| Human Molecular Genetics | Pages |
Introduction
Results
Sequence of the fmr2 gene
Fmr2 expression in the adult mouse brain
Developmental expression of fmr2 in the mouse embryo
Discussion
Materials And Methods
Isolation and sequencing of the mouse fmr2 cDNA and the human FMR2 cDNA
Northern hybridisation
RNA in situ hybridisation
Synthesis of digoxygenin (DIG) labelled riboprobes
Hybridisation of riboprobe to tissue sections
Post hybridisation washing
Antibody detection
Post antibody washes and histochemistry
Acknowledgements
References
Expression of the murine homologue of FMR2 in mouse brain and during development
DDBJ/EMBL/GenBank accession nos AJ001549, AJ001550
The expression of the FRAXE fragile site on the human X chromosome is associated with the expansion of a CCG repeat at the 5[prime] end of the FMR2 gene. The repeat expansion results in transcriptional silencing of the gene and this event has been found to be associated with mild mental handicap in families. We have previously shown that the gene is particularly abundantly expressed in the hippocampus and amygdala by northern analysis. Here we demonstrate the expression pattern of the homologous gene in adult mouse brain and early mouse embryos. High levels of fmr2 mRNA were noted in the hippocampus, the piriform cortex, Purkinje cells and the cingulate gyrus. Expression of fmr2 occurs on, or before, day 7 in the embryo and reaches its highest levels at 10.5-11.5 days. A more detailed analysis shows that the fmr2 expression in the embryo at 11 days is more specific and evident to the roof of the hind brain and the lateral ventricle of the brain. The coding sequence of the mouse fmr2 gene shows very high conservation with 88% amino acid identity to the human FMR2 sequence. The folate sensitive FRAXE is situated in Xq28 on the human X chromosome, 600 kb distal to FRAXA, the fragile site associated with fragile X syndrome (1-3). The fragility is caused by the expansion of a CCG trinucleotide repeat array (3) and this event is found to be associated with a rare form of mild mental handicap in families (4-14). Normal X chromosomes have 6-35 CCG repeat units at the FRAXE locus. Expansion of the repeat to >200 units leads to the expression of the fragile site and methylation of the adjacent CpG island is observed (3). Unlike in the fragile X syndrome where the vast majority of events on female transmission are expansions, the FRAXE repeat may expand or contract on transmission and males with the full mutation have had affected daughters with full mutations (4).
The FRAXE phenotype shows no consistent dysmorphology but there are problems with learning, particularly speech delay, reading and writing problems (4-14). There is also an increased incidence of behavioural deficits in FRAXE individuals. These include attention deficit, hyperactivity and in some cases autistic- like behaviour. The inconsistent phenotype has led to some debate as to whether FRAXE is associated with a phenotype or whether the observed features are a result of ascertainment bias. A study of 300 males referred for fragile X syndrome testing did not reveal a single FRAXE case suggesting that FRAXE may be a benign fragile site unrelated to any clinical phenotype (15). However, only 19 of the individuals tested were fragile site positive in Xq27-28. A subsequent study also screened 300 mentally retarded subjects. A FRAXE family was detected including a mother with an unmethylated premutation and seven children, six with methylated full mutations and the other with an unmethylated premutation. The six children with full mutations had social problems and mild to moderate mental retardation (6). Screening of boys with learning difficulties in the UK did not reveal any FRAXE full mutations and revealed only 1 in 4994 FRAXA full mutations (16).
Deletions within 200 kb of FRAXE in two boys with developmental disorders were reported by Gedeon et al. (17). This provided further evidence for a gene in the FRAXE region. Subsequently, it was shown that the CCG repeat lay at the 5[prime] end of a gene which was designated FMR2 (18-20). Thus it was likely that the methylation of the FRAXE locus associated with repeat expansion would cause the inactivation of this gene (3). Indeed, it has been shown that there is a loss of expression of FMR2 in fibroblast cell lines from FRAXE patients when compared with controls with normal sized repeat arrays at FRAXE (19,21).
Northern analysis using human samples shows expression of a 9.5 kb transcript in the adult human placenta and brain (18-20). A more detailed analysis of the brain by northern blotting reveals a widespread expression of FMR2 in different areas of the adult human brain with noticeably higher levels of the 9.5 kb transcript in the tissues of the hippocampus and amygdala (18). High levels of this transcript in these regions of the brain are consistent with a phenotype of learning and memory difficulties. The study presented here therefore aimed to look at the developmental expression of the mouse homologue of FMR2 (fmr2), in the brain of the adult mouse and to identify the structures and cell types which express the gene. The conservation of FMR2 was analysed in detail by comparing the mouse homologue of FMR2 (fmr2) with the human cDNA. A partial mouse cDNA clone was isolated from a mouse brain phage library (Uni-Zap, Stratagene). This clone was extended by RT-PCR to encompass the complete coding regions of the mouse fmr2. The cDNA sequence is 4504 bp and encodes 1272 amino acids. The predicted sequence of the mouse fmr2 protein is highly homologous to the human sequences with an amino acid identity of 88% (Fig. 1). The mouse sequence detects a 9.5 kb mRNA in brain and because of the homology to the human sequence is highly likely to be the mouse homologue of the human sequence. The GC-rich 5[prime] UTR in the mouse cDNA clone also shows high homology to the human equivalent but it does not contain a pure CCG repeat. There are 11 CCG elements but these are interrupted five times with other base sequences. All parts of the FMR2/fmr2 protein are highly conserved and seem therefore to be of importance for the function of FMR2. However, the N- and C-terminal parts are the regions most highly conserved between the human and mouse genes (Fig. 1). FMR2 has two homologues in the database, AF-4 and LAF-4 (22,23; Fig. 2), both putative transcription factors (22). In addition, two non-overlapping ESTs also homologous to FMR2 were recently identified (24). Highly conserved sequences in the N- and C-terminal regions of the three human members of this protein family are also conserved in the mouse fmr2 sequence (Fig. 2) and may be important for the common function of these proteins.
Figure
Figure
We also cloned the coding region of the human FMR2 gene. Our human sequence is virtually identical to the published human cDNAs (18-20), but differs in the combination of splice sites used, when compared with the recently reported gene structure of FMR2 (24). In the 5[prime] region of the mouse cDNA a small insert of 12 bp, encoding the four amino acids VAEY, was found, as indicated by a box in Figure 2. Interestingly, this insert was also found in one of the new human cDNA clones and not in another, corresponding to the use of exons 2 and 2A, respectively (24). The differentially spliced exon 5 and exon 7A (24) (indicated by arrowheads in Fig. 2) are found neither in the murine cDNA nor in our new human cDNA. Further evidence for splice variants is provided by the insertion of the dipeptide VN. This dipeptide (boxed in Fig. 2) is found in our human cDNA and in the human cDNA reported by Gecz et al. (19). It is not encoded in the human FMR2 cDNA sequence reported by Gu et al. (20), however, and may therefore in humans be due to different splicing at the 3[prime] end of exon 13. In the murine sequence the dipeptide is conserved as IN.
In addition, two likely polymorphisms were found in the new human sequence. These differences are boxed in Figure 2. In position 435 of the new human FMR2 protein an alanine (GCC) is found instead of a serine (TCC; 19) or a valine (GTC; 20). Also, at position 513, a glutamine (CAA) is substituted for the proline (CCA) found in the previously reported human cDNA sequences. The new human protein is identical to the murine protein in these two positions. This sequence was found in two independent but overlapping cDNAs. RNA in situ hybridisation techniques were used to get a more detailed picture of the cells in the brain which express fmr2. We used the previously cloned Ox19.3 fragment of the Ox19 short FMR2 human cDNA as a probe (18). This probe gives a clear specific band of 9.5 kb on northern blots (18-20; see below) and only detected sequences on the human X chromosome. The probe shows 85% homology to the mouse sequence and does not cross hybridise with other known members of the gene family.
Coronally cut sections of the hind brain revealed a labelling of the cingulate gyrus, hippocampus, piriform cortex and the Purkinje cell layer (Fig. 3). Sense probes showed no staining in these regions indicating that the hybridisation was specific.
Figure
The expression pattern during mouse development was performed to identify the precise time at which the fmr2 gene is switched on. The same survey can also narrow down the gestational stage at which levels of fmr2 expression might be most readily detected.
A mouse embryo northern blot (Clontech) was used for this study. Four tracks of poly A+ RNA on this blot corresponded to RNA from mouse embryos at days 7, 11, 15 and 17 p.c. (post coitum). The blot was hybridised with the Ox19.3 fragment. A single band of 9.5 kb in size was observed in all the tracks (Fig. 4). The signal remained on stringent washing at 0.1× SSC, 0.1% SDS. The strongest signal was observed in the 11 day p.c. mouse embryo track. Hybridisation with an actin probe shows equal loading (data not shown).
Figure
Mouse embryos at age 10.5, 12.5 and 15.5 days p.c. were examined by RNA in situ hybridisation. The Ox19.3 DIG labelled riboprobe was used for these analyses. Expression overall was at highest levels in the 10.5 day embryo. The embryo at 15.5 days p.c. had little, mainly diffuse, expression throughout but this may not have been specific since the sense probe also gave a very low level of signal.
At 10.5 days p.c. signal was seen around the telencephalic vesicle, mesencephalic vesicle and the fourth ventricle (Fig. 5A). There was also staining of the mandibular component of the first branchial arch and the third branchial pouch.
Figure
Below the head region the atrial chamber of the heart and first branchial arch are stained (Fig. 5C). Also, the left hindlimb bud shows expression. At the caudal end of the embryo the neuroepithelium of the neural tube shows expression along its length. The sections hybridised with sense probe had no staining (Fig. 5B and D).
At 12.5 days p.c. most of the staining seen at 10.5 days was no longer present, and there seemed to be a low level of diffuse expression in many of the tissues (data not shown). The only structures which showed staining at high levels were the roof of the hind brain and the lateral ventricle, including the roof of the neopallial cortex (future cerebral cortex) and corpus striatum mediale. There was also staining around the cochlea. The expression of FMR2 is absent in individuals with a mild mental handicap associated with expansion and methylation of a trinucleotide repeat at FRAXE (21). Therefore, it appears that the FMR2 gene may be involved in certain learning, memory and language learning processes. So far, very little is known about the function of FMR2. Information about the mouse fmr2 sequence and expression pattern can help to highlight the role of FMR2 in mammals.
The high level of conservation between human and mouse FMR2/fmr2 sequences makes the mouse a useful model for studying the expression of the FMR2 gene In the experiments described here we used a human cDNA probe which shows 85.4% identity to the mouse sequence. Both human and mouse probes identify identical bands in northern blots. Fmr2 expression was detected in several different areas of the adult mouse brain. The expression in the hippocampus of the mouse brain is concordant with the previously published northern analysis of human brain samples (18). The hippocampus is an essential component in learning (25). At first, the expression of fmr2 in the cingulate gyrus was surprising. However, the cingulate gyrus is intimately connected with the hippocampus and amygdala as part of the limbic system. The amygdala plays a role in emotional learning (26,27).
Other cells of the brain which express fmr2 are the Purkinje cells in sections of the hind brain, and the cells of the piriform cortex which lie in the cerebellar cortex. The piriform cortex is also composed of Purkinje cells which highlights a possible role for this type of cell in learning. Purkinje cells are commonly associated with motor and balance skills but are now increasingly thought to be involved in motor learning processes (28). The cells of the piriform cortex are linked to amygdala formation; the amygdala receives afferents from the piriform cortex. These data do not prove the association of fmr2 with learning and memory. However, they do provide a basis for analysis of the cell types in which fmr2 expression may play a key role.
FMR2 is believed to function as a transcription factor based on studies of homologous proteins (22,23). A striking similarity to two other proteins altered in acute leukaemia, AF-4 and LAF-4, may provide clues to the function of FMR2. Ma et al. (22) present evidence that LAF-4 is a nuclear protein and that it binds DNA with high affinity. Furthermore, both LAF-4 and AF-4 possess strong transcriptional activation properties. The conserved region in all the members of this protein family may indicate the position of an active domain. Experiments to test the function of fmr2 in this respect are underway. Total RNA from mouse brain was retrotranscribed and the resulting first strand cDNA was used in a PCR reaction utilising primers specific for FMR2. The PCR product was sequenced and shown to be part of the mouse FMR2 homologue (fmr2). Primers derived from this sequence were used to PCR screen a mouse fetal brain cDNA library (Unizap, Stratagene) essentially as described by Amarvardi and King (29). A positive lambda clone was isolated and the recombinant Bluescript SK+ plasmid was recovered by in vivo excision and sequenced. Additional regions of the predicted fmr2 cDNA sequence were obtained by direct sequencing of PCR products. These PCR products were generated by RT-PCR on mouse brain cDNA using primers chosen from the 3[prime] end of the lambda clone and from the 3[prime] coding and UTR of the FMR2 cDNA sequence (14). The coding region of the human FMR2 cDNA was obtained in three highly overlapping fragments by RT-PCR on human foetal brain mRNA. The fragments were generated using the Pwo polymerase and the primer pairs CTATGGATCTATTCGACTTTTTCAG and TTTGGTTTGGGCCAGGTAAA; GAGCCTGTGAAGACCTTGAC and GATTTGTCCAGTCATGGC- GATGT; CCTGTCATGCAAACTGAAATCCTGTC and CTACAACAAGTGGGCATCGATGCGCAGCCA. At least two clones of each fragment were sequenced at both strands. Both the cDNAs were sequenced using the Amplitaq FS Dye terminator kit and reactions were run on an ABI 377. A mouse embryo northern blot (Clontech) containing 2 µg poly A+ RNA from embryos of 7, 11, 15 and 17 days p.c. was hybridised with the human probe Ox19.3 using Expresshyb according to manufacturers' instructions (Clontech). The blot was washed to a stringency of 0.1× SSC and 0.1% SDS. This technique was essentially as described by Wilkinson (30). Reagents were supplied by Boehringer Mannheim unless stated otherwise. Mouse tissues were fixed in 4% paraformaldehyde. Embryos of an age greater than 12.5 days p.c were punctured through the epidermis at several points to allow diffusion of fixative into the internal organs. Adult mouse brains were divided into three coronally sectioned fragments. Tissues were equilibrated in 15% sucrose before being embedded and frozen in Tissue-tek embedding compound (BDH). Tissue blocks were cut to provide 0.8 µm sections using a cryostat (Leica). Sections were captured on super-frost microscope slides (BDH) and dried at room temperature before use. At least six different animals of 11.5 days p.c. were used. The Ox19.3 partial human cDNA clone was used to synthesise a riboprobe (5). The clone was linearised on either side of the probe to provide sense and antisense templates. Linearised plasmid was purified using a wizard DNA clean up column (Promega). The riboprobe synthesis reaction used 1 mg Ox19.3 linearised plasmid made up to a volume of 14 µl with sterile DEPC treated water, 1 µl of transcription buffer (400 mM Tris-HCl, pH 8.25; 60 mM MgCl2 and 20 mM spermidine), 1 µl 0.2 mM DTT, 2 µl nucleotide mix pH 8.0 (10 mM GTP/ATP/CTP, 6.5 mM UTP, 3.5 mM digoxygenin-UTP), 50 U RNAse inhibitor and 10 U SP6, T7 or T3 RNA polymerase. The reaction was incubated at 37°C for 2 h. One µl of the reaction was tested for synthesis and the remainder of the reaction was treated with 50 U of DNAse. The riboprobe was precipitated at -20°C with the addition of 100 µl water, 10 µl LiCl (10 M) and 300 µl absolute alcohol. The pellet was resuspended in 50 µl DEPC treated water. Prehybridisation of sections was carried out for 1 h. Each section was covered with hybridisation solution [50% formamide, 5× SSC, 2% Blocking powder, 0.1% Triton X-100, 0.5% CHAPS (Sigma), 1 mg/ml yeast RNA, 5 mM EDTA, 50 mg/ml heparin] and then incubated at 65°C in a humidified chamber for 1 h. Hybridisation was carried out with the addition of 0.2-1 mg/ml riboprobe to fresh hybridisation solution and the sections were incubated overnight at 65°C. Sections were washed at 65°C for 5 min with solution 1 (50% formamide, 5× SSC, 0.1% Triton X-100, 0.5% CHAPS). Then for 5 min with 70% solution 1, 30% 2× SSC and again for 5 min with 30% solution 1, 70% 2× SSC. Two further washes were carried out at the same temperature using 2× SSC, 0.1% CHAPS. To prepare for antibody detection sections were washed twice for 10 min with TBT at room temperature. Sections were preblocked with 10% sheep serum, 2% BSA in TBT for 30 min. This was then removed and replaced with anti-dioxygenin (Boehringer) antibody preabsorbed with mouse embryo powder overnight at 4°C. Sections were incubated overnight at 4°C. After overnight incubation, sections were washed with TBT for 5 min three times at room temperature and then three times for 30 min each. Finally they were washed with freshly made NTM three times for 5 min. The colour reaction was made up with 3.2 ml NBT and 3.5 ml BCIP per ml NTM and the colour was allowed to develop in the dark. The colour reaction was stopped with PBT. We would like to thank Dr Judith Skinner for help with the in situ hybridisation protocol, Dr Lee Buttery for advice on tissue fixation and sectioning and Dr Jon Tinsley for assistance with microscopy and for general discussion. We are also grateful to Dr Judith Skinner and William Miller for assistance in northern blot analysis. We would also like to thank Helen Blaber for help in the preparation of this paper. This work was supported by the Medical Research Council, UK, Wennergren-Center Foundation (J.B.) and the Action Research, UK. G.S.F. was a Norwegian Oxford Scholar.
This article has been cited by other articles:
INTRODUCTION
RESULTS
Sequence of the fmr2 gene
Fmr2 expression in the adult mouse brain
Developmental expression of fmr2 in the mouse embryo
DISCUSSION
MATERIALS AND METHODS
Isolation and sequencing of the mouse fmr2 cDNA and the human FMR2 cDNA
Northern hybridisation
RNA in situ hybridisation
Synthesis of digoxygenin (DIG) labelled riboprobes
Hybridisation of riboprobe to tissue sections
Post hybridisation washing
Antibody detection
Post antibody washes and histochemistry
ACKNOWLEDGEMENTS
REFERENCES
This page is run by Oxford University Press, Great Clarendon Street, Oxford OX2 6DP, as part of the OUP Journals
Comments and feedback: www-admin{at}oup.co.uk
Last modification: 14 Feb 1998
Copyright© Oxford University Press, 1998.
CiteULike 
Connotea 
Del.icio.us What's this?
E. Bitoun, P. L. Oliver, and K. E. Davies
The mixed-lineage leukemia fusion partner AF4 stimulates RNA polymerase II transcriptional elongation and mediates coordinated chromatin remodeling
Hum. Mol. Genet.,
January 1, 2007;
16(1):
92 - 106.
[Abstract]
[Full Text]
[PDF]
A. Urano, M. Endoh, T. Wada, Y. Morikawa, M. Itoh, Y. Kataoka, T. Taki, H. Akazawa, H. Nakajima, I. Komuro, et al.
Infertility with Defective Spermiogenesis in Mice Lacking AF5q31, the Target of Chromosomal Translocation in Human Infant Leukemia
Mol. Cell. Biol.,
August 1, 2005;
25(15):
6834 - 6845.
[Abstract]
[Full Text]
[PDF]
P. L. Oliver, E. Bitoun, J. Clark, E. L. Jones, and K. E. Davies
Mediation of Af4 protein function in the cerebellum by Siah proteins
PNAS,
October 12, 2004;
101(41):
14901 - 14906.
[Abstract]
[Full Text]
[PDF]
Y. Gu, K. L. McIlwain, E. J. Weeber, T. Yamagata, B. Xu, B. A. Antalffy, C. Reyes, L. Yuva-Paylor, D. Armstrong, H. Zoghbi, et al.
Impaired Conditioned Fear and Enhanced Long-Term Potentiation in Fmr2 Knock-Out Mice
J. Neurosci.,
April 1, 2002;
22(7):
2753 - 2763.
[Abstract]
[Full Text]
[PDF]
C. J. Cummings and H. Y. Zoghbi
Fourteen and counting: unraveling trinucleotide repeat diseases
Hum. Mol. Genet.,
April 1, 2000;
9(6):
909 - 916.
[Abstract]
[Full Text]
[PDF]
This Article 
Abstract
FREE Full Text (PDF)
Alert me when this article is cited
Alert me if a correction is posted
Services
Email this article to a friend
Similar articles in this journal
Similar articles in ISI Web of Science
Similar articles in PubMed
Alert me to new issues of the journal
Add to My Personal Archive
Download to citation manager
Search for citing articles in:
ISI Web of Science (20)
Request Permissions Google Scholar
Articles by Chakrabarti, L.
Articles by Davies, K. E.
Search for Related Content
PubMed
PubMed Citation
Articles by Chakrabarti, L.
Articles by Davies, K. E.
Social Bookmarking
What's this?