Decreased Expression of CoREST1 and CoREST2 Together with LSD1 and HDAC1/2 during Neuronal Differentiation

Abstract

CoREST (CoREST1, rcor1) transcriptional corepressor together with the histone demethylase LSD1 (KDM1A) and the histone deacetylases HDAC1/2 form LSD1-CoREST-HDAC (LCH) transcriptional complexes to regulate gene expression. CoREST1 belong to a family that also comprises CoREST2 (rcor2) and CoREST3 (rcor3). CoREST1 represses the expression of neuronal genes during neuronal differentiation. However, the role of paralogs CoREST2 and CoREST3 in this process is just starting to emerge. Here, we report the expression of all CoRESTs and partners LSD1 and HDAC1/2 in two models of neuronal differentiation: Nerve-Growth-Factor (NGF)-induced neuronal phenotype of PC12 cells, and in vitro maturation of embryonic rat cortical neurons. In both models, a concomitant and gradual decrease of LSD1, HDAC1, HDAC2, CoREST1, and CoREST2, but not CoREST3 was observed. As required by the study, full-length rat rcor1 gene was identified using in silico analysis of available rat genome. The work was also complemented by the analysis of rat RNA-seq databases. The analysis showed that all CoRESTs, including the identified four splicing variants of rat CoREST3, display a wide expression in adult tissues. Moreover, the analysis of RNA-seq databases showed that CoREST2 displays a higher expression than CoREST1 and CoREST3 in the mature brain. Immunofluorescent assays and immunoblots of adult rat brain showed that all CoRESTs are present in both glia and neurons. Regarding functional partnership, CoREST2 and CoREST3 interact with all LSD1 splicing variants. In conclusion, neuronal differentiation is accompanied by decreased expression of all core components of LCH complexes, but not CoREST3. The combination of the differential transcriptional repressor capacity of LCH complexes and variable protein levels of its different components should result in a finely tuned gene expression during neuronal differentiation and in the adult brain.

Fig 1. Annotation of rat rcor1-3 transcripts.

UCSC Genome Browser images show transcript annotation data at rcor1 (A), rcor2 (B) and rcor3 (C) loci. Gene model tracks of RefSeq and Ensembl are shown in blue and red, respectively. Assembled transcripts by Cufflinks are shown in black. Predicted ELM2 and SANT1/SANT2 domains are highlighted in yellow and red, respectively. Together, the alignments of RefSeq transcripts annotation from human and mouse with the PhyloP genomic conservation scores give the evolutionary information to support the existence of the new identified exons.

Fig 2. rcor1-3 transcripts expression profile across rat tissues.

The panel shows the expression levels (measured in FPKM) of rcor1-3 genes (A) and rcor3 isoforms (B) in the different rat tissues analyzed.

Fig 3. CoREST1 and CoREST2, but not CoREST3 are down-regulated during NGF-dependent neuronal differentiation of PC12 cells.

(A) Left: representative immunoblots of CoREST1, HDAC1 and LSD1 in equivalent protein amounts of cytosolic and nuclei fractions of control (-NGF) and NGF treated (50 mg/ml during 7 days) PC12 cells. Histone H3 was used as loading control for nuclear fraction and GAPDH for cytoplasmic fraction. Right: graph of relative protein levels of CoREST1, HDAC1 and LSD1 respect to histone H3 levels. Values correspond to the mean ± SEM of at least 3 independent experiments. **p<0.01. Statistical analysis performed by Mann-Whitney test. (B) Semiquantitative RT-PCR was performed for rat rcor2 and rcor3 genes, rpl19 gene was used as reference gene. Left: representative PCR reaction. Right: Quantification of rcor2 and rcor3 mRNA expression levels in control (-NGF) and NGF treated PC12 cells. Values correspond to the mean ± SEM. **p< 0.01. Statistical analysis performed by Mann-Whitney test.

Fig 4. CoREST1 and CoREST2, but not CoREST3 decrease during cortical neurons maturation.

(A) Rat embryonic cortical neurons were immediately processed (day 0) or mantained in vitro during 2, 4 and 6 days to detect CoREST1, CoREST3, HDAC1, HDAC2 and LSD1. Values correspond to the mean ± SEM. **p<0.01, *p<0.05 (DIV 0 v/s DIV 6). Statistical analysis performed by Kruskal-Wallis test followed by Dunn’s multiple comparison test. (B) Rat embryonic cortical neurons were mantained in vitro as described above. Total RNA was extracted and semiquantitative RT-PCR was performed for rat rcor2 and rcor3 transcripts; rpl19 was used as reference gene. Values correspond to the mean ± SEM of at least 3 independent experiments. *p<0.05 (DIV 0 v/s DIV 6). Statistical analysis performed by Kruskal-Wallis test followed by Dunn’s multiple comparison test.

Fig 5. rcor2 but not rcor3 is down-regulated during brain development.

Total RNA of E.14.5 and E.18.5 embryonic rat brain, and the cortex of adult male rats were subjected to semiquantitative RT-PCR to determine rcor2 and rcor3 mRNA expression. rpl19 was used as reference gene. Values correspond to the mean ± SEM of at least 3 independent experiments. ***p< 0.001, **p<0.01, according to two-way ANOVA and Bonferroni’s posthoc test. # P<0.05, according to one-way ANOVA and Bonferroni’s posthoc test.

Fig 6. All CoRESTs are expressed in neurons and glia.

Adult rat brain sections were incubated with specific antibodies against CoREST1, CoREST2, the neuronal marker β-III tubulin and glial marker GFAP. (A) The immunostaining shows CoREST1 or CoREST2 (green), β-tubulin III (red) and Hoechst (blue, nuclear marker) counterstaining in the Hilar region of the hippocampus. Scale bar = 0 μm. (B) Confocal immunofluorescent images of hippocampus sections from male adult rat brain. The immunostaining shows CoREST1 or CoREST2 (green, as indicated) and GFAP (red). (C) CoREST3 is detected in immunoblots of total protein extracts from mice glial cells in culture and and rat brain areas as indicated.

Fig 7. CoREST2 and CoREST3 interact with all LSD1 isoforms.

HEK293-T cells were co-transfected with myc-CoREST2 (myc-CoR2) or myc-CoREST3 (myc-CoR3) and HA-LSD1 isoforms (LSD1; LSD1-2a; LSD1-8a; LSD1 2a/8a). Forty-eight hours post-transfection, cells were harvested and total protein extract was immunoprecipitated with the indicated HA antibodies (m: monoclonal; p: policlonal) or IgG (control). Immunoblots were performed with anti-myc or anti-HA antibodies, as indicated.