September 1, 2011;
Eukaryotic initiation factor 6 (eif6) overexpression affects eye development in Xenopus laevis.
The translation initiation factor eif6
has been implicated as a regulator of ribosome assembly, selective mRNA translation and apoptosis. Many of these activities depend upon the phosphorylation of eif6
serine 235 by PKC. Previous data showed that eif6
binds to the 60S ribosomal subunit when unphosphorylated, inhibiting assembly with the 40S subunit. Phosphorylation of Ser235 releases eif6
from the 60S subunit and allows assembly. eif6
acts as an anti-apoptotic factor via regulation of the bcl2
balance and acts selectively upstream of bcl2
. This activity also depends upon phosphorylation of eif6
Ser235. One of the consequences of eif6
overexpression in Xenopus embryos is aberrant eye
development. Here we evaluate the eye
phenotype and show that it is transient. We show that the whole eye
, particularly the retina
layers, of the embryos injected with eif6
-encoding mRNA recover by stage 42
. Embryos over-expressing eif6
have normal expression of anterior
- and brain
-specific markers, indicating that outside the eye
field, other neural regions appear unaffected by the eif6
injection. No eye
defect was detected when morpholinos were used to reduce eif6
protein synthesis. We tested how two known pathways of eif6
function with respect to alteration of eye
development. We found that injection of bcl2
did not produce the eye
phenotype and eif6
co-injection did not rescue the eye
defect, suggesting that the eye
phenotype is not bearing on the anti-apoptotic role played by eif6
is not linked to its role as an anti-apoptotic factor. We also determined that PKC-dependant phosphorylation of Ser235 in eif6
is not required to produce defective eye
development. These results indicate that the aberrant eye
phenotype, produced by eif6
overexpression, is not directly linked to the PKC-regulated effects of eif6
on translation and ribosomal subunit interaction or on eif6
[+] show captions
Fig. 1. Effects of eif6 overexpression on eye formation. (a)–(f) Three different embryos injected with 400 pg eif6 mRNA and 300 pg GFP mRNA into one blastomere of the two-cell stage and harvested at stage 35. Micrographs of the two sides (a and b; c and d; e and f) of the same embryos are shown; (a), (c), and (e) are the injected sides. The eif6 overexpressors showed a range of the defective eye phenotype (arrowheads in a, c, e). The injected sides were marked by GFP fluorescence (inset in a, c, e). (g) GFP-injected side (fluorescence in the inset) and un-injected side (h) of the same embryo: overexpression of GFP alone did not affect eye development. (i)–(j) Micrographs of the two sides of the eif6-injected embryo harvested at stage 42. The eye of the injected side marked by GFP fluorescence (inset in i) recovers normal morphology. (k)–(p) Cryostat sections of stage 35 eif6-injected embryos showed a reduced eye in the injected side. Magnification of reduced eye in insets of figures k–n. (k) DAPI staining. Dasher lines border the entire thickness of neural retinas. The eye marker pax6 (l) was present in both injected and un-injected eye. However the retina layer markers showed a difference between injected and un-injected eye. Indeed, in the small injected eye otx2 (m) was diffuse and not localised in the central retina as in the un-injected eye. rbpms, (n) becomes evident only in the un-injected side. The encephalon structures were morphologically normal, forebrain (o) and midbrain (l) were labelled with pax6, hindbrain (p) was marked by egr2. (q)–(t) Cryostat sections of stage 42 eif6-injected embryos. (q) DAPI staining shows a normal stratification of retina in the injected eye compared to the control eye. (r)–(t) In situ hybridisation with rbpms, nrl, an outer nuclear layer marker and otx2, showed no difference between injected and un-injected eye, indicating that the recovered retina is structurally similar to the control retina. i.s :eif6-injected side.
Fig. 2. eif6 overexpression specifically inhibits eye development. (a)–(j) Embryos were injected with eif6 (400 pg) and βgal (400 pg) or GFP (400 pg) alone into one blastomere of the two-cell stage and harvested at the neurula stage. Whole mount in situ hybridisations were performed for rax (a), (b), pax6 (c), (d), otx2 (e), (f), six3 (g), (h) and sox2 (i), (j), fgf8 (k), (l), en2 (m), (n), egr2 (o), (p).GFP overexpression did not affect the expression of rax (a), pax6 (c), otx2 (e), six3 (g), sox2 (i), fgf8 (k), en2 (m), egr2 (o). Inset in (a) shows diffusion of GFP fluorescence in the whole injected side (i.s.) of a neurula. eif6 overexpression specifically reduced rax (b), pax6 (d), otx2 (f) and six3 (h) expression within the presumptive eye field (arrowheads), while neural expression of the same markers was unaffected (arrows). (j) No expression difference was found for the specific neural markers sox2, fgf8 , en2, egr2 between the eif6-injected and un-injected side.
Effects of eif6 downregulation. (a), (b) Serial cross-sections of stage 28 eif6-direct morpholino and GFP-injected embryo were immunostained with anti-GFP, with anti-eif6 antibodies and with TUNEL assay. The lower level of eif6 immunostaining (b) is present in the injected side (i.s.) marked by GFP (a). An increase of apoptotic nuclei is evident in the injected side (c). (d) Western blot with anti-eif6 antibody indicates that the level of protein in eif6 morpholino-injected embryos was lower than that in mispaired morpholino-injected embryos. Three sets of injected embryos were used in the blot. Normalisation was performed with anti-a-tubulin antibody. (e)–(h) Experimental and control side of the same embryo injected with eif6-direct morpholino (e), (f) or control mispaired morpholino (g), (h) in one blastomere at the two-cell stage and harvested at stage 35. No eye difference was evident between the injected side (e), (g) marked by GFP fluorescence (inset) and the un-injected side (f), (h). (i)–(n) In situ hybridisation of morpholino (i, k, m) or mispaired morpholino (j, l, n) injected embryos at the neurula stage. The expression of anterior markers rax (i), (j), pax6 (k), (l) and otx2 (m), (n) was not affected by depletion of eif6.
Fig. 4. Apoptosis reduction is not involved in the eye defect. Embryos were injected with bcl2 (300 pg) and GFP (200 pg) or GFP (400 pg) alone in one blastomere of the two-cell stage. (a) Western blot with anti-Bcl2 antibody indicates the greater presence of protein in bcl2-injected embryos with respect to GFP-injected embryos. Two sets of bcl2-injected embryos were used in the blot. Normalisation was performed with anti-a-tubulin antibody. GFP immunofluorescence (b) and TUNEL staining (c) on sections of bcl2-injected embryo. The injected side, marked by GFP fluorescence, showed fewer apoptotic nuclei compared with the control side. At the neurula stage, whole mount in situ hybridisations were performed for rax (d), (e), pax6 (f), (g) and otx2 (h), (i). bcl2 overexpression did not reduce rax (e), pax6 (g) or otx2 (i) expression in the injected side compared to the control side. (j)–(l) Embryo were injected with myc-bax (1 ng) and GFP (200 pg). The injected side, marked by GFP (j) showed an increase of TUNEL-positive nuclei (k). Western blot with anti-myc antibody indicates the presence of exogenous bax in the injected embryos (l). bax-eif6 co-injection does not restore normal expression of rax (m).
Effects of mutant S235A overexpression. Embryos injected with eif6 mutant S235A (300 pg) and GFP (200 pg) or GFP alone (400 pg) in one blastomere of the two-cell stage were harvested at stage 35 (a)–(h) or at the neurula stage (i)–(n). (a) In the Western blots using anti-β-catenin antibody, eif6 or GFP-injected embryos and two sets of S235A-injected embryos were used. A decrease in β-catenin level was found only in eif6 overexpressors. Normalisation was performed with the anti-tubulin antibody. (b) RT-PCR analysis showed that β-catenin mRNA level was unaffected in eif6 and S235A overexpressors compared with GFP-injected embryos. Normalisation was performed using histh4 primers. (c)–(h) Micrographs of the two sides (c and d; e and f; g and h) of the same stage 35 embryo are given. S235A overexpressors show a reduction in eye size (arrowheads) of the injected side (c), (e), marked by GFP fluorescence (inset), compared to the uninjected side (respectively d, f). (g), (h) Injection of GFP alone (fluorescence in inset of Figure e) did not affect eye development. (i)–(n) Whole mount in situ hybridisations were performed for rax (i), (j), pax6 (k), (l) and otx2 (m), (n) on GFP (i, k, m) or S235A (j, l, n) overexpressing embryos at neurula stage. The expression of the eye marker rax (j), pax6 (l) and otx2 (n) was inhibited in the eye field (arrows) of the S235A-injected side.