Blood Journal
Leading the way in experimental and clinical research in hematology

A recurrent in-frame insertion in a CEBPA transactivation domain is a polymorphism rather than a mutation that does not affect gene expression profiling–based clustering of AML

  1. Bas J. Wouters,
  2. Irene Louwers,
  3. Peter J. M. Valk,
  4. Bob Löwenberg, and
  5. Ruud Delwel

To the editor:

Mutations in CEBPA, the gene encoding the transcription factor CCAAT/enhancer binding protein alpha (C/EBPalpha), have been reported in multiple studies, and are found in approximately 8% of patients with acute myeloid leukemia (AML).1,2 Specific regions of the gene tend to be most commonly mutated: (1) in-frame insertions in the basic/leucine zipper (bZIP) region and (2) truncating out-of-frame insertions or deletions in the N-terminus.1,2 Although mutations are most frequently found in these 2 regions, other abnormalities have been described as well.2 Fröhling et al reported in 6 of 236 AML cases the existence of an in-frame insertion mutation of 6 nucleotides.3 This insertion is predicted to result in a histidine-proline duplication (HP196-197ins) in a transactivation domain of C/EBPalpha. In vitro studies have suggested that this proline-histidine–rich region may play a role in antiproliferative control, although this notion has not been supported by in vivo experiments.4,5 Remarkably, in none of the other initial CEBPA mutation studies was the insertion reported as either a mutation or a polymorphism, notwithstanding the fact that investigators frequently applied single-strand conformation polymorphism (SSCP) analysis or nucleotide-sequenced the complete CEBPA cDNA (see Leroy et al2 for references). More recently, one other group described the HP duplication in 20 (20%) of 100 AML samples.6 In this study, the insertion was reported in 7 (39%) of 19 healthy volunteers as well, questioning its role in AML.6

In a cohort of 285 AML cases, we previously selectively screened for the 2 major mutation types and identified 17 patients with mutations.7,8 Here, we asked whether this cohort also included cases with HP196-197ins. By means of a denaturing high-performance liquid chromatography (dHPLC) approach9 and subsequent nucleotide sequencing, we identified the heterozygous HP196-197ins in 9 patients (3.2% of 282 available samples). We also screened an independent second cohort of 305 AML cases, and again found 12 cases (3.9%) to present with this duplication. Finally, we analyzed a series of 274 nonleukemic blood samples and found 22 individuals (8.0%) to carry the same insertion.

Cases with CEBPA mutations were found predominantly in 2 distinct gene expression clusters8 (Figure 1). We asked whether cases with HP196-197ins associated with specific gene expression clusters as well. The 9 specimens carrying HP196-197ins in the first cohort of 285 AML cases did not cluster with CEBPA mutant cases. Moreover, they did not belong to one single previously defined cluster of AML, but were spread out over several subgroups instead (Figure 1).

Figure 1

Correlation view of 285 AML cases including CEBPA status. Pairwise correlations between samples are displayed using 2856 probe sets as described.8 Colors of boxes visualize Pearson correlation coefficient: red indicates higher positive correlation; blue indicates higher negative correlation. Bars next to each sample represent CEBPA status: mutation in bZIP region and/or N-terminus (red), presence of HP196-197ins (blue), or neither (green). For 3 specimens, depicted in white, no material was available for dHPLC analysis. The figure was generated using HeatMapper software (http:www.erasmusmc.nl/hematologie/heatmapper/).10

We conclude that HP196-197ins represents a common CEBPA polymorphism, rather than a mutation, that does not influence gene expression profiling–based clustering of AML specimens. Whether the higher percentage of HP196-197ins observed in nonleukemic samples compared with AML cases is due to chance, or represents an important difference, remains to be elucidated in larger series.

Acknowledgments

This work was supported by a grant from the Dutch Cancer Society “Koningin Wilhelmina Fonds” (EMCR 2006-3522).

Footnotes

  • Correspondence: Ruud Delwel,Erasmus Medical Center, Department of Hematology, Room Ee1342, Dr. Molewaterplein 50, 3015 GE Rotterdam, The Netherlands; e-mail: h.delwel{at}erasmusmc.nl.
  • Conflict-of-interest disclosure: The authors declare no competing financial interests.

References