Figure 1.
Loci of FUS gene and structure of SVA D located upstream.
A- Schematic of loci of the FUS gene located on chromosome 16. According to UCSC genome browser (Hg18) there are several transcripts with nearly all originating at the transcriptional start site indicated in the diagram. There is a SVA D 9.9kb upstream of this transcriptional start site of the FUS gene. This SVA D is present in human and chimpanzees but not other primates. ENCODE data from the genome browser UCSC is summarised indicating the presence of DNase1 clusters, expressed sequence tags (ESTs) and histone modifications associated with enhancers and promoter at this locus. B- Schematic showing the components contributing to the structure of the SVA D located upstream of the FUS gene. It contains an Alu-like sequence, a tandem repeat (TR) consisting of 7 copies of a 37–40 bp repeat, a variable number tandem repeat (VNTR) consisting of 3–4 copies of a 37–50 bp repeat and a SINE. This particular SVA is missing the CCCTCT hexamer repeat seen at the 5′ end of a canonical SVA. The fragments cloned into the reporter gene vector (pGL3P) are shown by the black line for the SVA (length 1240/1190 bp) and the grey line for the TR/VNTR (length 862/812 bp). C- Sequence of the 7 copies of the 5′ TR and 3–4 copies of the 3′ VNTR. The repeat underlined is the additional copy found in the long allele which is absent in the short allele of the SVA (sequence in UCSC genome browser corresponds to long allele).
Figure 2.
The SVA and the VNTR within show distinct functional properties in a reporter gene construct.
Reporter gene constructs containing each allele of the FUS SVA and TR/VNTR (long and short) were transfected into the neuroblastoma cell line, SK-N-AS. The fold values of activity demonstrated by each construct compared to pGL3P normalised to the internal control (pMLuc-2) to account for differences in transfection efficiency are displayed. Both alleles when tested as a complete SVA showed repressive function and were significantly different to each other. When the alleles were tested as a smaller fragment consisting of the central TR/VNTR region they both showed enhancer properties. One tailed t-test was used to measure the significance of fold activity of the FUS SVA and TR/VNTR over the minimal promoter of the pGL3P vector *P<0.05, ***P<0.001, and to compare the activity of the alleles of the SVA and the TR/VNTR to each other # P<0.05.
Figure 3.
Demonstration of the activity of the L-SVA and L-VNTR presumptive transcriptional regulator in the chick embryo model at stage 22 HH.
Chick embryos were electroporated with either a FUS proximal promoter GFP (ppGFP) reporter construct (A–C), L-SVA ppGFP-reporter (D–F) or L-TR/VNTR ppGFP-reporter (G–I) at stage 14HH and GFP expression analysed 48 hr later (stage 22HH). Expression could not be detected in the neural tube from the FUS proximal promoter sequences alone (B), however when either the L-SVA (E) or L-TR/VNTR (H) sequences were included, GFP reporter gene expression could readily be seen. Panels A, D and G show the corresponding bright field images. Panels C, F and I show the identical fields taken with a red filter to demonstrate the extent of successful electroporation of the neural tube using a control tomato marker expression plasmid. Scale bar in B is 2 mm and in E & H is 1 mm.
Figure 4.
Genotype frequencies of SVA located upstream of the FUS gene in a SALS and matched control cohort.
A- Example image of the two alleles of the TR/VNTR of the FUS SVA run on a 1.2% agarose gel after amplification using PCR. The two alleles identified were named long (L) and short (S) and the genotype of each individual was determined. One example of each allele was gel extracted and sequenced and corresponded to the previously sequenced and cloned alleles. L = 665 bp and S = 615 bp. B-. Table showing the percentage of each genotype in the SALS patients (241 individuals) and the matched controls (228 individuals).