*3.7. RG4-Forming Sequences*

Probable RG4-forming sequences were only predicted in the 5′UTRs of four ABCA genes (33%; *ABCA1*, *ABCA2*, *ABCA3*, and *ABCA7*); three of the sequences were located within the first third of the region. G-quadruplexes (G4) are secondary structures involving four nucleic acid strands that can be adopted by both DNA (DG4) and RNA (RG4) that contain guanine-rich sequences [40,41]. DG4 sequences have been mainly connected with the regulation of transcription and RG4 sequences regulation of translation; however, many other processes such as telomere maintenance, splicing, or RNA localization have also been discussed. Generally, RG4 sequences are believed to play roles as negative regulators of translation, although a positive effect of these structures has also been reported [17,42]. Thus, the regulatory effect of an RG4 sequence is significantly influenced by its position, the surrounding DNA topology, and other factors [43,44]. Huppert et al. [45] mapped the incidence of G4 in human 5′UTRs and calculated that 6.2% of these regions contained putative G4 sequences, with a density of approximately 0.3 per kb. Their data showed the highest density of putative G4 sequences at the 5′ -ends of the 5′UTRs, decreasing approximately linearly along the 5 ′UTR regions. According to the analyses performed by Maizels and Gray [46], in 10–15% of human genes, a G4 motif occurs in the region specifying the 5′UTR of the encoded mRNA. However, the data analyzed in the latter study showed a higher G4 motif frequency at the 3′ -ends of the 5′UTRs than at 5′ -ends. With a new algorithm named G4Hunter, Bedrat and coworkers [47] found a significantly higher occurrence of G4 sequences in the human genome, with a density of approximately 2.4 per kb. 53.3% of 5′UTRs contained at least 1 G4-forming sequence in their study. In light of the latest results published in the literature, we can conclude that RG4 sequences are present in below-average numbers within the 5′UTRs of ABCA genes.
