June 1, 2000;
Beta-catenin signaling activity dissected in the early Xenopus embryo: a novel antisense approach.
Xenopus embryos develop dorsal/ventral
axes as a result of the activity of a maternal Xwnt pathway, in which beta-catenin is an essential component, acting as a transactivator of transcription of zygotic genes. However, the questions of where and when beta-catenin is required in early embryogenesis have not been addressed directly, because no loss-of-function method has been available. Here we report the use of a novel antisense approach that allows us to target depletion of protein to individual blastomeres. When a "morpholino" oligo complementary to beta-catenin mRNA is injected into early embryos, it depletes beta-catenin protein effectively through the neurula stage
. By targeting the oligo to different cleavage
blastomeres, we block beta-catenin activity in different areas and at different times. Dorsal vegetal injection at the 2- and 4-cell stages blocks dorsal axis formation and at the 8-cell stage
formation, while A-tier injection at the 32-cell stage
causes abnormal cement gland
formation. This approach shows the complex involvement of Xwnt pathways in embryonic patterning and offers a rapid method for the functional analysis of both maternal and early zygotic gene products in Xenopus.
[+] show captions
FIG. 1. The effectiveness of beta-catenin MO in causing ventralization and reducing beta-catenin protein levels. 6 ng of morpholino oligo (Mo)
was injected into two cells (3 ng each) at the two-cell stage. This caused a ventralized phenotype compared to uninjected (Uninj) controls.
(B) Real-time RT PCR analysis of sibling embryos of those in A, for molecular markers of dorsal development siamois, Xnr3, and chordin
and the ventral marker Xwnt 8. The level of expression of these markers is compared in control, MO iniected , and MO + beta-catenin mRNA
at the early gastrula stage (see Materials and Methods for details). While dorsal markers were reduced by MO injection, the ventral marker
-was increased. When MO was injected together with beta-catenin mRNA, the ventralized phenotype was reversed and dorsal markers were
hyperexpressed. (C) When MO was injected together with 500 pg beta-catenin mRNA, the ventralized phenotype was reversed and dorsalized
embryos resulted (Mo + RNA). (D) Northern blotting shows that the morpholino oligo injected at 6 and 12 ng did not degrade beta-catenin mRNA at the blastula or neurula stages. A probe for plakoglobin mRNA was used as a loading control. (E) The degree of depletion of
beta-catenin protein at early gastruta (stage 10) and tailbud (stage 24) stages is compared in a Western blot. When the oligo was injected into
both cells of the two-cell stage embryo, or into the two dorsal cells of the four~cell stage embryo, it caused a significant reduction in total
beta-catenin protein at the early gastrula stage but not at the tailbud stage. ~-Tubulin is used as a loading control, cs, cell stage; dors, dorsal
injection. IF) MO caused a greater depletion of beta-catenin protein at the neurula stage than at the blastula stage and reduced protein levels
in both membrane and cytosolic fractions. Sibling embryos of those shown in A were Western blotted in pellet and supernatant fractions
at the stages shown. Significant reduction of protein levels occurred in both fractions compared to the uninjected control.
FIG. 2. beta-Catenin requirement differs for dorsoventral and anterior patterning. 6 ng of MO was injected at the four-cell stage either into
two dorsal or into two ventral blastomeres. While injections into dorsal blastomeres caused ventralization, injection into ventral
blastomeres caused ventralization in few cases. (B) 6 ng of MO was injected into two cells of each of the four quadrants of the embryo.
Injection into dorsal but not ventral quadrants caused embryos to develop with reduced heads but did not prevent the formation of
dorsoventral axes and tail structures. (C) Sibling embryos of those shown in B were analyzed by RT-PCR and compared for the degree of
expression of dorsal marl, ers at the early gastrula stage (stage 10). Siamois and Xnr3 were unaffected when ventral animal or vegetal cells
were targeted, but reduced after dorsal injections of MO. Xwnt8 was increased after dorsal MO injection. ODC, ornithine decarboxylase
used as a control for the efficiency of RNA synthesis in each sample. (D} One batch of embryos was injected with 10 ng of MO at the two-,
four-, or eight-cell stage, targeting the dorsal vegetal area. Embryos were analyzed at the early gastrula stage for dorsal markers siamois,
Xnr3, and chordin and head markers Xhex and cerberus. ODC was used as control for the efficiency of RNA synthesis in each sample.
Embryos injected with MO at the two-cell stage have both dorsal and head markers depleted to below 10% of the uninjected control level
at the gastrula stage. Eight-cell stage injections result in head marker Xhex and cerberus depletion~ while siamois, chordin, and Xnr3 are
FIG. 3. The phenotype caused by beta-catenin depletion at the eight-cell stage is less extreme than that caused by injection of the same dose
in the same volume at the four-cell stage, 10 ng of oligo was injected either into two dorsal cells at the four~cell stage or into four dorsal
cells at the eight-cell stage. Injection into two dorsal cells at the four-cell stage caused complete ventralization in all cases (B); injections
into two dorsal animal (C), two dorsal vegetal (D), or two dorsal animal + two dorsal vegetal cells (E), at the eight-cell stage, caused mainly
reduced heads, cs, cell stage; dots/an, dorsal animal cells; dors/veg, dorsal vegetal cells.
FIG. 4. Depletion of beta-catenin in animal cells at the 32-cell stage causes excessive anterior neural development. (A) Pairs of embryos at
the tailbud stage that received 15-20 ng of MO. MO was injected into 32-celt stage embryos, targeting either four dorsal (upper right) or
ventral (lower right) animal cells (A tier) or four C-tier cells (upper left). The lower left pair received 15 ng of otigo plus 100 pg of beta-catenin
mRNA. Injection of the C tier has no effect on cement gland development, while dorsal or ventral animal cell injection results in embryos with normal dorsal axes but enlarged and ectopic cement glands. The phenotype is rescued by the co-injection of beta-catenin mRNA along
with MO. IB] RT PCR analysis of sibling embryos of those shown in A analyzed at the gastruta stages for anterior neural markers Xagl and
OTX2 and the geueral neural marker, Xnrpl. Notice that X~rpl is expressed maternally as well as zygotically and increases in level due
to new zygotic transcription by the late gastrula stage, m uninjected controls. Embryos injected in dorsal or ventral animal cells with MO
at the 32-cell stage overexpress all three markers compared to controls. Injections into these cells also causes a 40-50% reduction in the
level of expression of Xnr3 at the early gastrula stage and enhances Xhex expression. ODC, ornithine decarboxylase is used as a control for
the efficiency of mRNA synthesis in each sample.