Article Text


Mosaicism for FMR1 and FMR2 deletion: a new case
  1. S Fengler,
  2. S Fuchs,
  3. R König,
  4. J Arnemann
  1. Institute of Human Genetics, Johann Wolfgang Goethe University Hospital, Theodor-Stern-Kai 7, D-60590 Frankfurt/Main, Germany
  1. Correspondence to:
 Dr J Arnemann, Institute of Human Genetics, Johann Wolfgang Goethe University Hospital, Theodor-Stern-Kai 7, D-60590 Frankfurt/Main, Germany;

Statistics from

Fragile X syndrome is the most common disorder causing mental retardation with an estimated prevalence of between 1 in 4000 to 1 in 6000 males.1, 2 In nearly all cases, the molecular basis of this X linked mental impairment is an amplification in excess of 200 copies of an unstable CGG trinucleotide repeat sequence within the 5` untranslated region of the FMR1 gene (FRAXA, MIM 309550) at Xq27.3.3 This causes an inhibition of the transcription leading to a loss of function effect in the male. In the female, the amplified gene is preferentially X inactivated by methylation.4 However, the X inactivation pattern can vary between different cell types and during development and thus, usually, causes a milder phenotype in female patients.5 In close distal proximity to the FMR1 gene, a second gene, FMR2 (FRAXE, MIM 309548) involved in mental retardation, has been described with amplification of an unstable GCC trinucleotide repeat sequence as the underlying pathological mechanism.6 This report adds a further case with mosaicism for a FMR1 and FMR2 deletion to the few described cases with a large FMR1 deletion.7–11


A boy of 3½12 years of age was referred to our Clinical Genetics Unit because of severe developmental delay, behavioural difficulties, and delayed speech development. He is the first child of healthy, unrelated Turkish parents. The boy was born two weeks before term with a birth weight of 2690 g, length of 49 cm, and OFC of 32.5 cm. He had feeding and eating problems which still persisted at the age of 3½12 years. At 3 months he was treated with physiotherapy for muscular hypotonia.

Clinical investigation at 3½12 years showed height (97.5 cm) and weight (14.2 kg) between the 25th-50th centile, a long face with prominent forehead and small and pointed chin, periorbital fullness, big ears, a short nose with anteverted nares, a long, smooth philtrum (fig 1), microdolichocephaly (OFC 48.5 cm, <3%), marked joint hypermobility, broad hands and feet, short toes, and truncal hypotonia. He only speaks a few words. He is hyperactive and his behaviour is characterised by aggression and self-mutilation, but he calms down while listening to music. There is no history yet of epilepsy as has been described in similar cases.7, 12 He suffers from recurrent otitis media with effusion which may have caused subsequent language and articulation deficits.

Figure 1

The proband. Note the long face with prominent forehead, pointed chin, periorbital fullness, big ears, short nose, and long, smooth philtrum. (Photograph reproduced with permission.)

Cytogenetic analysis of peripheral lymphocytes at a 700 band level showed a normal male karyotype, 46,XY. Routine molecular analysis for trinucleotide amplification of the FMR1 gene by Southern blot hybridisation of genomic DNA from white blood cells with an FMR1 probe (GL30, a kind gift from A Poustka's laboratory) failed to give any result for our patient, leaving the possibility of a deletion. Hybridisation with an FMR2 probe (Oxe20, a kind gift from Kay Davies's laboratory) confirmed a deletion covering at least FMR1 and FMR2. The same results were obtained testing genomic DNA from the patient's fibroblasts. However, when PCR based assays for the non-variable regions of the FMR1 (FRAXA-S: CTGCAGGAGGCGGCCCGG; FRAXA-A: GCTAGCAGGGCTGAAGAGAAGATG) and FMR2 genes (FMR2-S: GTCACACACAGAGCTACCGGC; FMR2-A: CACGTGGAGGGTTTAACCGAG) were applied to genomic DNA from leucocytes and from fibroblasts, clear bands were obtained which suggested the existence of a cellular mosaicism as a likely explanation of the laboratory findings. Given the established sensitivity of our Southern blot hybridisation conditions, we estimated the presence of the second cell line to be less than 10% (based on other data not shown) in the tested tissues. This level, however, does not necessarily reflect the state of mosaicism in the brain. Standard Southern blot hybridisation as well as methylation analysis showed a carrier pattern with amplification of the trinucleotide repeats within the premutation range for his mother, while his father was normal. Testing for a possible amplification of the CGG repeats in the remaining cell line by PCR (FMR1-A: GGAACAGCGTTGATCACGTGACGTGGTTTC; FMR1-571R: GGGGCCTGCCCTAGAGCCAAGTACCTTGT) did not give a normal sized band in the patient, suggesting that the FMR1 gene of the remaining cell line most likely originated from the mother's premutated gene. The deletion in the prominent cell line must have been a postzygotic event covering the region between the FMR1 and FMR2 genes with an estimated size of at least 900 kb, based on knowledge of the distance between FMR1 and FMR2 as reported in the Genetic Location Database (LDB;


A few cases with large deletions in a complete or mosaic pattern have been published, which were preferentially de novo or inherited deletions,10, 11, 13–15 while small mosaic deletions are usually associated with premutations.8 The basic events giving rise to these deletions are not yet understood.

The small head, the severe developmental delay, and the persistent feeding and eating problems are not typical for fragile X syndrome, but might reflect a functional or developmental imbalance caused by the deletion of additional genes in the majority of cells. Similar clinical phenotypes with moderate to severe mental retardation, epilepsy, and deletion of the FMR1 and FMR2 genes have been described previously.7, 10 Microcephaly is atypical in fragile X syndrome; however, the patient has a fragile X-like facial appearance. Two other published patients with a large deletion10 had head circumferences at or below the 3rd centile as well, while a third patient with a deletion of the FMR1 and FMR2 genes7 was macrocephalic. While the clinical findings of our patient suggest that the deletion may be larger, encompassing more genes than just FMR1 and FMR2, the deletion has apparently arisen on the background of an inherited triplet repeat mutation. Usually, in previous reports, deletions arising in the presence of a triplet repeat expansion have been small while large deletions have not been associated with triplet repeat expansion as in our case.16 We thus speculate that our patient has mosaicism of cells with an FMR1 gene functionally inactivated by repeat amplification and cells with a large deletion of FMR1, FMR2, and possibly neighbouring genes, thus spreading the classical fragile X phenotype.


View Abstract

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.