Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.
Learn more at Review Commons
Referee #3
Evidence, reproducibility and clarity
The manuscript by Wong et al., provides the first report of a homozygous loss-of-function mutation of the RAF1 gene in humans. The mutation (T543M) was found in two siblings of a consanguineous family in association with perinatal death and multiple developmental abnormalities. These abnormalities show strong similarities with a rare congenital malformation syndrome with unknown aetiology named Acro-cardio-facial syndrome (ACFS, MIM600460). Conversely, reported abnormalities are different from those observed in RASopathies, congenital diseases caused by gain-of-function mutations in either the RAF gene or genes implicated in the same signaling pathway (MAPK) and that induce its ectopic/over-activation. By performing functional experiments in cellular systems (cell line where the mutated RAF was either overexpressed or knocked-in) and in Xenopus Laevis embryos, the authors demonstrate that the reason for the phenotypic differences compared to RASopathies is that the RAF1T543M variant impairs the signaling activity of RAF and blunts MAPK pathway activation. In particular, the RAF1T543M variant: 1) is not actively phosphorylated at key activating residues, 2) is unable to transduce MAPK signaling towards MEK/ERK substrates, 3) is inherently unstable and prone to proteasome-mediated degradation and 4) is unable to block stress-induced apoptosis. On the basis of increased apoptosis detected in cellular systems and some morphological defects observed in the probands, the author classify this novel syndrome as a segmental progeroid syndrome.
Major comments:
The authors perform a thorough analysis of the structural and functional defects of the RAF1T543M protein, using in silico analyses and in vitro systems. The data are corroborated by an elegant and clear-cut experiment in Xenopus embryos that demonstrates that the mutated RAF is not able to transduce MAPK signaling when overexpressed in an in vivo model. The lack of patients' material prevents a validation in human cells, but I think the evidence collected in the manuscript is supportive of the loss-of-function mutation in RAF as the causative mutation of the observed phenotype.
The concept illustrated in the last sentence of the discussion, i.e. that different mutations in the same genes can either cause cancer (when overactivating) or premature aging (when blunting the activity of the enzyme) is fascinating. I think discussing this concept in the context of this manuscript is appropriate. However, the authors classify the syndrome caused by the RAF1T543M variant as a "novel segmental progeroid syndrome" in the title, abstract and first sentence of the discussion. I don't think that presented data are convincing for this classification. Indeed, two affected siblings die at very early post-natal stages (7 and 50 days, respectively) likely because of malformations that are not compatible with life and that are due to altered in uthero development. Progeroid syndromes are a heterogeneous group of syndromes, a minority of which is characterized by malformations at birth. The more general concept is that physical abnormalities are progressively acquired during post-natal life, in specific tissues that show typical features associated with aging, including senescence, decline of the stem cell compartment, increased inflammation. In the case of individuals carrying the RAF1T543M variant, none of these tissue abnormalities have been reported (due to the lack of material from patients) and the classification as a progeroid syndrome is based on external inspection of organs. Reported phenotypes are not specific and a failure of in uthero development of heart, limbs and other organs seems like the absolutely predominant trait. This in uthero phenotype is perfectly consistent with the physiological role of the MAPK pathway downstream multiple RTKs that transduce morphogenetic, in addition to mitogenic, signals. As for the authors words: "endogenous FGF/FGFR signaling is required for proper mesoderm and neural induction" and supports the author claim that the analysis of individuals carrying the RAF1T543M variant "underscores the importance of RTK signaling during human development". I think that the "progeroid" phenotypes are less clear. It is still important to present the phenotypes reminiscent of progeroid syndromes and discuss them and perhaps more clearly explain the logical connection with the increased apoptosis observed in in vitro experiments.
In regard to the apoptosis experiment in figure 4, it supports the claim that RAF1 activity is necessary for protection from stress-induced apoptosis. However, the cartoon presented in figure 4C also shows that this process is mediated by increased ASK1/MST2 signaling. This part of the cartoon is based on the literature and has not been formally demonstrated. The authors can try to rescue the apoptotic status by silencing ASK1 in the RAF1T543M cells, similar to what has been done in Yamaguchi, O., 2004, Cardiac-specific disruption of the c-raf-1 gene induces cardiac dysfunction and apoptosis. J. Clin. Invest. 114, 937-943. Literature also suggests that RAF1-mediated inhibition of apoptosis is kinase-independent. This part is particularly interesting since the variant produces a protein that is both kinase-inactive and unstable. It would be a nice addition to the matuscript if the authors could clarify, at least in their cellular model, whether the increased apoptosis is due to the loss of either the protein or the kinase activity.
Minor comments:
A minor comment to the Xenopus ISH: the number of embryos that have been analyzed per condition is not reported anywhere. Figure legend explains that presented images are "representative pictures", but knowing the number of tested embryos and the penetrance of the phenotype would help undertsand the relevance of the RAF mutation in the signaling pathway under investigation.
Significance
This work by Wong and colleagues provides conceptual advance and will be of interest for researchers dealing with RTK/MAPK signaling in multiple contexts including oncology, developmental biology and cardiomyopathy.
The gene under investigation is widely studied in cancer, where gain-of-function, oncogenic mutations are common. The role of RAF1 during embryonic development is less known. A couple of studies have investigated the developmental phenotype of mouse models carrying either a knock-out allele (Mikula, M, et al. Embryonic lethality and fetal liver apoptosis in mice lacking the c-raf-1 gene. EMBO J. 2001. 20:1952-1962) or an allele producing a truncated protein (Wojnowski L et al., (1998) Craf-1 protein kinase is essential for mouse development. Mech Dev, 76, 141-149). A few studies have investigated the conditional inactivation in specific tissues, such as the cardiac muscle (Yamaguchi, O., 2004, Cardiac-specific disruption of the c-raf-1 gene induces cardiac dysfunction and apoptosis. J. Clin. Invest. 114, 937-943.). And one study has found heterozygous carriers of a loss-of-function mutation among cohorts of children affected by dilated cardiomyopathy (Dhandapany, P.S., et al. (2014). RAF1 mutations in childhood-onset dilated cardiomyopathy. Nat. Genet. 46, 635-639). The manuscript reports the spectrum of defects acquired during embryonic development by carriers of a pathogenic mutation in the RAF1 gene. The mutation impairs both stability and kinase activity of the protein. The manuscript points out the non-redundant role of RAF1-mediated signaling in specific organs during embryonic development.
The person who is reviewing the manuscript has expertise in cellular signaling, mouse embryonic development and human aging.
