Ribosome profiling in mouse hippocampus: plasticity-induced regulation and bidirectional control by TSC2 and FMRP

BACKGROUND

Mutations in TSC2 are the most common cause of tuberous sclerosis (TSC), a disorder with a high incidence of autism and intellectual disability. TSC2 regulates mRNA translation required for group 1 metabotropic glutamate receptor-dependent synaptic long-term depression (mGluR-LTD) and behavior, but the identity of mRNAs responsive to mGluR-LTD signaling is largely unknown.

METHODS

We utilized Tsc2+/- mice as a mouse model of TSC and prepared hippocampal slices from these animals. We induced mGluR-LTD synaptic plasticity in slices and processed the samples for RNA-seq and ribosome profiling to identify differentially expressed genes in Tsc2+/- and following mGluR-LTD synaptic plasticity.

RESULTS

Ribosome profiling reveals that in Tsc2+/- mouse hippocampal slices, the expression of several mRNAs was dysregulated: terminal oligopyrimidine (TOP)-containing mRNAs decreased, while FMRP-binding targets increased. Remarkably, we observed the opposite changes of FMRP binding targets in Fmr1-/y hippocampi. In wild-type hippocampus, induction of mGluR-LTD caused rapid changes in the steady-state levels of hundreds of mRNAs, many of which are FMRP targets. Moreover, mGluR-LTD failed to promote phosphorylation of eukaryotic elongation factor 2 (eEF2) in TSC mice, and chemically mimicking phospho-eEF2 with low cycloheximide enhances mGluR-LTD in TSC mice.

CONCLUSION

These results suggest a molecular basis for bidirectional regulation of synaptic plasticity and behavior by TSC2 and FMRP. Our study also suggests that altered mGluR-regulated translation elongation contributes to impaired synaptic plasticity in Tsc2+/- mice.

Regulation of the cardiomyocyte transcriptome vs translatome by endothelin-1 and insulin: translational regulation of 5' terminal oligopyrimidine tract (TOP) mRNAs by insulin

BACKGROUND

Changes in cellular phenotype result from underlying changes in mRNA transcription and translation. Endothelin-1 stimulates cardiomyocyte hypertrophy with associated changes in mRNA/protein expression and an increase in the rate of protein synthesis. Insulin also increases the rate of translation but does not promote overt cardiomyocyte hypertrophy. One mechanism of translational regulation is through 5' terminal oligopyrimidine tracts (TOPs) that, in response to growth stimuli, promote mRNA recruitment to polysomes for increased translation. TOP mRNAs include those encoding ribosomal proteins, but the full panoply remains to be established. Here, we used microarrays to compare the effects of endothelin-1 and insulin on the global transcriptome of neonatal rat cardiomyocytes, and on mRNA recruitment to polysomes (i.e. the translatome).

RESULTS

Globally, endothelin-1 and insulin (1 h) promoted >1.5-fold significant (false discovery rate < 0.05) changes in expression of 341 and 38 RNAs, respectively. For these transcripts with this level of change there was little evidence of translational regulation. However, 1336 and 712 RNAs had >1.25-fold significant changes in expression in total and/or polysomal RNA induced by endothelin-1 or insulin, respectively, of which approximately 35% of endothelin-1-responsive and approximately 56% of insulin-responsive transcripts were translationally regulated. Of mRNAs for established proteins recruited to polysomes in response to insulin, 49 were known TOP mRNAs with a further 15 probable/possible TOP mRNAs, but 49 had no identifiable TOP sequences or other consistent features in the 5' untranslated region.

CONCLUSIONS

Endothelin-1, rather than insulin, substantially affects global transcript expression to promote cardiomyocyte hypertrophy. Effects on RNA recruitment to polysomes are subtle, with differential effects of endothelin-1 and insulin on specific transcripts. Furthermore, although insulin promotes recruitment of TOP mRNAs to cardiomyocyte polysomes, not all recruited mRNAs are TOP mRNAs.

All translation elongation factors and the e, f, and h subunits of translation initiation factor 3 are encoded by 5'-terminal oligopyrimidine (TOP) mRNAs

Terminal oligopyrimidine (TOP) mRNAs (encoded by the TOP genes) are identified by a sequence of 6-12 pyrimidines at the 5' end and by a growth-associated translational regulation. All vertebrate genes for the 80 ribosomal proteins and some other genes involved, directly or indirectly, in translation, are TOP genes. Among the numerous translation factors, only eEF1A and eEF2 are known to be encoded by TOP genes, most of the others having not been analyzed. Here, we report a systematic analysis of the human genes for translation factors. Our results show that: (1) all five elongation factors are encoded by TOP genes; and (2) among the initiation and termination factors analyzed, only eIF3e, eIF3f, and eIF3h exhibit the characteristics of TOP genes. Interestingly, these three polypeptides have been recently shown to constitute a specific subgroup among eIF3 subunits. In fact, eIF3e, eIF3f, and eIF3h are the part of the functional core of eIF3 that is not conserved in Saccharomyces cerevisiae. It has been hypothesized that they are regulatory subunits, and the fact that they are encoded by TOP genes may be relevant for their function.

AMPK represses TOP mRNA translation but not global protein synthesis in liver

The AMP-activated protein kinase (AMPK) represses signaling through the mammalian target of rapamycin complex 1 (mTORC1). In muscle, repression of mTORC1 leads to a reduction in global protein synthesis. In contrast, repression of mTORC1 in the liver has no immediate effect on global protein synthesis. In the present study, signaling through mTORC1 and translation of specific mRNAs such as those bearing a 5'-terminal oligopyrimidine (TOP) tract and were examined in rat liver following activation of AMPK after treadmill running. Activation of AMPK repressed translation of the TOP mRNAs encoding rpS6, rpS8, and eEF1alpha. In contrast, neither global protein synthesis nor translation of mRNAs encoding GAPDH or beta-actin was changed. Basal phosphorylation of the mTORC1 target 4E-BP1, but not S6K1 or rpS6, was reduced following activation of AMPK. Thus, in liver, AMPK activation repressed translation of TOP mRNAs through a mechanism distinct from downregulated phosphorylation of S6K1 or rpS6.

Chordin expression, mediated by Nodal and FGF signaling, is restricted by redundant function of two beta-catenins in the zebrafish embryo

Using embryos transgenic for the TOP-GFP reporter, we show that the two zebrafish beta-catenins have different roles in the organizer and germ-ring regions of the embryo. beta-Catenin-activated transcription in the prospective organizer region specifically requires beta-catenin-2, whereas the ventrolateral domain of activated transcription is abolished only when both beta-catenins are inhibited. chordin expression during zebrafish gastrulation has been previously shown in both axial and paraxial domains, but is excluded from ventrolateral domains. We show that this gene is expressed in paraxial territories adjacent to the domain of ventrolateral beta-catenin-activated transcription, with only slight overlap, consistent with the now well-known inhibitory effects of Wnt8 on dorsal gene expression. Eliminating both Wnt8/beta-catenin signaling and organizer activity by inhibition of expression of the two beta-catenins results in massive ectopic circumferential expression of chordin and later, by formation of a distinctive embryonic phenotype ('ciuffo') that expresses trunk and anterior neural markers with correct relative anteroposterior patterning. We show that chordin expression is required for this neural gene expression. The Nodal gene squint has been shown to be necessary for optimal expression of chordin and is sufficient in some contexts for its expression. However, chordin is not normally expressed in the ventrolateral germ-ring despite robust expression of squint in this domain. We show the ectopic circumferential expression of chordin and other dorsal genes to be completely dependent on Nodal and FGF signaling, and to be independent of a functional organizer. We propose that whereas the axial domain of chordin expression is formed by cells that are derived from the organizer, the paraxial domain is the result of axial-derived anti-Wnt signals, which relieve the repression that otherwise is set by the Wnt8/beta-catenin/vox,vent pathway on latent germ-ring Nodal/FGF-activated expression.

Tuberous sclerosis complex proteins 1 and 2 control serum-dependent translation in a TOP-dependent and -independent manner

The tuberous sclerosis complex (TSC) proteins TSC1 and TSC2 regulate protein translation by inhibiting the serine/threonine kinase mTORC1 (for mammalian target of rapamycin complex 1). However, how TSC1 and TSC2 control overall protein synthesis and the translation of specific mRNAs in response to different mitogenic and nutritional stimuli is largely unknown. We show here that serum withdrawal inhibits mTORC1 signaling, causes disassembly of translation initiation complexes, and causes mRNA redistribution from polysomes to subpolysomes in wild-type mouse embryo fibroblasts (MEFs). In contrast, these responses are defective in Tsc1(-/-) or Tsc2(-/-) MEFs. Microarray analysis of polysome- and subpolysome-associated mRNAs uncovered specific mRNAs that are translationally regulated by serum, 90% of which are TSC1 and TSC2 dependent. Surprisingly, the mTORC1 inhibitor, rapamycin, abolished mTORC1 activity but only affected approximately 40% of the serum-regulated mRNAs. Serum-dependent signaling through mTORC1 and polysome redistribution of global and individual mRNAs were restored upon re-expression of TSC1 and TSC2. Serum-responsive mRNAs that are sensitive to inhibition by rapamycin are highly enriched for terminal oligopyrimidine and for very short 5' and 3' untranslated regions. These data demonstrate that the TSC1/TSC2 complex regulates protein translation through mainly mTORC1-dependent mechanisms and implicates a discrete profile of deregulated mRNA translation in tuberous sclerosis pathology.

S6K1(-/-)/S6K2(-/-) mice exhibit perinatal lethality and rapamycin-sensitive 5'-terminal oligopyrimidine mRNA translation and reveal a mitogen-activated protein kinase-dependent S6 kinase pathway

Activation of 40S ribosomal protein S6 kinases (S6Ks) is mediated by anabolic signals triggered by hormones, growth factors, and nutrients. Stimulation by any of these agents is inhibited by the bacterial macrolide rapamycin, which binds to and inactivates the mammalian target of rapamycin, an S6K kinase. In mammals, two genes encoding homologous S6Ks, S6K1 and S6K2, have been identified. Here we show that mice deficient for S6K1 or S6K2 are born at the expected Mendelian ratio. Compared to wild-type mice, S6K1(-/-) mice are significantly smaller, whereas S6K2(-/-) mice tend to be slightly larger. However, mice lacking both genes showed a sharp reduction in viability due to perinatal lethality. Analysis of S6 phosphorylation in the cytoplasm and nucleoli of cells derived from the distinct S6K genotypes suggests that both kinases are required for full S6 phosphorylation but that S6K2 may be more prevalent in contributing to this response. Despite the impairment of S6 phosphorylation in cells from S6K1(-/-)/S6K2(-/-) mice, cell cycle progression and the translation of 5'-terminal oligopyrimidine mRNAs were still modulated by mitogens in a rapamycin-dependent manner. Thus, the absence of S6K1 and S6K2 profoundly impairs animal viability but does not seem to affect the proliferative responses of these cell types. Unexpectedly, in S6K1(-/-)/S6K2(-/-) cells, S6 phosphorylation persisted at serines 235 and 236, the first two sites phosphorylated in response to mitogens. In these cells, as well as in rapamycin-treated wild-type, S6K1(-/-), and S6K2(-/-) cells, this step was catalyzed by a mitogen-activated protein kinase (MAPK)-dependent kinase, most likely p90rsk. These data reveal a redundancy between the S6K and the MAPK pathways in mediating early S6 phosphorylation in response to mitogens.

Alternative patterns of transcription and translation of the ribosomal protein L32 mRNA in somatic and spermatogenic cells in mice

The patterns of transcription and translation of the ribosomal protein L32 (Rpl32) mRNA differ greatly in adult testis and somatic tissues. Northern blots reveal that the levels of Rpl32 mRNA are four- to five-fold higher in prepubertal and adult testes, and purified pachytene spermatocytes and round spermatids than in a variety of nongrowing adult somatic tissues. 5' RACE demonstrates that transcription in 8-day prepubertal testis, which lacks meiotic and haploid cells, strongly prefers the same start site in the 5' terminal oligopyrimidine tract (5' TOP) that is used is somatic cells. The 5' TOP is a cis element that inhibits translation of many mRNAs in nongrowing somatic cells. Although the sizes of deadenylated Rpl32 mRNAs are indistinguishable in somatic and spermatogenic cells, transcription initiates at 11 sites over a 31-nt segment in adult testis and approximately 62% of Rpl32 mRNAs lack a 5' TOP. In agreement with previous studies, low levels of cycloheximide increase the proportions and sizes of polysomes in absorbance profiles, and increase the proportions and sizes of polysomes translating four 5' TOP mRNA species including the Rpl32 mRNA in 8-day seminiferous tubules. In contrast, cycloheximide has little or no effect on the absorbance profiles and distribution of Rpl32 mRNA and 5' TOP mRNAs in adult seminiferous tubules. The failure of cycloheximide to increase the size of polysomes in adult seminiferous tubules implies a block in the pathway by which ribosomes are recruited onto translationally active mRNAs.

Translation of both 5'TOP and non-TOP host mRNAs continues into the late phase of Baculovirus infection

Complete cDNA sequences were obtained for ribosomal protein (rp) L15 and eukaryotic initiation factor eIF2alpha from the lepidopteran insect Spodoptera frugiperda, and for elongation factor eEF2 from S. exigua. The presence of a 5' terminal oligopyrimidine (TOP) tract classified the lepidopteran rpL15 transcript as a TOP mRNA. For eEF2, two types of transcripts were observed, one of which had a 5'TOP tract. The transcript levels for rpL15, eEF2 and eIF2alpha decreased following baculovirus infection. Polysome analysis showed that the corresponding mRNAs remained to be translated until at least 16 h post-infection for both TOP and non-TOP mRNAs. Baculovirus-induced host shut-off therefore appears to be regulated at the level of RNA abundance rather than at the translational level.