparently balanced translocations can provide a very valuable resource for positional cloning of genes of interest.38 The breakpoint on chromosome 10 is within a chromosomal region associated with several neurological disorders. Well established methods for gene identification or isolation of transcribed sequences, such as direct cDNA selection or exon trapping, could be applied to the isolated YAC DNA as a next step to characterise the altered genomic region, which is most likely responsible for the clinical anomalies found in the proband.

In summary, FISH analysis with band specific probes derived from contigs from chromosomes 3 and 10 showed an unusual cryptic insertion and allowed very detailed characterisation of three out of four breakpoints in this CCR between chromosomes 2, 3, and 10. This CCR strongly suggests that a region of chromosome 10q defined by a single YAC (806_h_8) is likely to contain a gene for early onset paraplegia.

The molecular characterisation and biochemical studies will contribute to a better understanding of the pathogenic mechanism of this important neurological disorder.