p53 activation by knockdown technologies.
Robu ME, Larson JD, Nasevicius A, Beiraghi S, Brenner C, Farber SA, Ekker SC. · 2007
Read p53 activation by knockdown technologies. by Robu ME, Larson JD, Nasevicius A, Beiraghi S, Brenner C, Farber SA, Ekker SC. (2007) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/europmc-pmc1877875.txt.
Introduction
Robu, Ekker and co-authors warned in 2007 that common gene-silencing tools in zebrafish trigger a hidden side effect. Morpholinos and small interfering RNAs are injected into embryos to switch off one gene at a time. Both can also hit unintended targets, and side effects are hard to separate from the true result.
Comparing injected embryos with known mutants, the team found the same off-target pattern for both tool types and several targeting styles. The main effect was unwanted cell death driven by activation of p53, the cell's damage guardian. Blocking p53 at the same time erased that off-target death. It did not mask genuine effects, such as the death caused by losing the gene chordin. The authors urge routine p53 co-blocking as a control in knockdown studies.
Catalog record
| CREATOR | Robu ME, Larson JD, Nasevicius A, Beiraghi S, Brenner C, Farber SA, Ekker SC. |
|---|---|
| YEAR | 2007 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 57 KB |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
p53 Activation by Knockdown Technologies By Mara E Robu, Jon D Larson, Aidas Nasevicius, Soraya Beiraghi, Charles Brenner, Steven A Farber, Stephen C Ekker, Mary Mullins Published in: PLoS Genetics DOI: 10.1371/journal.pgen.0030078 ABSTRACT Morpholino phosphorodiamidate antisense oligonucleotides (MOs) and short interfering RNAs (siRNAs) are commonly used platforms to study gene function by sequence-specific knockdown. Both technologies, however, can elicit undesirable off-target effects. We have used several model genes to study these effects in detail in the zebrafish, Danio rerio. Using the zebrafish embryo as a template, correct and mistargeting effects are readily discernible through direct comparison of MO-injected animals with well-studied mutants. We show here indistinguishable off-targeting effects for both maternal and zygotic mRNAs and for both translational and splice-site targeting MOs. The major off-targeting effect is mediated through p53 activation, as detected through the transferase-mediated dUTP nick end labeling assay, acridine orange, and p21 transcriptional activation assays. Concurrent knockdown of p53 specifically ameliorates the cell death induced by MO off-targeting. Importantly, reversal of p53-dependent cell death by p53 knockdown does not affect specific loss of gene function, such as the cell death caused by loss of function of chordin. Interestingly, quantitative reverse-transcriptase PCR, microarrays and whole-mount in situ hybridization assays show that MO off-targeting effects are accompanied by diagnostic transcription of an N-terminal truncated p53 isoform that uses a recently recognized internal p53 promoter. We show here that MO off-targeting results in induction of a p53-dependent cell death pathway. p53 activation has also recently been shown to be an unspecified off-target effect of siRNAs. Both commonly used knockdown technologies can thus induce secondary but sequence-specific p53 activation. p53 inhibition could potentially be applicable to other systems to suppress off-target effects caused by other knockdown technologies. INTRODUCTION Morpholino phosphorodiamidate oligonucleotides (MOs) [1] and short inhibitory RNAs (siRNAs) [2] have been instrumental to induce sequence-specific gene knockdown in multiple systems. However, the use of both technologies is sometimes limited by induction of off-target effects [3–7]. About 15–20% of MOs used in zebrafish show off-targeting effects [3], represented by a signature neural death peaking at the end of segmentation (1 day post-fertilization [dpf]). The affected embryos grow with smaller heads and eyes, exhibit somite and notochord abnormalities, and eventually display craniofacial defects. These MO-induced developmental defects are target-independent because they are not displayed by characterized mutants in the respective genes [3]. We show here that the off-target effects of MOs are mediated through p53-induced apoptosis. Concurrent knockdown of p53 with various MOs significantly alleviates off-target neural death. Importantly, however, p53 MO did not affect specific phenotypes induced by a variety of MOs. We propose the use of p53 knockdown as a tool to attenuate off-target effects and facilitate the study of specific loss of function phenotypes. RESULTS MOs Exhibit Off-Target Neural Apoptotic Effects General morphological features of MO-induced off-target neural death have been previously described [3]. We further investigated the nature of this cell death and the mechanism of MO mistargeting. For this report, we focused primarily on MOs designed against two gene targets for which mutants have been previously described, wnt5/pipetail (ppt) [8] and smoothened/slow muscle omitted (smu) [9,10], to facilitate the discrimination between specific and nonspecific effects. A translational MO against smoothened (Smo MO) induces characteristic smu phenotype (spinal curvature, U-shaped somites) (Figure 1C). A splice-site wnt5 MO (Wnt5 MO1) induces tail and body-axis shortening and somite compression (Figure 1E), characteristic of the wnt5/ppt mutant (Figure 1K). What both Smo MO- and Wnt5 MO-injected embryos (morphants) have in common is an additional and very similar neural death (Figure 1, arrows). This neural death is target-independent, since it is not exhibited by the respective mutants (Figure 1K) [9,10]. Nonetheless, this neural death appears to be sequence-specific, since a completely different splice-site wnt5 MO (Wnt5 MO2) shows no neural death, but readily induces the characteristic wnt5/ppt phenotype (Figure 1G). We tested another type of knockdown molecule based on an alternating trans-4-hydroxy-L-proline/phosphate polyamide backbone called gripNA [11]. Interestingly, a gripNA targeting wnt5 (similar in sequence to Wnt5 MO1) also induces neural death along with the characteristic wnt5/ppt phenotype (Figure 1I). A gripNA against smoothened also causes additional neural death (unpublished data), supporting the idea that the off-targeting effects are not limited to the MO chemistry, but represent a common feature to these knockdown technologies. [Figure: p53 MO Attenuates Cell Death Induced by MOs and GripNAs(A–G and I–N) Brightfield images of 28 hpf embryos injected with demonstrative MOs and gripNAs. The arrows (C, E, I, and M) indicate neural death that is significantly attenuated to normal head size and morphology by co-knockdown of p53 (D, F, J, and N). Interestingly, Wnt5 MO2 shows no significant neural death (G) even at 6 ng, a higher dose than Wnt5 MO1 (E) (3 ng), but can elicit a highly penetrant Wnt5 phenotype even at 1.5 ng (unpublished data).(H) Effect of p53 MO on Wnt5 splicing. We carried out RT-PCR using primers spanning the exon 5–exon 6 junction targeted by Wnt5 MO1. Proper splicing was completely inhibited by Wnt5 MO1. p53 co-knockdown did not affect the efficiency of Wnt5 MO1 inhibition. Embryos injected with Wnt5 MO2, which targets the previous junction (exon 4–exon 5), still exhibited some properly spliced transcript at the exon 5–exon 6 junction. β-actin was used as a loading control.] The off-target neural death induced by MOs is highly reminiscent of the neural death induced by a published Mdm2 MO (Figure 1M). Mdm2 is a negative regulator of the tumor suppressor p53, the gene most frequently mutated in human cancers [12]. Mdm2 knockout in mice is an embryonic lethal [13] due to extensive p53 upregulation and p53-induced apoptosis. Mdm2-targeted MO in zebrafish was reported to induce apoptotic neural death [14]. We examined the mechanism of MO-induced off-target neural death by testing for apoptosis in multiple MO-injected zebrafish embryos using a transferase-mediated dUTP nick end labeling (TUNEL) assay (Figure 2) and by staining with acridine orange (unpublished data). Our results suggest that the neural death induced by off-targeting MOs is apoptotic in nature and was indistinguishable from the cell death observed in the Mmd2 knockdown [14]. We tested the specificity of Mdm2 MO-induced cell death by overexpressing a Mdm2 RNA construct. However, we did not observe any significant rescue of the Mdm2 MO-induced cell death with the Mdm2 RNA construct (unpublished data). Therefore, it is possible that the cell death phenotype induced by the Mdm2 MO is also primarily an off-targeting effect. We performed an in-depth analysis of MO off-targeting to examine the concordance between the phenotypes observed by light microscopy and apoptosis patterns observed by TUNEL staining. We analyzed zebrafish embryos injected with Wnt5 MO1 at 14 hpf (the onset of cell death, [3]), 22 hpf, 26 hpf, and 30 hpf (Figures 3 and 4; Figure S1). Brightfield images show the signature appearance of opaque-looking discolored tissue around the eyes and in the nervous system in embryos injected with Wnt5 MO1 (Figures 3 and 4; Figure S1). The extent of the opaque tissue increased at later time points and with increasing MO dose. The cell death could be more easily visualized using darkfield microscopy (Figures 3A, 3C, 3F, 3I, 3L, 3O, 3R, 3U, 3X, 4C, 4F, 4I, 4L, 4O, 4R, 4U, and 4X). This analysis