Frataxin mRNA isoforms in FRDA patients and normal subjects: effect of tocotrienol supplementation

Abstract

Friedreich's ataxia (FRDA) is caused by deficient expression of the mitochondrial protein frataxin involved in the formation of iron-sulphur complexes and by consequent oxidative stress. We analysed low-dose tocotrienol supplementation effects on the expression of the three splice variant isoforms (FXN-1, FXN-2, and FXN-3) in mononuclear blood cells of FRDA patients and healthy subjects. In FRDA patients, tocotrienol leads to a specific and significant increase of FXN-3 expression while not affecting FXN-1 and FXN-2 expression. Since no structural and functional details were available for FNX-2 and FXN-3, 3D models were built. FXN-1, the canonical isoform, was then docked on the human iron-sulphur complex, and functional interactions were computed; when FXN-1 was replaced by FXN-2 or FNX-3, we found that the interactions were maintained, thus suggesting a possible biological role for both isoforms in human cells. Finally, in order to evaluate whether tocotrienol enhancement of FXN-3 was mediated by an increase in peroxisome proliferator-activated receptor-γ (PPARG), PPARG expression was evaluated. At a low dose of tocotrienol, the increase of FXN-3 expression appeared to be independent of PPARG expression. Our data show that it is possible to modulate the mRNA expression of the minor frataxin isoforms and that they may have a functional role.

Figure 1

(a) A representative scheme of frataxin isoform cDNAs and primer sequences. Frataxin isoform 1 (FXN-1) and isoform 2 (FXN-2) share the left primer sequence, while the right primer was specific for each isoform. Frataxin isoform 3 (FXN-3) right primer sequence was designed in the region downstream the 702 base pair. (b) Enzymatic digestion of FXN-1 and FXN-2 PCR products. A representative polyacrylamide gel electrophoresis (10%) of FXN-1 and FXN-2 PCR products digested or undigested with BstNI restriction enzyme. Lane 1: DNA ladder; lane 2: undigested FXN-1 PCR product; Lane 3: BstNI digested FXN-1 PCR product; lane 4: undigested FXN-2 PCR product; lane 5: BstNI digested-FXN-2 PCR product.

Figure 2

(a) Semiquantitative PCR analysis of frataxin isoforms in a control subject. Equal amount of a control subject cDNA was used for FXN-1, FXN-2, and FXN-3 semiquantitative PCR analysis. PCR products were run in an agarose gel electrophoresis (2.5%). M: DNA ladder; NC: negative control. (b) qRT-PCR expression of frataxin isoforms. Frataxin isoform mRNA expression was analyzed by qRT-PCR in five healthy subjects (Ctr) and in five FRDA patients before (white columns) and after (grey columns) two-month tocotrienol supplementation. Data were normalized for two housekeeping genes, beta-actin and GAPDH; for each gene target, the normalized expression value of one control subject arbitrarily chosen was set to 1, and all other gene expression data were reported to that sample. Data are expressed as mean of technical triplicates ± SD and analyzed by ANOVA. Following tocotrienol supplementation, FXN-3 mRNA increased 3.49-fold in FRDA patients (p ≤ 0.00000000342).

Figure 3

qRT-PCR expression of PPARG. Peroxisome proliferator-activated receptors-γ (PPARG) mRNA expression was analyzed by qRT-PCR in five healthy subjects (Ctr) and in five FRDA patients before (white columns) and after (grey columns) two-month tocotrienol supplementation. Data were normalized for two housekeeping genes, beta-actin and GAPDH; for each gene target, the normalized expression value of one control subject arbitrarily chosen was set to 1, and all other gene expression data were reported to that sample. Data are expressed as mean of technical triplicates ± SD and analyzed by ANOVA.

Figure 4

Structural comparison of human FXN-1 with FXN-2 and FXN-3. Sequence and structural alignment of FXN-1 (1EKG, depicted in blue) with the computed 3D models of FXN-2 and FXN-3 (colored in green and red, resp.). The secondary structure of FXN-1 is also reported along the structural alignment (helices: red line; beta sheets: yellow arrow). Functional residues of the anionic patch and residues involved in protein stability are highlighted in black and orange, respectively, and are conserved in all the isoforms.

Figure 5

Docking of human FXN-1, FXN-2, and FXN-3 on the human tetrameric NFS1/ISCU complex. (a) The human iron-sulfur assembly complex was modeled adopting the E. coli counterpart as a template (PDB code: 3LVL). The NFS1/ISCU complex includes two copies of ISCU (surface representation in purple) and two copies of NFS1 (surface in cyan and yellow, resp.). The backbone of the three frataxin isoforms is depicted in blue (FXN-1), in green (FXN-2), and in red (FXN-3). The three backbones overlap with a pairwise root mean square deviation (RSMD) of about 1.3 Å. (b) Important residues essential for functional interaction are conserved in all isoforms and highlighted in black. Relevant interactions on the complex are highlighted in orange. Basic residues of NFS1 involved in the interaction with the anionic patch of frataxin (R218, R219, R221, and R223) are colored in light green. (c) Zooming on conserved residues among human frataxins was experimentally proven in yeast and humans to play a critical role in the interaction with the ISCU complex: W155 N146 and R165 (R161 in FXN-2 and FXN-3).