Page 321 - Haematologica March 2020

NGS of F7 and minigene studies identify molecular bases of FVII deficiency
events leading to two subsequent founder effects. In par- ticular, one haplotype was found in patients of Maghreb origin whereas the second one was compatible with genotypes of patients of European and Lebanese origins.
Concerning the c.572-392C>G change, the aberrant splicing profile displayed approximately 70% of correctly spliced transcripts, which is not apparently consistent with the FVII:C levels (10%) observed in homozygous patient #330. However, the mutation was associated with the less frequent A2 and M2 polymorphic alleles, which have been demonstrated to halve the FVII expression when present in the homozygous state.33-37
The c.571+78G>A and c.572-392C>G mutations were also found in patients with moderately reduced FVII lev- els (#28, #377, #284). In patients #377 and #284, this can be explained by the fact that: i) the mutations are present in heterozygous condition; and ii) by the additional con- tribution of the functional polymorphisms A2 and M2. For patient #28, heterozygous for the c.571+78G>A mutation, the residual expression could arise from the allele bearing the p.Arg59Trp change, which, however, has never been characterized.
In contrast to the c.571+78G>A and c.572-392C>G vari- ants, the in vitro splicing analysis did not reveal detrimental
A
B
effects on splicing for the c.806-329G>A change. Therefore, we were unable to explain the genotype-phe- notype relationship for three patients (#31, #262 and #341) who were not carriers of the c.571+78G>A or c.572- 392C>G mutations. Since we have not identified other candidate pathogenic variants besides the mutations pre- viously identified by conventional sequencing, it is tempt- ing to speculate that the genetic defect could be in the unexplored highly repetitive rich GC region of intron 2 or in the 5’ or 3’ regulatory regions of the F7 gene.
Knowledge of the alternative splicing patterns and of the mechanisms involved offers the opportunity to design correction strategies that could have therapeutic implica- tions.20-22 Here, we exploited variants of the U7 small nuclear RNA, the RNA component of the U7 small nuclear ribonucleoprotein that is biologically involved in histone RNA 3' end processing.46 By changing the Sm consensus sequence of the endogenous U7snRNA, it is possible to express U7snRNA variants (U7smOPT) that no longer modulate histone processing, but can bind to RNA targets though base-pair interaction and efficiently accumulate into nucleus as snRNP.41 Therefore, by chang- ing the 5’ tail of the U7smOPT, it is possible to target a desired RNA sequence and avoid its recognition by splic-
Figure 4. Investigation of aberrant splicing mechanisms by using antisense U7smOPT variants. (A) Schematic representation of engineered U7smOPT exploited in this study. The sequence of intronic mRNA and of the engineered 5’tail of U7smOPT with relative base-pairing is reported. The nucleotide changes identified by next- generation sequencing are indicated in bold and red, as well as the corresponding base in U7smOPT antisense sequence. (B) Splicing pattern analysis in HEK293T cells transiently transfected with pIVS6 wild type (wt) and variants (c.571+78A and c.572-392G) alone or in combination with engineered U7smOPT (pU7). PCR with T7bisF-F7ex7R oligonucleotides performed at 32 cycles (top) or, for semi-quantitative evaluation, at 25 cycles (bottom). M: 100 bp ladder.
haematologica | 2020; 105(3)
835