Morpholino-mediated Knockdown of DUX4 Toward Facioscapulohumeral Muscular Dystrophy Therapeutics

Abstract

Derepression of DUX4 in skeletal muscle has emerged as a likely cause of pathology in facioscapulohumeral muscular dystrophy (FSHD). Here we report on the use of antisense phosphorodiamidate morpholino oligonucleotides to suppress DUX4 expression and function in FSHD myotubes and xenografts. The most effective was phosphorodiamidate morpholino oligonucleotide FM10, which targets the polyadenylation signal of DUX4. FM10 had no significant cell toxicity, and RNA-seq analyses of FSHD and control myotubes revealed that FM10 down-regulated many transcriptional targets of DUX4, without overt off-target effects. Electroporation of FM10 into FSHD patient muscle xenografts in mice also down-regulated DUX4 and DUX4 targets. These findings demonstrate the potential of antisense phosphorodiamidate morpholino oligonucleotides as an FSHD therapeutic option.

Figure 1

DUX4 and DUX4 target gene knockdown in facioscapulohumeral muscular dystrophy (FSHD) myotubes in vitro. (a) Schematic demonstrating DUX4-fl transcript and relative targets of phosphorodiamidate morpholino oligonucleotides (PMOs); PAS, polyadenylation signal. (b) FSHD biomarker expression analysis of 4-day FSHD myotube cultures treated with standard control or DUX4-targeting morpholinos. Significant decreases in CCNA1, MBD3L2, TRIM43, and ZSCAN4 biomarker expression were observed in myotube cultures treated with 10 umol/l FM10 (*P-value < 0.05, Student's t-test performed on ΔCts normalized to RPL13A); an intermediate effect was observed in FM9-treated cultures. MYH1 expression was not affected by PMO treatment. Data are presented as the mean fold change ± standard error of mean (SEM) relative to standard control morpholino. (c) DUX4 protein expression analysis of 4-day FSHD myotube cultures treated with standard control or DUX4-targeting morpholinos. DUX4 protein was detected by immunostaining with C-terminal DUX4-specific P4H2 antibody. The greatest decrease in DUX4-positive nuclei was observed in myotube cultures treated with 10 umol/l FM10; an intermediate effect was observed in FM9-treated cultures.

Figure 2

RNA-seq analysis of facioscapulohumeral muscular dystrophy (FSHD) versus unaffected myotube cultures and FM10 versus control phosphorodiamidate morpholino oligonucleotide (PMO)-treated myotube cultures. These volcano plots show log₂(fold-change) versus -log₁₀(P-value) for tests of differential expression of ~ 37,000 genes detected by RNA-seq, for three comparisons: (a) myotube cultures from FSHD subjects versus unaffected first-degree relatives (n = 6 pairs); (b) FM10 versus control (CTRL) PMO-treated myotube cultures from FSHD subjects (n = 2); and (c) FM10 versus control (CTRL) PMO-treated myotube cultures from unaffected subjects (n = 2). The two subjects in b are first-degree relatives of the two subjects in c, and are unrelated to the 12 subjects in a. To aid in comparisons between plots, the color-coding of points in all plots is determined by direction and significance of changes in a: genes whose expression is higher in FSHD than control myotubes are colored red if P-value < 10⁻⁴ and pink otherwise; genes whose expression is lower in FSHD than control myotubes are colored blue if P-value < 10⁻⁴ and light blue otherwise. In all plots, black squares are overlaid on the DUX4 targets from Table S1 of Yao et al. Of the 46 genes that were significantly decreased by FM10 in b, 41 had elevated expression in FSHD versus control myotubes in a, significantly so for 32; and 30 were among the DUX4 targets from Table S1 of Yao et al. Conversely, 116 of the 121 genes that were significantly elevated in FSHD versus unaffected myotubes in a, and 162 of the 185 detected DUX4 target genes from Table S1 of Yao et al. had at least some reduction by FM10 in b. P-values are from likelihood-ratio tests for negative binomial regression (R package edgeR), and prior counts of 1 were used to avoid infinite log fold-changes; see Materials and Methods for details.

Figure 3

In vivo validation of FM10 knockdown of DUX4-fl and DUX4 target genes in a human facioscapulohumeral muscular dystrophy (FSHD) xenograft model. (a) Schematic of the in vivo study design. FSHD patient muscle was transplanted into the tibialis anterior space of NRG mice. Xenografts fully regenerated in the mouse leg for more than 4 months, were treated with Control- or FM10- phosphorodiamidate morpholino oligonucleotides (PMO) via electroporation, and harvested after 2 weeks. (b) Histology of a four-month FSHD xenograft. Immunohistochemistry of the human-specific muscle membrane protein spectrin and nuclear envelope protein lamin A/C confirmed that the xenograft was well regenerated. Human spectrin (red), human lamin A/C (green), DAPI (blue). Scale bar: 200 μm. (c) Expression analysis of DUX4-fl and (d) of DUX4 target genes MBD3L5 and ZSCAN4 in xenografts from biopsy and autopsy donor groups, all with FSHD, treated with Control- and FM10-PMOs. Statistical tests were performed on ΔCts, normalized to housekeeping genes PPIA and GUSB. Significant decreases were observed for all three genes with FM10 treatment (**P < 0.01; *P < 0.05, all tests two-tailed). No correlation was detected between housekeeping gene expression and treatment. The biopsy group consisted of xenografts from three FSHD donors (n = 4 Ctrl; n = 6 FM10); to address between-donor rather than within-donor variance in response to FM10, ΔCts for xenografts from each donor were averaged, and a paired t-test performed on the per-donor averages. All xenografts in the autopsy group were from a single FSHD donor (n= 10 Ctrl; n = 6 FM10), and an unequal variance t-test was used for this data. Expression levels for each xenograft are shown as fold-changes relative to the per-donor mean ΔCt for the Ctrl PMO (2^−ΔΔCt), with mean of the per-donor averages (biopsy) or mean ± standard error of mean (SEM) (autopsy) also indicated.