The DNMT3A R882H mutation does not cause dominant negative effects in purified mixed DNMT3A/R882H complexes

Abstract

The DNA methyltransferase DNMT3A R882H mutation is observed in 25% of all AML patients. DNMT3A is active as tetramer and the R882H mutation is located in one of the subunit/subunit interfaces. Previous work has reported that formation of mixed wildtype/R882H complexes leads to a strong loss of catalytic activity observed in in vitro DNA methylation assays (Russler-Germain et al., 2014, Cancer Cell 25:442–454). To investigate this effect further, we have prepared mixed wildtype/R882H DNMT3A complexes by incubation of individually purified subunits of the DNMT3A catalytic domain and full-length DNMT3A2. In addition, we have used a double affinity tag approach and specifically purified mixed catalytic domain complexes formed after co-expression of R882H and wildtype subunits in E. coli cells. Afterwards, we determined the catalytic activity of the mixed complexes and compared it to that of purified complexes only consisting of one subunit type. In both settings, the expected catalytic activities of mixed R882H/wildtype complexes were observed demonstrating an absence of a dominant negative effect of the R882H mutation in purified DNMT3A enzymes. This result suggests that heterocomplex formation of DNMT3A and R882H is unlikely to cause dominant negative effects in human cells as well. The limitations of this conclusion and its implications are discussed.

Comprehensive structure-function characterization of DNMT3B and DNMT3A reveals distinctive de novo DNA methylation mechanisms

“…2e). We then validated these observations by in vitro methylation analysis on 30-mer oligonucleotide substrates 34 with CpG sites in different trinucleotide flanking sequence context, which reveals methylation preferences of both enzymes in strong agreement with the results from the NGS profile ( Supplementary Fig. 3a, b).…”

“…1e). Furthermore, we assayed human DNMT3A-and DNMT3B-mediated methylation on 30-mer oligonucleotide substrates 34 containing a CpG with either the SatII 6-bp flanking sequence (TCCATTCGAT GATG) or a DNMT3B-disfavored reference substrate (AGGC GCCC, rank 4092 of 4096 in the B/A preference, Fig. 1d).…”

Functional Analysis of DNMT3A DNA Methyltransferase Mutations Reported in Patients with Acute Myeloid Leukemia

“…The ability of enzymatically inactive regulatory factor Dnmt3L to activate WT and mutant Dnmt3a-CD was examined ( Figure 5B). Dnmt3a-CD in complex with Dnmt3L forms a linear heterotetramer consisting of two Dnmt3a-CD and two Dnmt3L subunits [34]. Dnmt3L positions the Dnmt3a catalytic loop and improves the catalysis [3].…”

Comprehensive structure-function characterization of DNMT3B and DNMT3A reveals distinctivede novoDNA methylation mechanisms

“…1e). Furthermore, we assayed human DNMT3A-and DNMT3Bmediated methylation on 30-mer oligonucleotide substrates 34 containing a CpG with either the SatII 6-bp flanking sequence (TCCATTCGATGATG) or a DNMT3B-disfavored reference substrate (AGGCGCCC, rank 4092 of 4096 in the B/A preference, Fig.1d). Strikingly, DNMT3B showed an ~11-fold higher activity on the SatII substrate, whereas DNMT3A was ~15-fold more active on the reference substrate ( Fig.…”

“…3, 7a,b, 13 were measured using 1 µM biotinylated double stranded 30-mer oligonucleotides containing a single CpG site basically as described 52 . Using a standard substrate GAG AAG CTG GGA CTT CCG GGA GGA GAG TGC 34 , flanking sequences selected to be preferred or disfavored by DNMT3A and DNMT3B were inserted as described in the text and figure legends. In all substrates, the lower DNA strand was biotinylated.…”