The ribosome biogenesis factor Nol11 is required for optimal rDNA transcription and craniofacial development in Xenopus

Abstract

The production of ribosomes is ubiquitous and fundamental to life. As such, it is surprising that defects in ribosome biogenesis underlie a growing number of symptomatically distinct inherited disorders, collectively called ribosomopathies. We previously determined that the nucleolar protein, NOL11, is essential for optimal pre-rRNA transcription and processing in human tissue culture cells. However, the role of NOL11 in the development of a multicellular organism remains unknown. Here, we reveal a critical function for NOL11 in vertebrate ribosome biogenesis and craniofacial development. Nol11 is strongly expressed in the developing cranial neural crest (CNC) of both amphibians and mammals, and knockdown of Xenopus nol11 results in impaired pre-rRNA transcription and processing, increased apoptosis, and abnormal development of the craniofacial cartilages. Inhibition of p53 rescues this skeletal phenotype, but not the underlying ribosome biogenesis defect, demonstrating an evolutionarily conserved control mechanism through which ribosome-impaired craniofacial cells are removed. Excessive activation of this mechanism impairs craniofacial development. Together, our findings reveal a novel requirement for Nol11 in craniofacial development, present the first frog model of a ribosomopathy, and provide further insight into the clinically important relationship between specific ribosome biogenesis proteins and craniofacial cell survival.

Fig 1. Expression of nol11 during vertebrate development.

A) Wild type nol11 expression pattern during Xenopus tropicalis development. Note the strong expression in developing neural folds (NF) and the presumptive CNC at stages 16 and 22 (lateral [left] and dorsal [right] views presented). Expression is strongly associated with the migrating and differentiating CNC at subsequent stages, and is also detected in the region of the ventral blood islands (BI) and isthmus (Is) at stage 28. BA, branchial arch; Ht, heart; Lv, liver region; Op, optic placode. B) and C) Anterior transverse dissections showing expression of nol11 in neural folds and premigratory CNC of stage 14 and 16 embryos respectively. Plane of dissection is represented by the red dotted line marked c in A. D) Horizontal dissection (shown as dotted red line marked d in A) of nol11 expression in the branchial arches of stage 28 Xenopus embryos. E) nol11 expression in E8.5, E9.5 and E10.5 wild type mouse embryos. At E8.5 expression is strongly detected in the neural folds. Transcripts are associated with CNC positive regions at both E9.5 and E10.5. BA2, 2^nd branchial arch; FNP, frontonasal prominence; Ht, heart; mdBA1, mandibular BA1; mxBA1, maxillary BA1; Op, optic placode; Ot, otic placode; T, trigeminal region.

Fig 2. The nol11 craniofacial phenotype.

A) Gross morphology and cartilage staining of UC, nol11 whole embryo, nol11 one-sided knockdowns and CMO one-sided knockdown embryos. Note the reduced cartilage size and abnormal morphology in nol11 morphants (red arrowheads) while CMO injected embryos are unaffected. B) Craniofacial cartilage size is significantly reduced in nol11 but not CMO morphants. C) Co-injection of human NOL11 RNA can rescue the cartilage phenotype in approximately 75% of treated embryos. Cartilage staining of an RNA rescued embryo; nol11 MO was injected at the one cell stage and human NOL11 RNA was injected into one cell at the two cell stage. Green arrowheads highlight rescued side.

Fig 3. Knockdown of nol11 disrupts cranial neural crest development.

At stages 14 and 24 neural and CNC development appears normal in one side treated embryos, as assayed by expression of key marker genes including sox3, cytokeratin, twist and slug. By stage 28 however, reductions are apparent in the expression of numerous CNC genes. The branchial arches are also smaller on the MO treated side at stage 28. Graph displays number of embryos exhibiting abnormal gene expression in control and nol11 morphant embryos.

Fig 4. Increased apoptosis underlies the nol11 cartilage defects.

A) nol11 knockdown results in a progressive increase in apoptosis. At stage 14 no significant difference was observed in rates of TUNEL staining between knockdown and control halves of the embryo. At stages 18 and 28 increased apoptosis was evident on the treated side of whole mount and sectioned paraffin embedded embryos. Note that this increased apoptosis occurs primarily within the craniofacial ectomesenchyme. The graph represents the relative quantification of apoptosis rates at stages 14, 18 and 28. This stage specific increase in apoptosis was confirmed by a similar increase in p53 protein levels in 1 cell injected embryos as assayed by western blot (lower right panel). Dotted red lines mark the embryonic midline. B) No significant change in proliferation rates was noted following nol11 knockdown. C) Inhibition of apoptosis by p53 MO results in a partial rescue of cartilage size and morphology. Each pair of columns in the graph compares cartilage size measured in bilateral halves of embryos. The blue pair reveals no significant difference in cartilage measurements in the left vs right side of the UC embryonic head. In the second pair (red), cartilage size is seen to be comparable on either side of the nol11 morphant head. The final pair illustrates that cartilage size is significantly improved on the side of nol11 morphants rescued with p53 MO (green) relative to the side that received nol11 MO only (red). D) Western blot demonstrating that the p53 MO efficiently reduces p53 protein levels in nol11 morphants.

Fig 5. Nol11 depletion impairs rDNA transcription and pre-rRNA processing in X. tropicalis.

A) Scheme of pre-rRNA processing pathways in X tropicalis. The pre-rRNA is transcribed by RNAPI as a 40S polycistronic precursor. Several cleavages are required to separate the mature rRNAs. The locations of oligonucleotide probes used for northern blots are indicated by lettered lines (a, c) and the cleavage sites indicated. This scheme was adapted from [–75]. B) Morpholino (MO) depletion of Nol11 impairs pre-rRNA transcription at stage 28. The northern blot was hybridized with probe a (Fig. 5A) and with a probe to the 7SL RNA as a loading control (lower panel). Bands were quantified and analysed by RAMP ([60]; S6A,B Fig) C) Morpholino (MO) depletion of Nol11 impairs pre-rRNA transcription and processing. The northern blot was hybridized with probe c (Fig. 5A) and with a probe to the 7SL RNA as a loading control (lower panel). Bands were quantified and analysed by RAMP ([60]; S6C,D, E, F Fig). D) Depletion of Nol11 leads to increased p53 levels. The expression of p53 from control and nol11 depleted embryos was analysed by western blot with anti-p53 antibodies. GAPDH levels were used as a loading control. Values for p53 expression normalized to GAPDH are represented in the bar graph. E) MO-resistant human NOL11 (hNOL11) mRNA but not p53 depletion rescues pre-rRNA levels. Embryos injected as shown by + and—in the figure at stages 22 and 28. The pre-rRNAs were visualized with probe a on a northern blot; hybridization to the 7SL RNA was used as a loading control.