Fig. 4

DNMT3a2 protein expression. a Four different antibodies were used in these experiments, two with antigens in the DNMT3a1-specific N-terminal region and two with antigens in the DNMT3a1/DNMT3a2 common C-terminal region. b As a positive control and to provide a reference molecular weight, the full-length human DNMT3a protein with a FLAG tag was overexpressed in HEK cells. Cell lysates with the overexpression vector and empty vector were probed with anti-FLAG antibody. DNMT3a was detected at ~140 kDa (arrow), but a non-specific band at ~75 kDa was also detected. Therefore, the FLAG-tagged DNMT3a was affinity purified on an anti-FLAG column. c The purified protein (0.1 μg) was probed with all four primary antibodies, returning a single band at ~140 kDa (arrows) demonstrating that all four antibodies could detect DNMT3a at the correct molecular size for future experiments. d Protein lysates from E11, E18, PN1 brain and 3 M hippocampus (15 μg) were probed with all four primary antibodies. The full-length DNMT3a was detected with all antibodies though with very weak signal in 3 M hippocampal samples (data not shown). To increase the likelihood of detecting DNMT3a in the hippocampal sample, immunoblots were performed again but with twice the protein loading in hippocampal samples (30 μg) as the other samples (15 μg). Full-length DNMT3a was detected including in hippocampal samples (arrows)—though the signal was very low for SC-36769 and 64B1446 antibodies. As the C-terminal antibodies should also detect the shorter form of DNMT3a, DNMT3a2, the PA511141 and 64B1446 blots were examined for banding not found with the N-terminal antibodies. While a ~95-kDa band was observed, it was also evident with the N-terminal antibodies (open circles). No unambiguous banding was evident that could be attributed to DNMT3a2