Commentary on "Poor evidence for host-dependent regular RNA editing in the transcriptome of SARS-CoV-2".

Analysis of the SARS-CoV-2 transcriptome has revealed a background of low-frequency intra-host genetic changes with a strong bias towards transitions. A similar pattern is also observed when inter-host variability is considered. We and others have shown that the cellular RNA editing machinery based on ADAR and APOBEC host-deaminases could be involved in the onset of SARS-CoV-2 genetic variability. Our hypothesis is based both on similarities with other known forms of viral genome editing and on the excess of transition changes, which is difficult to explain with errors during viral replication. Zong et al. criticize our analysis on both conceptual and technical grounds. While ultimate proof of an involvement of host deaminases in viral RNA editing will depend on experimental validation, here, we address the criticism to suggest that viral RNA editing is the most reasonable explanation for the observed intra- and inter-host variability.

Mechanisms for evolving hypervariability: the case of conopeptides.

Hypervariability is a prominent feature of large gene families that mediate interactions between organisms, such as venom-derived toxins or immunoglobulins. In order to study mechanisms for evolution of hypervariability, we examined an EST-generated assemblage of 170 distinct conopeptide sequences from the venoms of five species of marine Conus snails. These sequences were assigned to eight gene families, defined by conserved elements in the signal domain and untranslated regions. Order-of-magnitude differences were observed in the expression levels of individual conopeptides, with five to seven transcripts typically comprising over 50% of the sequenced clones in a given species. The conopeptide precursor alignments revealed four striking features peculiar to the mature peptide domain: (1) an accelerated rate of nucleotide substitution, (2) a bias for transversions over transitions in nucleotide substitutions, (3) a position-specific conservation of cysteine codons within the hypervariable region, and (4) a preponderance of nonsynonymous substitutions over synonymous substitutions. We propose that the first three observations argue for a mutator mechanism targeted to mature domains in conopeptide genes, combining a protective activity specific for cysteine codons and a mutagenic polymerase that exhibits transversion bias, such as DNA polymerase V. The high D:(n)/D:(s) ratio is consistent with positive or diversifying selection, and further analyses by intraspecific/interspecific gene tree contingency tests weakly support recent diversifying selection in the evolution of conopeptides. Since only the most highly expressed transcripts segregate in gene trees according to the feeding specificity of the species, diversifying selection might be acting primarily on these sequences. The combination of a targeted mutator mechanism to generate high variability with the subsequent action of diversifying selection on highly expressed variants might explain both the hypervariability of conopeptides and the large number of unique sequences per species.

Evolving better brains: a need for neurotrophins?

The NGF family of neurotrophins has a crucial role in regulating neuron numbers during vertebrate development. Six years ago the prediction was made that invertebrates with simple nervous systems, such as Caenorhabditis elegans, would lack neurotrophins. Surprisingly, it now appears that not only C. elegans but also Drosophila melanogaster, lack homologs of the neurotrophins or their trk receptors. Furthermore, functional studies indicate that control of neuronal numbers in Drosophila is primarily dependent on steroids. By contrast, a recognizable trk homolog exists in molluscs, a phylum that includes species with the most complex nervous systems in the invertebrate kingdom. This suggests that neurotrophic signaling mechanisms might be one of the prerequisites for evolution of complex nervous systems. Expansion of the genome projects to other invertebrates, such as molluscs and coelenterates, should provide new insights on the molecular correlates of building complex brains.

Structural features of the rat GFAP gene and identification of a novel alternative transcript.

The glial fibrillary acidic protein (GFAP) is expressed in a cell-specific manner and represents the major subunit of intermediate filaments of astroglial cells. The knowledge of the gene structure is an important step for further understanding the mechanisms of cell-specific expression. In the present study, we report the complete sequence of the rat GFAP gene and provide evidence for the existence, in the rat brain, of a novel alternative transcript. Since three different transcripts, indicated as GFAPalpha, beta, and gamma, have been previously reported (Feinstein et al. [1992] J. Neurosci. Res. 32:1-14; Zelenika et al. [1995] Mol. Brain Res. 30:251-258), we called this novel mRNA isoform GFAPdelta. It is generated by the alternative splicing of a novel exon located in the classic seventh intron. This alternative exon (called VII+) contains a 101-bp coding sequence in frame with exon VII and interrupted by a stop codon TAA at position +5451. Therefore, the novel GFAPdelta transcript encodes for an hypothetical GFAP where the forty-two carboxy-terminal amino acids encoded by exon VIII and IX are replaced by thirty-three amino acids encoded by exon VII+. Northern blot analysis with a specific probe for exon VII+ revealed a 4.2-kb mRNA, expressed in several brain areas, but absent in extracerebral tissues (lung, heart, kidney, liver, spleen). The previously discovered GFAP isoforms (alpha, beta, and gamma) produce hypothetical translation products differing in the amino-terminal Head domain. The present data suggest, for the first time, the possible existence of GFAP isoforms differing in the carboxy-terminal Tail domain.

GFAPbeta mRNA expression in the normal rat brain and after neuronal injury.

GFAPbeta mRNA is an alternative transcript of the glial fibrillary acidic protein (GFAP) gene, whose transcriptional start site is located 169 nucleotides upstream to the classical GFAPalpha mRNA. By an RT-PCR method with primers on separate exons, we were able to confirm the presence of GFAP transcripts with a longer 5' untranslated region in all the examined areas of rat brain and in primary cultures of astroglial cells. Northern blot analysis, using an oligoprobe specific for the 5' region of GFAPbeta, revealed a single hybridization band of 2.9 kb in all the brain regions examined and in primary cultures of astroglial cells. The availability of the quantitative Northern blot assay allowed further studies on the regulation of GFAPbeta expression in vivo. Since it is well-known that neuronal brain injury is one of the most powerful inducers of GFAP, we examined the expression of GFAPalpha and beta after a neurotoxic lesion in the rat hippocampus. Results obtained show a parallel increase in both GFAP transcripts with an identical time-course, suggesting that regulatory regions of the gene influence in similar way the rate of transcription at the two different start sites (alpha and beta) or that a similar post-transcriptional mechanism is involved in regulating both mRNA isoforms.

A neural-specific hypomethylated domain in the 5' flanking region of the glial fibrillary acidic protein gene.

In the present study we examined the methylation status of the glial fibrillary acidic protein (GFAP) gene promoter, analyzing various CG sites in both the human and rat gene in GFAP-expressing and nonexpressing tissues. Moreover, we studied the methylation of specific CG sites in different rat brain areas during postnatal development, in cell cultures highly enriched in specific neural- or non-neural-cell types (fibroblasts), and in human gliomas. The obtained results do not support a simple correlation between demethylation and expression of the GFAP gene but help to identify a cluster of CG sites in the 5'flanking region (from -1176 to -1471 in the rat) that are hypomethylated in neural cell types and localized in a region highly conserved between rat, mouse and human GFAP promoters. Neural-specific hypomethylation of this conserved zone can be observed also in the human GFAP gene both in normal brain tissue and neoplastic glial cells. A higher demethylation of the -1176 site at early stage of postnatal life was observed in specific rat brain areas, such as hippocampus and cerebellum. The most dramatic differences were observed in the cerebellum where a peak of demethylation of the -1176 site was detected at 15 days of postnatal life, followed by an intense remethylation of this site. Results of experiments in the CG4 glial progenitor cell line showed that demethylation of the -1176 site is already established before transcriptional activation of the GFAP gene. Moreover, results of experiments in primary cell cultures show that in neuronal cell types, such as cerebellar granule cells and embryonic cerebral hemisphere neurons, the level of demethylation of the -1176 site is comparable to that observed in cultured astrocytes. In contrast a high level of methylation can be observed in cultured non-neural cell types (fibroblasts). Such neural-specific hypomethylation could be established in a very early stage in the progression along the neural cell lineage and could play a role in maintaining a local open chromatin conformation which is then necessary to allow the interaction with specific regulatory factors present in astroglial cells.