Limits...
BAK1 gene variation and abdominal aortic aneurysms--results may have been prematurely overrated.

Küry S, Airaud F, Piloquet P, Bézieau S - Hum. Mutat. (2010)

View Article: PubMed Central - PubMed

AUTOMATICALLY GENERATED EXCERPT
Please rate it.

The most confusing point fits in the commentary on the article by Hatchwell... We did not understand why the authors had not compared the same type of nucleic acid between blood (DNA) and matching aortic tissues (RNA)... Indeed, Gottlieb et al.'s Table 2 suggests that allelic variants at position 42 (rs1051911), 52 (rs1051912), and 81 (rs1051913) are likely in very strong or even complete linkage disequilibrium, because only two haplotype combinations are represented (major haplotype CA/GC/AT vs. variant haplotype CA/GC/AT)... By the way, it is not clear whether the genotypes reported in Gottlieb et al.'s Table 2 were found in a homozygous state; it seems odd that the sequences of the aortic tissues do not exhibit traces of the major allelic variants, which are largely predominant in blood (of note, rs1051912 and rs1051913 variant alleles were never found in any of the HapMap or SNP500 cancer populations)... Gottlieb et al. excluded that BAK1 cDNA sequences of abdominal aortic tissues may have been contaminated by BAK2 sequences, because of the RNAse treatment used for reverse transcription... This would confirm the hypothesis of a sequencing artifact [Hatchwell, ], given that the RNA sequences of BAK1 and BAK2 differ exactly at the points of variation (single nucleotide polymorphisms [SNPs]) observed by Gottlieb et al. (Fig. 1)... Besides, in order to refute the hypothesis of a contamination by a sequence other than BAK1, the authors use two variants concerning codons 2 (ATG/GC/TCG for BAK2 vs... In conclusion, like Dr. Hatchwell, we are not convinced that the cDNA sequence presented by Gottlieb et al. in the aortic tissues corresponds to BAK1 and not to pseudogene BAK2... Yet, even if the contamination was confirmed there would still remain one enigma: why would the authors find a homozygous variant sequence, because the oligonucleotidic primers that they used for their experiments perfectly match both BAK1 and BAK2 mRNA sequences? One explanation could be an allele-specific amplification due to an undescribed polymorphism in one of the primer sequences... All in all, Gottlieb et al. contains an accumulation of inaccuracies and paradoxical results that put a doubt on their conclusions... We do not pretend that their interpretations are wrong, but we believe that there is too huge a contrast between the strong message delivered by the authors and the quality of the experiments and report that they proposed, which did not meet the high-level standard expected for a major work—considering also the obvious translation errors of original Table 2 of Gottlieb et al.'s article corrected in the Erratum published online on 22 November 2009... It is also noteworthy that the quite neutral title of the article, “BAK1 gene variation and abdominal aortic aneurysms,” does not reflect the revolutionary idea relayed by the media... This suggests that the authors probably did not expect such an impact for their work, and that their results might have been prematurely overrated.

Show MeSH
Comparison of the different sequences retrieved from NCBI for BAK2 gene. Focus on the two variants in codons 2 and 145 used by Gottlieb et al. [2010] to refute a crosscontamination between BAK1 and BAK2 sequences. We show here that two of the five BAK2 reference sequences presented do not differ from BAK1 at codon 2.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC2992694&req=5

fig02: Comparison of the different sequences retrieved from NCBI for BAK2 gene. Focus on the two variants in codons 2 and 145 used by Gottlieb et al. [2010] to refute a crosscontamination between BAK1 and BAK2 sequences. We show here that two of the five BAK2 reference sequences presented do not differ from BAK1 at codon 2.

Mentions: A more plausible hypothesis remains the crosscontamination between BAK1 transcript and genomic DNA of pseudogene BAK2 (BAK1P1) suggested by Hatchwell [2010]. Gottlieb et al. [2010] excluded that BAK1 cDNA sequences of abdominal aortic tissues may have been contaminated by BAK2 sequences, because of the RNAse treatment used for reverse transcription. They consider thereby that BAK2 exists in its genomic form only [Gottlieb et al., 2010]. However, according to the databank reference sequence XM_002348050, BAK2 would have been predicted according to similarities to 1 mRNA, 18 ESTs, and 5 proteins. The fact that one supporting evidence is an mRNA suggests that an mRNA transcript and even a translation is not unlikely for BAK2. This would confirm the hypothesis of a sequencing artifact [Hatchwell, 2010], given that the RNA sequences of BAK1 and BAK2 differ exactly at the points of variation (single nucleotide polymorphisms [SNPs]) observed by Gottlieb et al. [2009] (Fig. 1). Besides, in order to refute the hypothesis of a contamination by a sequence other than BAK1, the authors use two variants concerning codons 2 (ATG/GCC/TCG for BAK2 vs. ATG/GCT/TCG for BAK1) and 145 of both BAK1 and BAK2 that would enable to discriminate the two genes [Gottlieb et al., 2010]. The problem is that the variant at codon 2 is not present in all the BAK2 sequences of reference found in the NCBI databank (Fig. 2), which makes very doubtful the validity of the reference sequence for BAK2 and thus the demonstration of Gottlieb et al. [2009,2010]. Moreover, the authors do not comment on the six other variants that differ between the sequences of BAK1 (XM_002348050) and BAK2 (NG_000850), which would be useful to make a discrimination between the two gene sequences.


BAK1 gene variation and abdominal aortic aneurysms--results may have been prematurely overrated.

Küry S, Airaud F, Piloquet P, Bézieau S - Hum. Mutat. (2010)

Comparison of the different sequences retrieved from NCBI for BAK2 gene. Focus on the two variants in codons 2 and 145 used by Gottlieb et al. [2010] to refute a crosscontamination between BAK1 and BAK2 sequences. We show here that two of the five BAK2 reference sequences presented do not differ from BAK1 at codon 2.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC2992694&req=5

fig02: Comparison of the different sequences retrieved from NCBI for BAK2 gene. Focus on the two variants in codons 2 and 145 used by Gottlieb et al. [2010] to refute a crosscontamination between BAK1 and BAK2 sequences. We show here that two of the five BAK2 reference sequences presented do not differ from BAK1 at codon 2.
Mentions: A more plausible hypothesis remains the crosscontamination between BAK1 transcript and genomic DNA of pseudogene BAK2 (BAK1P1) suggested by Hatchwell [2010]. Gottlieb et al. [2010] excluded that BAK1 cDNA sequences of abdominal aortic tissues may have been contaminated by BAK2 sequences, because of the RNAse treatment used for reverse transcription. They consider thereby that BAK2 exists in its genomic form only [Gottlieb et al., 2010]. However, according to the databank reference sequence XM_002348050, BAK2 would have been predicted according to similarities to 1 mRNA, 18 ESTs, and 5 proteins. The fact that one supporting evidence is an mRNA suggests that an mRNA transcript and even a translation is not unlikely for BAK2. This would confirm the hypothesis of a sequencing artifact [Hatchwell, 2010], given that the RNA sequences of BAK1 and BAK2 differ exactly at the points of variation (single nucleotide polymorphisms [SNPs]) observed by Gottlieb et al. [2009] (Fig. 1). Besides, in order to refute the hypothesis of a contamination by a sequence other than BAK1, the authors use two variants concerning codons 2 (ATG/GCC/TCG for BAK2 vs. ATG/GCT/TCG for BAK1) and 145 of both BAK1 and BAK2 that would enable to discriminate the two genes [Gottlieb et al., 2010]. The problem is that the variant at codon 2 is not present in all the BAK2 sequences of reference found in the NCBI databank (Fig. 2), which makes very doubtful the validity of the reference sequence for BAK2 and thus the demonstration of Gottlieb et al. [2009,2010]. Moreover, the authors do not comment on the six other variants that differ between the sequences of BAK1 (XM_002348050) and BAK2 (NG_000850), which would be useful to make a discrimination between the two gene sequences.

View Article: PubMed Central - PubMed

AUTOMATICALLY GENERATED EXCERPT
Please rate it.

The most confusing point fits in the commentary on the article by Hatchwell... We did not understand why the authors had not compared the same type of nucleic acid between blood (DNA) and matching aortic tissues (RNA)... Indeed, Gottlieb et al.'s Table 2 suggests that allelic variants at position 42 (rs1051911), 52 (rs1051912), and 81 (rs1051913) are likely in very strong or even complete linkage disequilibrium, because only two haplotype combinations are represented (major haplotype CA/GC/AT vs. variant haplotype CA/GC/AT)... By the way, it is not clear whether the genotypes reported in Gottlieb et al.'s Table 2 were found in a homozygous state; it seems odd that the sequences of the aortic tissues do not exhibit traces of the major allelic variants, which are largely predominant in blood (of note, rs1051912 and rs1051913 variant alleles were never found in any of the HapMap or SNP500 cancer populations)... Gottlieb et al. excluded that BAK1 cDNA sequences of abdominal aortic tissues may have been contaminated by BAK2 sequences, because of the RNAse treatment used for reverse transcription... This would confirm the hypothesis of a sequencing artifact [Hatchwell, ], given that the RNA sequences of BAK1 and BAK2 differ exactly at the points of variation (single nucleotide polymorphisms [SNPs]) observed by Gottlieb et al. (Fig. 1)... Besides, in order to refute the hypothesis of a contamination by a sequence other than BAK1, the authors use two variants concerning codons 2 (ATG/GC/TCG for BAK2 vs... In conclusion, like Dr. Hatchwell, we are not convinced that the cDNA sequence presented by Gottlieb et al. in the aortic tissues corresponds to BAK1 and not to pseudogene BAK2... Yet, even if the contamination was confirmed there would still remain one enigma: why would the authors find a homozygous variant sequence, because the oligonucleotidic primers that they used for their experiments perfectly match both BAK1 and BAK2 mRNA sequences? One explanation could be an allele-specific amplification due to an undescribed polymorphism in one of the primer sequences... All in all, Gottlieb et al. contains an accumulation of inaccuracies and paradoxical results that put a doubt on their conclusions... We do not pretend that their interpretations are wrong, but we believe that there is too huge a contrast between the strong message delivered by the authors and the quality of the experiments and report that they proposed, which did not meet the high-level standard expected for a major work—considering also the obvious translation errors of original Table 2 of Gottlieb et al.'s article corrected in the Erratum published online on 22 November 2009... It is also noteworthy that the quite neutral title of the article, “BAK1 gene variation and abdominal aortic aneurysms,” does not reflect the revolutionary idea relayed by the media... This suggests that the authors probably did not expect such an impact for their work, and that their results might have been prematurely overrated.

Show MeSH