Wnt1 morpholino induces Tp53-dependent apoptosis. Uninjected (A, D, G, J), wnt1 MO-injected (B, E, H, K), and wnt1 + tp53 MO-injected (C, F, I, L) embryos were stained to detect dying cells (A–C, TUNEL, green channel), segment-specific marker EphA4 (A–C, red channel), and Tp53 transcriptional targets tp53 (D–F), p21 (G–I), and mdm2 (J–L). Wnt1 MO-injected embryos show increased cell death (compare A and B, green channel), which is rescued when tp53 MO is co-injected (compare B and C). Tp53 targets are induced in wnt1 MO-injected embryos compared to uninjected controls (compare E, H, K with D, G, J), and this expression is restored to control levels when tp53 MO is co-injected (compare E, H, K and F, I, L).

Tp53 knockdown rescues wnt1 MO phenotypes. In situ hybridization was performed to detect deltaA (A, H, O), deltaB (B, I, P), neurog1 (C, J, Q), ascl1a (D, K, R), ascl1b (E, L, S), rfng (F, M, T), and sema3gb (G, N, U) in embryos injected with control (A–G), wnt1 (H–N), and wnt1 + tp53 (O–U) morpholinos. Loss of expression of deltaA, deltaB, neurog1, ascl1a, and ascl1b in the hindbrain of wnt1 morphant embryos (compare A–E with H–L) is rescued when Tp53 is knocked down (compare H–L with O–S). Expanded expression of boundary markers rfng and sema3gb in wnt1 morphant embryos (compare F and G with M and N) is also rescued by Tp53 knockdown (compare M and N with T and U).

Proneural genes are not required to restrict boundary marker expression to rhombomere interfaces. Embryos injected with control (A–C), neurog1 + ascl1a + ascl1b (triple, D–F) or triple + tp53 (G–I) morpholinos were hybridized with RNA probes to deltaA (A, D, G), deltaB (B, E, H), and rfng (C, F, I). Effective proneural gene knockdown is inferred from the decrease in downstream genes deltaA and deltaB in triple MO-injected embryos (compare A and B with G and H). Boundary marker rfng expression does not expand to non-boundary regions in the absence of proneural genes when cell death is blocked (compare C and I). The expansion of rfng expression in triple proneural gene knockdown alone (compare C and F) is therefore dependent on Tp53 activity and/or apoptosis (compare F and I).

Pro-apoptotic Bcl protein activity is necessary and sufficient for ectopic expression of boundary markers. 16ss embryos treated for 2.5 h with 20 μM Bcl2 inhibitor HA14-1 undergo extensive apoptosis (anti-active caspase3 staining in green channel, compare A and D), and express ectopic rfng (compare B and E). Ectopic rfng expression induced by HA14-1 treatment is still present in HA14-1-treated tp53 morphants (compare E and F), demonstrating that Tp53 is not required. To assess the requirement for Bcl activity in MO induced ectopic rfng expression, wnt1 morphant embryos were co-injected with either Bcl2-GFP mRNA (G, I, K) or a morpholino to knockdown the pro-apoptotic Bcl gene puma (H, J, L). Both approaches rescued the expanded rfng expression (decreased ectopic rfng in K compared to I, L compared to J). All four experiments place ectopic rfng downstream of pro-apoptotic Bcl activity.

Pro-apoptotic Bcl genes puma and bax-a are required for normal boundary marker expression. Embryos injected with either of 2 different puma morpholinos (A–I) or a bax-a morpholino (JvL) show reduction or loss of rfng mRNA expression at the 21ss. Puma knockdown in wild-type (A–C) and tp53 mutant (D–I) embryos showed equivalent reductions in rfng expression (compare B, C and E, F, H, I). A krox20::GAL4 driver line was crossed to a UAS::puma line to force expression of Puma in rhombomeres 3 and 5. Double-positive embryos showed increased apoptosis (data not shown) and ectopic expression of rfng within these 2 segments (compare M and N, O). Similar ectopic rfng expression was detected when this GAL4 line was crossed to a UAS linked auto-activating Caspase3a (Rev-Casp3a, compare N, O and Q, R).

Additional morpholinos cause Tp53-dependent boundary domain expansion. Embryos injected with control (A–C), tcf3b (D), deltaA (E), rfng (F), tcf3b + tp53 (G), deltaA + tp53 (H), and rfng + tp53 (I) morpholinos, and hybridized with probe to boundary marker rfng. Tcf3b, deltaA, and rfng MOs alone induce an expansion of rfng expression (compare A–C with D–F), that is rescued when tp53 MO is co-injected (compare D–F with G–I).

Tp53 mutant fish do not display phenotypes of morpholino toxicity. Wild-type (A, B, C) and tp53 mutant (D, E, F) embryos were injected with control or wnt1 MOs. In situ hybridization for rfng mRNA revealed extensive ectopic rfng expression in the wnt1 MO-injected wild-type embryos, while no ectopic rfng was observed in the wnt1 MO-injected tp53 mutant fish (compare C and F).

Puma and bax-a morpholino efficacy. Schematic of puma genomic locus, splicing pattern and observed spliced and unspliced products (A). RT-PCR (B) using primers to detect spliced (primers F + R2, SPLICED) and unspliced (primers F + R1, UNSPLICED) puma transcripts was performed on control MO (Ct) and puma SB MO (Pu)-injected embryos (0.56 pmol each). Loss of spliced puma (SPLICED, compare Ct and Pu) and increase in unspliced puma (UNSPLICED, compare Ct and Pu) upon puma SB MO injection was observed. PCR using primers to EF1a confirmed near-equivalent amounts of cDNA were used. Puma splice-blocking and bax-a morpholinos protect against gamma irradiation induced apoptosis (C). Control MO-injected embryos (C, Ctrl MO) show extensive Caspase3 activation (anti-active Caspase3, green) after exposure to 12.5 Gray irradiation, while both pumaSB and bax-a MO-injected embryos show much reduced levels. Embryos were irradiated at 18ss, and subsequently grown for 3 h at 28.5 °C.

Acknowledgments
This image is the copyrighted work of the attributed author or publisher, and ZFIN has permission only to display this image to its users. Additional permissions should be obtained from the applicable author or publisher of the image.

Reprinted from Developmental Biology, 350(2), Gerety, S.S., and Wilkinson, D.G., Morpholino artifacts provide pitfalls and reveal a novel role for pro-apoptotic genes in hindbrain boundary development, 279-289, Copyright (2011) with permission from Elsevier. Full text @ Dev. Biol.