High expression of nuclear factor 90 (NF90) leads to mitochondrial degradation in skeletal and cardiac muscles

Abstract

While NF90 has been known to participate in transcription, translation and microRNA biogenesis, physiological functions of this protein still remain unclear. To uncover this, we generated transgenic (Tg) mice using NF90 cDNA under the control of β-actin promoter. The NF90 Tg mice exhibited a reduction in body weight compared with wild-type mice, and a robust expression of NF90 was detected in skeletal muscle, heart and eye of the Tg mice. To evaluate the NF90 overexpression-induced physiological changes in the tissues, we performed a number of analyses including CT-analysis and hemodynamic test, revealing that the NF90 Tg mice developed skeletal muscular atrophy and heart failure. To explore causes of the abnormalities in the NF90 Tg mice, we performed histological and biochemical analyses for the skeletal and cardiac muscles of the Tg mice. Surprisingly, these analyses demonstrated that mitochondria in those muscular tissues of the Tg mice were degenerated by autophagy. To gain further insight into the cause for the mitochondrial degeneration, we identified NF90-associated factors by peptide mass fingerprinting. Of note, approximately half of the NF90-associated complexes were ribosome-related proteins. Interestingly, protein synthesis rate was significantly suppressed by high-expression of NF90. These observations suggest that NF90 would negatively regulate the function of ribosome via its interaction with the factors involved in the ribosome function. Furthermore, we found that the translations or protein stabilities of PGC-1 and NRF-1, which are critical transcription factors for expression of mitochondrial genes, were significantly depressed in the skeletal muscles of the NF90 Tg mice. Taken together, these findings suggest that the mitochondrial degeneration engaged in the skeletal muscle atrophy and the heart failure in the NF90 Tg mice may be caused by NF90-induced posttranscriptional repression of transcription factors such as PGC-1 and NRF-1 for regulating nuclear-encoded genes relevant to mitochondrial function.

Figure 1. Structure of the NF90 transgene and body growth rates of wild-type (WT) and NF90 transgenic (Tg) mice.

(A) NF90 transgene construct used to generate transgenic mice. (B) Growth curves of WT and two lines NF90 Tg male (n = 9) and female (n = 7) mice (lines TG1 and TG2) from the age of 5 weeks through 15 weeks. Data are expressed as means ± SD. *, p < 0.0001 relative to WT by a two-tailed Student’s t test. (C) Photograph of whole body from WT and NF90 Tg male mice (line TG1) at 12 weeks of age.

Figure 2. NF90 Tg mice display muscular atrophy.

(A) Lateral view of hindlimbs from WT and NF90 Tg mice (line TG1) at 12 weeks of age. An arrow indicates muscular atrophy in the NF90 Tg mice. (B) Axial computed tomography images of the center of distal hindlimbs in WT and NF90 Tg mice (line TG1) at 12 weeks of age. (C and D) Comparison of volumes of triceps surae muscles from left (C) and right (D) hindlimbs between WT and NF90 Tg mice (line TG1) at 12 weeks of age. The muscle volumes were measured by using X-ray computed tomography. Data are expressed as means ± SD (n = 5). *, p < 0.005 relative to WT by a two-tailed Student’s t test. (E) Peak force measurements (g) of grip strength of WT and NF90 Tg mice (line TG1) at 16 weeks of age. Data are expressed as means ± SD (n = 5). *, p < 0.001 relative to WT by a two-tailed Student’s t test.

Figure 3. NF90 Tg mice display heart failure.

(A to E) Measurement of heart rate, blood pressure (BP) and brain natriuretic peptide (BNP) of WT and NF90 Tg mice (line TG1) at 14 weeks to 19 weeks of age. Heart rate and BP of mice were measured by a programmable sphygmomanometer using tail-cuff method. (A) HR, heart rate; (B) SBP, systolic blood pressure; (C) DBP, diastolic blood pressure; (D) MBP, mean blood pressure. All data are expressed as means ± SD (n = 7 per group). *, p < 0.01 relative to WT by a two-tailed Student’s t test. (E) The expression of BNP in NF90 Tg mice. Total RNAs isolated from cerebellum, heart and kidney of WT and NF90 Tg mice (line TG1) were analyzed by RT-PCR with specific primers for BNP or HPRT. HPRT was used as an internal control. W-1 and W-2, wild-type; T-1 and -2, NF90 Tg mice (line TG1).

Figure 4. Skeletal and cardiac muscles of NF90 Tg mice exhibit mitochondrial degradation.

(A) Haematoxylin and eosin (HE)-stained sections of skeletal (I to III) and cardiac (VI to IV) muscles from WT (I and VI) and NF90 Tg mice (line TG1: II and V, line TG2: III and IV) at 15 to 17 weeks of age. Arrows highlight vacuolations. Scale bars show 10 mm at the inset. (B) Transmission electron microscopy analysis of skeletal (I and II) and cardiac (III and IV) muscles from WT (I and III) and NF90 Tg mice (II: line TG2, IV: line TG1) at 13 to 17 weeks of age. Arrows indicate degenerating mitochondria.

Figure 5. Autophagy is induced in the skeletal muscles of the NF90 Tg mice.

Immunoblot analysis of LC3A and LC3B in tissue extractions of skeletal muscles of WT (n = 6) and NF90 Tg mice (line TG1) (n = 4). Anti-α-tubulin was used as loading control. W-1 to -6, WT; T-1 to -4, NF90 Tg mice (line TG1).

Figure 6. NF90 represses protein synthesis rate.

(A) Isolation of NF90-associated complexes. The resulting peptide mass fingerprints were compared to those in a data base, and the identity of each protein is shown to the right of the gel. The molecular markers are indicated on the left in kilodaltons (kDa). (B and C) Protein synthesis rate in untransfected (C) and Flag-NF90-HEK 293 stable cells (S1, S2, and S3) (B) and primary cells of skeletal muscles from WT (n = 2) and NF90 Tg mice (line TG1) (n = 3) at 11 to 13 weeks of age (C). Protein synthesis was measured in cells incubated in medium containing [³H]methionine. The DNA amounts in the cells used in this assay were quantitated by the DABA method for normalization of the data. Data are expressed as [³H]methionine incorporated (DPM/DNA amount corresponding to 10⁵ cells) and are expressed as means ± SD of three independent experiments. *, p < 0.01 relative to control by a two-tailed Student’s t test.

Figure 7. The translation of PGC-1 is depressed in the skeletal muscles of the NF90 Tg mice.

(A) Immunoblot analysis of PGC-1 in tissue extractions of skeletal muscles of WT (n = 6) and NF90 Tg mice (line TG1) (n = 4). Anti-α-tubulin was used as loading control. W-1 to -6, WT; T-1 to -4, NF90 Tg mice (line TG1). Intensities of specific bands in imunoblotting analysis were measured with a densitometer and are presented as a graph. Data are expressed as means ± SD. *, p < 0.01 relative to WT by a two-tailed Student’s t test. (B) RNAs isolated from skeletal muscles of WT (n = 4) and NF90 Tg mice (line TG1) (n = 3) were analyzed for expressions of PGC-1α and PGC-1β by RT-qPCR. HPRT was used as an internal control and for normalization of the data. Data are expressed as means ± SD.

Figure 8. Expression analysis of mitochondria-related proteins, COX-2, COX-4 and NRF-1 in the skeletal muscles of WT and NF90 Tg mice.

(A) RNAs isolated from skeletal muscles of WT (n = 6) and NF90 Tg mice (line TG1) (n = 4) were analyzed for expressions of COX-2, COX-4 and NRF-1 by RT-qPCR. HPRT was used as an internal control and for normalization of the data. Data are expressed as means ± SD. *, p < 0.01 relative to WT by a two-tailed Student’s t test. (B) Immunoblot analysis of NRF-1 in tissue extractions of skeletal muscles of WT (n = 6) and NF90 Tg mice (line TG1) (n = 4). Anti-α-tubulin was used as loading control. W-1 to -6, WT; T1-1 to -4, NF90 Tg mice (line TG1). Intensities of specific bands in imunoblotting analysis were measured with a densitometer and are presented as a graph. Data are expressed as means ± SD. *, p < 0.01 relative to WT by a two-tailed Student’s t test.