Article Text

PDF

FMRP expression studies in blood and hair roots in a fragile X family with methylation mosaics
  1. B B A de Vries1,
  2. L-A Severijnen2,
  3. A Jacobs1,
  4. R Olmer2,
  5. D J J Halley2,
  6. B A Oostra2,
  7. R Willemsen2
  1. 1Department of Human Genetics, University Medical Centre Nijmegen, The Netherlands
  2. 2Department of Clinical Genetics, Erasmus University Rotterdam, The Netherlands
  1. Correspondence to:
 Dr B B A de Vries, Department of Clinical Genetics, University Medical Centre Nijmegen, P O Box 9101, 6500 HB Nijmegen, The Netherlands; 
 b.devries{at}antrg.umcn.nl

Statistics from Altmetric.com

The fragile X syndrome is a common cause of familial mental retardation with an estimated prevalence of 1/4000-1/6000 for males in western countries.1–3 This X linked disorder is characterised by mental retardation with additional features like a long face with large protruding ears, macro-orchidism, and eye gaze avoidance.4–6 The causative mutation is an amplification of a trinucleotide (CGG) repeat in the 5′ UTR of the FMR1 gene. Normal people have between six and 54 CGG repeats, carriers of the premutation have between 55 and 200, and affected subjects have more than 200 CGG repeats in their FMR1 gene, the so called full mutation.7,8 The latter expansion is accompanied by hypermethylation of the repeat and its upstream region resulting in a shutdown of transcription and absence of the FMRP.9–11

In fragile X patients, two special subclasses of mosaicism can be distinguished on the basis of size and methylation pattern: (1) subjects with a premutation in a proportion of their cells in addition to a full mutation, often referred to as “size mosaics”; this pattern can be observed in 20–40% of male patients12,13; (2) subjects with intercellular variations in the methylation status of a full mutation, “methylation mosaics”.14 In a large multicentre study, “methylation mosaicism” was observed in 3% of the males with a full mutation.12

Key points

  • In fragile X patients two special subclasses of mosaicism can be distinguished on the basis of size and/or methylation pattern: patients with full mutation and premutation, called “size mosaics”, and patients with intercellular variations of the methylation status, called “methylation mosaics”.

  • Within a known fragile X family, three brothers with methylation mosaic patterns were studied using the FMRP antibody test on both blood smears and hair roots. The index patient aged 10 years (case 1) was diagnosed at the age of 5 years; he was mildly retarded and had some clinical fragile X features. DNA analysis showed a full mutation (200–250 repeats) with 86% unmethylated mutations. His 5 year old brother (case 2) had a mutation of 177 repeats that was unmethylated in 67% of cells. His (early) development was considered normal and he lacked additional fragile X features. The third brother (case 3) had a mutation of 183–187 repeats that was unmethylated in 86% of cells. His development was normal at the age of 3 years and he did not have any significant fragile X features.

  • In cases 1, 2, and 3, FMRP expression in blood (in duplicate) was 20% and 10%, 22% and 10%, and 7% and 2%, respectively, and 67%, 85% and 88% of their hair roots expressed FMRP.

  • These results suggest that FMRP expression in hair roots gives a better reflection of the mental development than FMRP expression in leucocytes, which is consistent with the common embryonic origin of hair roots and neuronal cells.

In 1995, Willemsen et al15 developed an FMRP antibody test for detecting the presence or absence of FMRP in lymphocytes and later in hair roots.16 This test allowed for screening for the fragile X syndrome among mentally retarded males17 and in addition made quantification of the number of FMRP expressing cells in patients possible. Using this technique in blood smears, Tassone et al18 were able to find correlations between FMRP expression and IQ in males with size mosaicism and methylation mosaicism. Very recently, a highly significant correlation has been found between FMRP expression in hair roots and cognitive functioning in females carrying a full mutation (R Willemsen, in press).

Here, a study of FMRP expression in blood and hair roots is reported in three male sibs with a varying degree of methylation mosaicism.

PATIENTS AND METHODS

The three brothers who are the subjects of this report are from a fragile X family known to the Department of Clinical Genetics, Rotterdam. The family was ascertained through a son of the mother’s sister who had the classical clinical presentation of the fragile X syndrome confirmed by a fully methylated full mutation of the FMR1 gene.

DNA analysis

Genomic DNA was isolated19 from blood leucocytes digested with HindIII and the methylation sensitive enzyme EagI and hybridised with probe pP2 according to standard protocols.20

Sizing of the Southern blot and densitometry were done using a Kodak Electrophoresis Documentation and Analysis System 120.

Protein analysis

Blood smears were made from one drop of blood within two hours after collection. Slides were air dried. Hairs were plucked from different locations on the scalp and analysed within 24 hours. The FMRP was visualised by using monoclonal antibodies 1A1 against FMRP.21 Further immunoincubations were performed according to procedures described previously15,22 (http://www.eur.nl/FGG/CH1/frax/). A total of 100 leucocytes were analysed per patient and scoring for FMRP expression was performed by two people independently.

Determination of IQ levels

The Wechsler Intelligence Scale for Children-Revised (WISC-R), the McCarthy Scales of Children’s Abilities (MSCA), the Wechsler Preschool and Primary Scale of Intelligence (WIPPSI-R), and the Peabody Picture Vocabulary Test-Revised (PPVT-R) were used to test the intellectual abilities by one examiner (AJ) who was not informed about the genetic status of the children tested. The WISC-R is suitable for children from the age of 6 years whereas the MSCA, the WIPPSI-R, and PPVT-R should be used for younger children. The verbal, performance, and full scale IQ scores were calculated.

RESULTS

Case reports

Case 1

This boy was born after a normal pregnancy and delivery with a birth weight of 3250 g. In his first year of life he had frequent ENT related problems which disappeared after tonsillectomy and the insertion of grommets. His early development was somewhat slow but within the normal range: he sat at 8 months, walked unaided at 18 months, and spoke his first words at 1 year. At the age of 2 years hyperactive behaviour was noted which disappeared at 4 years of age. However, at that age he appeared to be unable to attend normal school. For further evaluation he was referred to our centre and because of the family history (the mother’s sister had two sons with the fragile X syndrome) DNA analysis of the FMR1 gene was performed.

At the age of 5 years he had normal physical measurements, height 1.17 m (70th centile) and head circumference 52.5 cm (75th centile). He had a long, narrow face with a normal chin, normally shaped and sized ears, periorbital fullness, and normal teeth (fig 1). His testes were mildly enlarged for his age (4 ml/4 ml). He had hyperextensible finger joints and relatively broad and short halluces. His behaviour was normal with normal eye contact.

Figure 1

(A) Case 1 at the age of 5 years, (B) case 2 at the age of 5 years, and (C) case 3 at the age of 3 years.

Psychological testing at the age of 10 years, using the WISC-R, showed a full scale IQ score of 55 with a verbal IQ score of 56 and a performance IQ score of 61.

Case 2

This boy is the 4 year younger brother of case 1. He was also born after an uneventful pregnancy and delivery with a normal birth weight. His early development was normal: he sat at 9 months, stood at 10 months, and walked unaided at 14 months. Because of the diagnosis in his older brother, he was tested for the fragile X syndrome at the age of 1 year 3 months.

At the age of 5 years he had normal physical measurements, height 110 m (25th centile) and head circumference 51.8 cm (60th centile). He had a normal face, except for a broad forehead which was observed in the father as well, some periorbital fullness, and normal ears (fig 1). His genitals were normal. His behaviour was normal with normal eye contact.

Psychological testing at the age of 5 years showed a full scale IQ score of 81 with the MSCA and 75 with the WIPPSI-R.

Case 3

This boy is the 6 years younger brother of case 1. He was born after a normal pregnancy and delivery with a birth weight of 3750 g. His development was normal: he walked at 14 months and spoke normal sentences at 3 years.

At the age of 3 years he had normal physical measurements, height 1.00 m (50th centile) and head circumference 51.5 cm (70th centile). He had no dysmorphic facial features, except for a broad forehead which was observed in the father as well; he had normal sized and shaped ears. His genitals were normal. He had some hyperextensibilty of MCP V. His behaviour was normal with normal eye contact. Psychological testing at the age of 3 years 8 months showed a full scale IQ score of 91 with the MSCA and 97 with the PPVT-R.

Molecular findings

In case 1, a full mutation was found in his leucocytes using Southern blot analysis: a 14% methylated 560 bp larger than normal band (~217 repeats) and a 86% unmethylated 670 bp larger than normal band (~253 repeats) (fig 2).

Figure 2

Analysis with probe pP2 of HindIII/EagI digested DNA (leucocytes) of the mother, cases 1, 2, and 3, and a normal sib.

Protein analysis in blood smears (in duplicate) showed 10% and 20% of the leucocytes and 67% of the hair roots (10/15) expressing FMRP.

In case 2, a mutation that was smaller than in case 1 was found in leucocytes using Southern blot analysis: a 33% methylated 440 bp larger than normal band (~177 repeats) and a 67% unmethylated 440 bp larger than normal band (~177 repeats).

Protein analysis in blood smears (in duplicate) showed 10% and 22% of the leucocytes and 85% of the hair roots (17/20) expressing FMRP.

In case 3, a mutation similar in size to that of case 2 was also found in leucocytes using Southern blot analysis: a 14% methylated 470 bp larger than normal band (~187 repeats) and an 86% unmethylated 460 bp larger than normal band (~183 repeats) (fig 2).

Protein analysis in blood smears (in duplicate) showed 2% and 7% of the leucocytes and 88% of the hair roots (23/26) expressing FMRP.

The test results are summarised in table 1.

Table 1

Summary test results in the three brothers

DISCUSSION

According to the definition of mosaicism, all fragile X males are mosaic as they have different amplified repeat sequences in the FMR1 gene in different cells. However, two special subclasses can be distinguished; 20–40% of fragile X males have a premutation in a proportion of cells and the full mutation in the remaining (majority) of cells. In general, the proportion of cells with a premutation is lower than <30% and various studies have shown that the cognitive functioning of these fragile X patients is not significantly better than the males with a full mutation, suggesting that the number of FMRP expressing cells with a premutation is insufficient.12,23 However, high functioning males with a size mosaic pattern have been described24–26 and Tassone et al18 detected a correlation between FMRP expression in blood smears and IQ in mosaic males.

For the second group of mosaic patients, the so called “methylation mosaics” the situation is different. These patients have a full mutation in all cells but in a proportion of cells the full mutation is unmethylated. The cells with an unmethylated full mutation are able to produce FMRP and can therefore function normally. Depending on their proportion, they are able to compensate for the loss of functioning of the cells with a methylated full mutation. Various reports on patients with methylation mosaicism suggest that a proportion of cells with an unmethylated full mutation of at least 40% of normal is likely to be required for normal cognitive functioning.24,27–36 This is supported by actual FMRP studies in blood smears of patients with methylation mosaicism that showed that all mosaic patients with a normal IQ had FMRP in ⩾50% of lymphocytes.18,32 This situation is reminiscent of the situation in females with a full mutation where cognitive function is related to the X inactivation pattern.12,14,37–39

FMRP expression studies in blood smears of methylation mosaic males have been reported. Smeets et al31 reported normal protein expression in cell lines of two normal functioning adults with an unmethylated full mutation. De Vries et al32 reported three cousins with 75%, 40%, and 10% cells expressing FMRP who had an unmethylated full mutation in 90%, 35%, and 10% of the cells, respectively. The latter two were both retarded whereas the adult males with 75% FMRP expressing cells had a normal IQ. Tassone et al18 found a correlation between IQ and FMRP expression in blood smears in 13 males with a partially methylated full mutation. They also found three non-retarded mosaic males with expression of FMRP in ⩾50% of lymphocytes.18 The findings of normal FMRP expression in partially unmethylated full mutations were in contrast with the report of Feng et al,40 who found markedly diminished FMRP production in fibroblast clones from transcripts with more than 200 repeats. These conflicting findings raise the question of whether unmethylated full mutations have normal or diminished FMRP expression, what is the relation to cognitive functioning, and what is the correct tissue to study. Interestingly, Tassone et al41 reported a six-fold increase of FMR1 mRNA levels in methylation mosaic males suggesting the existence of a compensatory response to impeded FMRP production.

In the oldest of the three reported brothers, the size of the (un)methylated alleles are all in the full mutation range whereas the other two brothers have (un)methylated alleles in the high premutation range; thus the latter two do have partially methylated premutation sized alleles which is quite rare.

The proportion of cells expressing FMRP in a blood smear ascertained by the FMRP antibody test did not correspond very well with the proportion of unmethylated FMR1 alleles (pre- or full mutation sized) as ascertained by DNA blotting analysis. It suggests that in leucocytes the translation might also be hampered in the large unmethylated premutation sized alleles. It also shows that accurate prediction of mental functioning in males with an intercellular variation of the methylation status through FMRP studies in blood smears is, like DNA analysis, less valid. However, FMRP expression in hair roots did reflect the cognitive functioning in the three brothers. Both brothers with normal IQs (81 and 91) had a high proportion of FMRP expressing hair roots (85% and 90%, respectively). This is consistent with the common embryonic origin, ectoderm, of hair roots and neuronal cells whereas blood is of mesodermal origin. Of course a larger number of males with (un)methylated full mutations need to be tested to assess the validity of the relationship between FMRP expression in hair roots and mental functioning.

Acknowledgments

We thank the patients and their family for their kind support.

REFERENCES

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.