Fig. 2: Sparse and random Dnmt3a2 overexpression does not allocate or enhance memory.
From: Neuronal ensemble-specific DNA methylation strengthens engram stability
a Schematic representation of the viral constructs used to express GFP or HA-Dnmt3a2 in a random and sparse subset of DG neurons14. b Experimental schedule used to assess the effect of DG random and sparse Dnmt3a2 overexpression in contextual fear memory. Graph represents % time freezing in the conditioned context. Mice were injected with CaMKII-mycCreERT2 and hSyn-loxP-GFP (Cre-loxP-GFP) or hSyn-loxP-HA-Dnmt3a2 (Cre-loxP-HA-Dnmt3a2) (Cre-loxP-GFP (n = 10) vs Cre-loxP-HA-Dnmt3a2 (n = 8), t(16) = 1.022, p = 0.3222 by unpaired t-test). c Experimental schedule used to assess the effect of DG random and sparse Dnmt3a2 overexpression in neuronal ensemble allocation. d, e Representative images of immunohistochemical analysis of GFP, HA-Dnmt3a2 and Arc expressions of brain slices of mice injected with Cre-loxP-GFP or Cre-loxP-HA-Dnmt3a2. Scale bar represents 50 μm. f Observed over chance overlap between GFP or HA-Dnmt3a2 and Arc expressions in the DG of mice injected with Cre-loxP-GFP or Cre-loxP-HA-Dnmt3a2 (Observed overlap vs chance overlap: GFP: n = 4, t(3) = 0.5203, p = 0.6388; HA-Dnmt3a2: n = 4, t(3) = 0.768, p = 0.4984 by paired t-test; Observed/chance GFP (n = 4) vs HA-Dnmt3a2 (n = 4), t(6) = 0.4584, p = 0.6628 by unpaired t-test). Dashed line represents the chance level. g Percentage of Arc positive neurons in the DG of mice injected with Cre-loxP-GFP or Cre-loxP-HA-Dnmt3a2 (n = 4, t(6) = 0.2721, p = 0.7947 by unpaired t-test). ITR: inverted terminal repeat, w- week, DG: dentate gyrus of the hippocampus, CFC: contextual fear conditioning, rAAVs: recombinant adeno-associated viruses. ns: not significant by the respective statistical test. Error bars represent s.e.m. Source data are provided as a Source Data file.
