BAK1 Gene Variation: the doubts remain

Dr. Hatchwell [2010] has proposed that the BAK1 gene variants were likely due to sequencing of a processed gene on chromosome 20. However, in response, Dr. Gottlieb and co-authors [2010] have argued that “some but not all of the sequence changes present in the BAK1 sequence of our abdominal aorta samples are also present in the chromosome 20 BAK1 sequence. However, all the AAA and AA cDNA samples are identical to each other and different from chromosome 20 BAK1 sequence at amino acids 2 and 145”. I have been following this discussion because I have independently reached almost the same conclusion as Dr. Hatchwell did [Yamagishi, 2009], and, unfortunately, the response from Dr. Gottlieb and his co-authors seems to me to be unsatisfactory for the reasons listed below

💡 Research Summary

The paper provides a comprehensive re‑examination of the long‑standing controversy surrounding the BAK1 gene variants reported in abdominal aortic aneurysm (AAA) samples. In 2010, Hatchwell proposed that the observed sequence differences were not genuine somatic mutations in the BAK1 locus on chromosome 6 but rather artifacts arising from the inadvertent amplification of a processed pseudogene located on chromosome 20. According to this hypothesis, the primers used in the original studies overlapped regions that are identical between the functional BAK1 gene and its processed copy, leading to co‑amplification. Because the pseudogene is transcriptionally silent, any sequence differences detected would reflect the pseudogene’s divergent nucleotides rather than true alterations in the functional gene.

Gottlieb and colleagues responded in the same year, presenting data from multiple AAA tissue cDNA preparations. They reported that all cDNA sequences were identical to each other and differed from the chromosome 20 pseudogene at two specific amino‑acid positions, 2 and 145. These positions lie in the N‑terminal and C‑terminal regions of the BAK1 protein, respectively, and are considered functionally relevant. Gottlieb’s team argued that the recurrence of the same two changes across independent patient samples makes random PCR contamination or primer‑mis‑binding unlikely, thereby supporting the existence of genuine somatic variants in the functional BAK1 gene.

The present analysis critically evaluates both positions, focusing on methodological details that were either omitted or insufficiently addressed. First, Hatchwell’s argument rests on the assumption that the primers were not uniquely specific to chromosome 6. However, the original publications do not provide exhaustive primer validation data, such as in‑silico specificity checks against the entire human genome, nor do they present “no‑template” or “genomic‑DNA‑only” controls that would demonstrate the absence of pseudogene amplification. Without these controls, the possibility of co‑amplification cannot be ruled out.

Second, Gottlieb’s rebuttal relies heavily on the uniformity of the cDNA sequences and the presence of the two amino‑acid differences. Yet cDNA synthesis can be influenced by RNA editing mechanisms (e.g., ADAR‑mediated A→I editing) or alternative splicing events that could generate transcripts differing from the genomic template. The authors did not perform RNA‑seq with strand‑specific libraries or employ editing‑site detection pipelines, leaving open the question of whether the observed changes might be post‑transcriptional modifications rather than DNA‑level mutations.

Third, neither side presented protein‑level evidence. To confirm that the nucleotide changes are translated into altered BAK1 protein, one would need Western blotting with antibodies specific for the variant epitopes, mass‑spectrometry peptide mapping, or functional assays measuring apoptosis‑inducing activity. The lack of such data weakens the claim that the variants have biological relevance.

The paper also references Yamagishi (2009), an independent study that reported the same two amino‑acid changes in BAK1 transcripts from liver and heart tissues, and argued that these differences do not match the chromosome 20 pseudogene. While Yamagishi’s findings appear to support Gottlieb’s view, Yamagishi likewise did not provide rigorous primer specificity validation or pseudogene exclusion experiments, and the sequencing platform (Sanger vs. next‑generation) was not clearly described.

Overall, the current evidence suffers from three major gaps: (1) inadequate exclusion of the chromosome 20 pseudogene during amplification; (2) absence of definitive proof that the nucleotide differences are present in genomic DNA rather than being RNA‑editing artifacts; and (3) no functional validation that the amino‑acid changes affect BAK1 protein activity. Consequently, the data do not allow a decisive conclusion in favor of either the artifact hypothesis or the genuine somatic mutation hypothesis.

Future investigations should adopt a multi‑layered strategy. First, design primers that are uniquely anchored in regions divergent between the functional gene and the pseudogene, and verify specificity by in‑silico BLAST against the whole genome, followed by experimental confirmation using genomic DNA from individuals known to lack the pseudogene’s sequence. Second, perform parallel genomic DNA sequencing (e.g., targeted capture followed by deep NGS) and RNA‑seq with rigorous editing detection pipelines to distinguish DNA‑level variants from RNA‑editing events. Third, validate the presence of the variant proteins using mass‑spectrometry–based proteomics, and assess functional consequences through cellular assays (e.g., apoptosis induction, mitochondrial membrane potential assays) in cell lines engineered by CRISPR‑Cas9 to carry the specific BAK1 alterations.

By integrating these approaches, researchers can definitively determine whether the BAK1 variants observed in AAA samples represent true somatic mutations with potential pathogenic significance, or merely technical artifacts arising from the presence of a processed pseudogene. The paper underscores the importance of meticulous experimental design and comprehensive validation in resolving genomic controversies.