RESPONSE: Re: Prognostic Significance of a Short Sequence Insertion in the MCL-1 Promoter in Chronic Lymphocytic Leukemia

The Correspondences by Freeman et al., Vargas et al., Iglesias-Serret et al., Nenning et al., and Dicker et al. address four key points regarding MCL-1 promoter nucleotide insertions. Because our control group was too small (n = 18) to determine population prevalence, we simply referred to these promoter changes as nucleotide insertions. Based on an analysis of 55 additional control samples, we have now found that the allele frequencies in the Saskatchewan population are 0.98 for the wild-type allele and 0.01 each for 6- and 18-nucleotide insertion alleles ( Fig. 1A and B ). The presence of these nucleotide insertions in other healthy populations is also reported in four of the five letters. Vargas et al. refer to their presence in the healthy controls; the frequencies of the +18 allele were 11.8% in Freeman et al., 17.5% in Iglesias-Serret et al., and 27.5% in Dicker et al., and the frequencies of the +6 allele were 20.3%, 12.5%, and 23.7%, respectively. The frequency of these nucleotide insertions in 134 patients with acute lymphoblastic leukemia reported by Nenning et al. is similar to what we observed in patients with chronic lymphocytic leukemia (CLL). Interestingly, Dicker et al. also report a novel 12-nucleotide insertion with a frequency of 1.4% in CLL case patients. Although there can be exceptions, the commonly accepted definition of a mutation is that the least common allele has a frequency of less than 1%. The presence of these nucleotide insertions in more than 1% of healthy population suggests that the MCL-1 promoter nucleotide insertions are polymorphisms, not mutations. Allelic frequency of MCL-1 promoter in control and CLL populations. ( A ) MCL-1 promoter changes in genomic DNA amplified from peripheral blood, amplified by polymerase chain reaction (PCR), resolved on a polyacrylamide gel, and analyzed using the Gel-Doc 2000 system after staining with ethidium bromide. M = 100-bp Gene Ruler marker; N = negative control (PCR amplification without DNA); lanes = amplified DNA samples. We did not observe any of the heteroduplex bands reported by Vargas et al., possibly because we used different primers and PCR conditions. ( B ) Allelic frequency of MCL-1 without and with 6- or 18-nucleotide insertions as determined by PCR amplification and gel electrophoresis/sequencing of genomic DNA extracted from the peripheral blood of healthy individuals (n = 73; 18 are from the original study, and 55 are new control subjects) and CLL patients (n = 58). ( C ) Schematic of the predicted MCL-1 promoter recognition using genetic algorithm discriminant analysis. The program recognizes eukaryotic gene promoters based on an assessment of dinucleotide frequency distribution by discriminant analysis; x -axis numbers correspond to GenBank promoter sequence (AF147742). The y axis shows the promoter activity score. The vertical line represents the predicted peak score for the transcription start site for the wild-type promoter sequence, with the highest peak score, 0.87, at this position. With the +6 insertion, the predicted activity score increased from 0.87 (score for the wild-type allele) to 0.99 (up arrow); with the +12 insertion, the predicted score increased from 0.87 to 0.96 (not shown); and with the +18 insertion, the predicted score decreased from 0.87 to 0.57 (down arrow). The exact transcriptional start site of the MCL-1 promoter is controversial ( 1 – 3 ) . The nucleotide insertions can be inferred to be either at position −190, as reported by us and ( 1 ) , or at −180 ( 2 ) . We are not entirely sure of the origin of the reference to position −188 in Freeman et al. Moreover, the references they cite for E2F1-mediated MCL-1 transcriptional regulation refer to promoter regions spanning −143/+10 and −117/+10, both of which are outside the nucleotide insertion-harboring region that we described. It has been shown that the MCL-1 promoter that contains the nucleotide insertion–harboring region has statistically significantly higher activity in response to phorbol 12-myristate 13-acetate (PMA), serum, and granulocyte-macrophage colony-stimulating factor than the promoter with this region deleted ( 2 ) . Our original publication supports these findings because we found an association between the +6 and +18 nucleotide insertions (considered together) and higher mRNA and protein levels. By contrast, Freeman et al. report reduced PMA-stimulated transcription activity with the MCL-1 nucleotide insertions in two transfected human cell lines. The promoter region being evaluated and the length of the transfected segment are likely to influence the findings ( 2 ) . Because we have not performed luciferase assays, we can neither confirm nor refute their findings. However, using a genetic algorithm for eukaryotic promoters (RegScan [ http://wwwmgs.bionet.nsc.ru/mgs/ ]), we found that the predicted MCL-1 promoter activity score was 0.87 for the wild-type promoter, 0.99 for the +6 insertion, 0.96 for the +12 insertion, and 0.57 for the +18 nucleotide insertion ( Fig. 1C ). The validity of this algorithm is supported by a high concordance between the predicted value and luciferase activity for a polymorphic allele (G125A) of the BAX promoter in CLL ( 4 ) and publications on the computer program ( 5 ) . Can a theoretical algorithm prediction be used to refute practical laboratory data? Not entirely, although it gives a reason to explore alternative explanations. Other factors may regulate the effect of nucleotide insertions on gene transcription, and these might also be tissue specific. Moreover, the +6 and +18 nucleotide insertions are likely to have different effects. Therefore, at present, it is not possible to draw definitive conclusions regarding the effect of nucleotide insertions on MCL-1 gene expression. Even if the nucleotide insertions negatively regulate gene expression, they may still be markers for other stimulatory influences on MCL-1 gene expression. Future studies focused on promoter activity and regulatory factors would help to resolve this issue. Iglesias-Serret et al. assert that, because these insertions are polymorphisms, they are unlikely to predispose to CLL. This was never our conclusion. We simply reported the association of these insertions with clinical endpoints in CD38-negative CLL patients. The assertion by Freeman et al. that these nucleotide insertions are of no prognostic importance in CLL simply because these are polymorphisms is not consistent with available information. The effect of polymorphisms on prognosis in cancer is well known ( 6 , 7 ), as is acknowledged by Vargas et al. This is further supported by a recent paper ( 8 ) that confirms the critical importance of a BAX promoter single-nucleotide polymorphism, G125A, first reported by us ( 9 ), in influencing treatment response and overall survival in CLL.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.