We first verified that both MMRRC and Jackson TetTag mice express tTA, while GFP is only found in the latter line. To do this we performed PCR (Figure 1). In MMRRC transgenic mice, but not their wildtype (WT) littermates, we observed a band corresponding to tTA but not GFP. In Jackson transgenic mice, there was a band for tTA and GFP that was not found in their littermate controls (Figure 1A). Next, we fear conditioned mice from each line and stained brain slices for GFP. As expected, GFP positive neurons were observed throughout the dorsal hippocampus in the Jackson TetTag line but not in MMRRC animals (Figure 1B). Interestingly, the number of GFP positive neurons in the Jackson mice appeared to be much higher than the levels of endogenous c-Fos expression we typically observe after context fear conditioning (Tanaka et al., 2014; Nakazawa et al., 2016). We therefore conducted a series of experiments comparing c-Fos expression in these two lines.

Having determined that Jackson fos-tTA/shGFP mice may express a fusion molecule of c-Fos and GFP, we next evaluated whether the presence of this molecule impacts c-Fos expression as measured with immunohistochemistry (IHC) by staining for c-Fos in Jackson fos-tTA/shGFP and MMRRC fos-tTA mice. To determine whether the transgenic mice express c-Fos at levels similar to their wild type litter mates, we used IHC to examine endogenous c-Fos expression in the dorsal dentate gyrus in behaviorally naïve home-cage mice. Dentate gyrus was chosen because it is a commonly studied area in “engram” studies. Unexpectedly, there were more c-Fos + neurons in the Jackson fos-tTA/shGFP mice than in MMRRC fos-tTA mice or wild type mice from either strain (F(3, 23) = 23.1, p < 0.0001; all post hoc comparisons versus Jackson fos-tTA/shGFP, p < 0.0001; Figure 2A). We also examined c-Fos expression after contextual fear memory retrieval and found similar results; there were more c-Fos + neurons in the Jackson fos-tTA/shGFP mice than in any other group (F(3,21) = 99.4, p < 0.0001; all post hoc comparisons versus Jackson fos-tTA/shGFP, p < 0.0001; Figure 2C). Comparison of c-Fos between home cage and fear conditioned animals revealed significant effects of fear conditioning (F(1, 45) = 4.4, p < 0.05) and genotype (F(3, 45) = 83.8, p < 0.001), but no interaction (F (3, 45 = 1.1, p > 0.05). Although we quantified c-Fos expression in the dentate gyrus, the effect is also qualitatively apparent in other hippocampal subregions like CA1 (Figures 2B, D).

To determine if the apparent overexpression of c-Fos, as measured by IHC, is the result of hyperexcitability or global increases in IEG expression, we quantified Zif268 and Arc expression in the dorsal DG of Jackson fos-tTA/fos-shGFP mice and wild-type animals from the same colony after contextual fear conditioning. After normalizing for the amount of stained cells in wild-type animals for each IEG, we found a difference only in the number of c-Fos + cells between transgenic and wild-type animals, as reported above (genotype X IEG interaction: F(2, 27) = 50.6, p < 0.0001; Figures 3A, B). We found no differences in the number of Zif268 + cells (t(27) = 0.25, p > 0.99) or Arc + cells (t(27) = 0.07, p > 0.99), suggesting the effect observed for c-Fos is not the result of increased neural activity or broad changes in expression of IEGs.

Given that elevated c-Fos expression is not observed in the MMRRC TetTag transgenic line, we suspected it may be related to the fos-EGFP transgene that is only expressed in the Jackson line. Previous papers have reported that this transgene contains the c-Fos promoter as well as several exons/introns from the endogenous c-Fos gene (Kim et al., 2015). If this produces a c-Fos-EGFP fusion protein it could be detected by our c-Fos antibody. The detection of both endogenous c-Fos and the c-Fos-EGFP protein could explain why levels of this immediate early gene, but not others, appear to be elevated. Therefore we tested for the expression of a fused cFos-Exon1-GFP RNA molecule in Jackson fos-tTA/fos-shGFP mice and mice from the MMRRC fos-tTA line (Figure 4). qPCR showed an increase in detection of such a molecule in the Jackson fos-tTA/fos-shGFP mice, but not in mice from the MMRRC fos-tTA line (p < 0.05). By contrast, qPCR for RNA corresponding to cFos-Exon1 and cFos-Exon3 showed no difference between the two mouse lines (respectively, p = 0.167 and p = 0.548). This indicates that mRNA corresponding to a fused cFos-Exon1-GFP molecule is present in the Jackson fos-tTA/fos-shGFP mice but not the MMRRC fos-tTA mice.

To determine if there are any functional behavioral consequences of the presence of the c-Fos -Exon1-GFP molecule, we examined memory acquisition and retrieval in Jackson wild type (n = 4) and fos-tTA/fos-shGFP (n = 4) animals. The mice underwent contextual fear conditioning and received a 5-min memory test 2 days later (Figure 5A). As expected, freezing increased after the shocks during training, but there were no differences in baseline or post-shock freezing between groups (Main effect of time F(1, 14) = 37.8, p < 0.0001; No effect of genotype, F(1, 14) = 1.1, p > 0.05; No time x genotype interaction, F(1, 14) = 0.96, p > 0.05) (Figure 5B). Similarly, freezing decreased across time during the memory test, but no differences were observed between groups [Main effect of time F(4, 56) = 12.01, p < 0.0001; No effect of genotype, F(1, 14) = 1.44, p > 0.05; No time × genotype interaction, F(4, 56) = 0.50, p > 0.05] (Figure 5C). Thus, the acquisition and retrieval of contextual fear memory is unaltered in Jackson fos-tTA/fos-shEGFP transgenic mice compared to wild type animals.

Subjects in this study were 62 (28 male and 34 female) triple transgenic fos-tTA/fos-shEGFP/tetO-H2B-GFP (TetTag) or double transgenic fos-tTA/fos-shEGFP mice. The generation of these mice has been described previously (Reijmers et al., 2007; Tayler et al., 2013). Briefly, fos-tTA/fos-shGFP mice (#018306, Jackson Laboratory) were crossed with tetO-H2B-GFP mice (#005104, Jackson Laboratory). The resulting animals express H2B-GFP under the tetO promoter and the tetracycline-transactivator (tTA) protein under the c-fos promoter. The H2B-GFP transgene was not utilized in this study, and we found no differences between H2B-GFP + and H2B-GFP- mice. No sex differences in C-Fos expression were observed; therefore, male and female mice were pooled for analyses. All mice were then maintained on a C57BL/6J background. Control, wild type animals were mice from the same colony, but negative for the transgenes. fos-tTA mice without the fos-shEGFP transgene were a gift from Dr. Mark Mayford and were treated the same as all other mice. The mice were raised on a diet of chow incorporated with low concentrations of doxycycline (DOX chow, 40 mg/kg, Envigo). Mice were on a 12 h light/dark cycle (lights on at 7am) and had ad libitum access to food and water throughout the experiment. At 21 days, pups were weaned and small tail clippings were collected for genotyping. All mice were 2–4 months old at the beginning of experimental manipulations. Two weeks before the experiment began, mice were single housed and allowed to acclimate to the within-laboratory vivarium. All experiments were conducted in accordance with the UC Davis Institutional Animal Care and Use Committee (IACUC).

gDNA of each mouse in the colony was isolated from tail clips using the Zymo Research Miniprep Plus Kit (Zymo Research #D4068) following the manufacturer’s protocol. DNA samples previously identified as either wild-type or fos-tTA positive from both the MMRRC TetTag and the Jackson TetTag lines, or water control, were added to a PCR mix of GoTaq Hot Start Green Master Mix (Promega #M5122), Nuclease-Free water, fos-tTA (forward: ACCTGGACATGCTGTGATAA, reverse: TGCTCCCATTCATCAGTTCC) and fos-EGFP (forward: AAGTTCATCTGCACCACCG, reverse: TCCTTGAAGAAGATGGTGCG) primers. Resulting PCR samples were amplified with a GoTaq Hot Start PCR cycle. A 4% Agarose gel with Ethidium Bromide was prepared and PCR products were loaded onto the gel alongside a 1 Kb DNA ladder. DNA fragments were separated for 30 min by gel electrophoresis powered to 130 V and imaged by exposing to UV light.

The behavioral apparatus has been described previously (Tayler et al., 2011; Wilmot et al., 2019). Briefly, contextual fear conditioning occurred in a conditioning chamber (30.5 cm × 24.1 cm × 21.0 cm) within a sound-attenuating box (Med Associates). The chamber consisted of a front-mounted scanning charge-coupled device video camera, stainless steel grid floor, a stainless steel drop pan, and overhead LED lighting capable of providing broad spectrum or infrared light. The conditioning chamber was lit with both broad spectrum and infrared light, scented with 70% ethanol, and background noise was generated with a fan in the chamber and HEPA filter in the room. The chamber was cleaned with 70% ethanol before each session.

All behavioral procedures occurred during the light portion of the light/dark cycle. All animals were habituated to handling for 5 min/day for 5 days prior to the beginning of behavioral experiments. Off DOX mice were taken off doxycycline for 48 h before training, while On DOX mice remained on doxycycline chow for the duration of the experiment. Mice in the fear conditioning groups were trained at the end of this 48h period. Training consisted of a 3 min baseline period of context exposure followed by 4 foot shocks (2 s, 0.75 mA) separated by a 1 min intertrial interval. Mice were removed from the chamber 1 min after the last shock. Immediately following training, Off DOX mice were given high concentration DOX chow (1 g/kg) for 24 h to rapidly suppress tTA and then remained on low concentration DOX for the rest of the study. No differences were found between On DOX and Off DOX mice, so these groups were pooled for analyses. Mice were returned to the conditioning chamber 2 days after training for a 5 min memory test. Freezing behavior was measured automatically using VideoFreeze (Med Associates). Home cage mice remained in their home cage for the duration of the experiment until they were sacrificed.

Three mice from the TetTag line, three wild type mice, and seven mice from the Mayford line were randomly selected for qPCR experiments. Mouse cortex was collected and immediately dissected into RNAlater and stored at −20°C. RNA was subsequently extracted using the Qiagen RNeasy kit (Qiagen #74104) following the manufacturer’s protocol. RNA was reverse transcribed into cDNA using the Qiagen QuantiTect Reverse Transcription Kit (Qiagen #205311) following the manufacturer’s protocol. qPCR was conducted on cDNA using Applied Biosystems Power SYBR Green Master Mix (Thermo Fisher #4367659) and the following primers:

Three technical qPCR replicates were used for each mouse/RNA molecule combination. The range of cycle counts across technical replicates was calculated for each condition. If the range was higher than two, the technical replicate that was furthest from the other two replicates was excluded. Following this, the range of cycle counts for technical replicates within each condition was less than or equal to two, and all conditions had at least two technical replicates remaining. Expression levels were calculated for each RNA molecule of interest using the 2^–ΔΔCt method (Livak and Schmittgen, 2001) and using GAPDH as a housekeeping gene. First, the mean cycle count for GAPDH for each mouse was subtracted from the mean cycle count for the mouse/condition of interest (Δ Ct). Then, the mean Δ Ct for wild type mice was subtracted from the ΔCt for each Mayford or TetTag mouse/condition (ΔΔ Ct). Finally, fold changes of RNA molecule expression over that observed in wild type mouse were calculated by raising 2 to the power of −ΔΔCt for each Mayford or TetTag mouse/condition (2^–ΔΔCt).

Ninety minutes after behavioral testing, mice were transcardially perfused with 4% PFA. Following 24 h of post-fixation, 40 μm coronal sections were cut and stained for c-Fos. Slices were washed three times in 1X phosphate buffered saline (PBS) at the beginning of the procedure and after all antibody and counterstaining steps. All antibodies and counterstains were diluted in a blocking solution containing.2% Triton-X and 2% normal donkey serum in 1X PBS, unless otherwise indicated. First, sections were incubated for 15 min in the blocking solution. Then, slices were incubated for 24 h at four degrees in anti-c-Fos (1:5000, rabbit, ABE457, Millipore), anti-Arc (1:1000, rabbit, 156 002, Synaptic Systems), anti-EGR1 (1:750, rabbit, 4153, Cell Signaling) or anti-GFP (1:5000, 13970, Abcam) primary antibodies. Next, slices were placed in biotinylated donkey anti-rabbit or anti-chicken secondary antibody (1:500, Jackson ImmunoResearch) for 1 h at room temperature, followed by Streptavidin-Cy2 or Cy3 (1:500, Jackson ImmunoResearch) for 45 min. Finally, sections were stained with DAPI (1:10,000 in PBS, Life Technologies) for 10 min, mounted on slides, and coverslipped with Vectashield anti-fade mounting media (Vector Labs).

Images were acquired at 20× magnification using a fluorescence slide scanner (BX61VS, Olympus), then cropped to include the dorsal dentate gyrus. A blinded experimenter performed cell counts on the entire suprapyramidal blade of the dentate gyrus in 3–4 sections from each animal (6–8 hemispheres). Cells were counted using the multi-point tool in Image-J. Counts were averaged across slices to obtain one value per animal.

Behavioral data was analyzed using repeated-measures ANOVA. For the training session, baseline freezing is the average freezing during the 3 min of context exposure prior to the first shock. Post-shock freezing is the freezing averaged across all 1-min interstimulus intervals between shocks. H2B-GFP cell count data were analyzed using unpaired t-tests. IEG cell count data was analyzed using ANOVA followed by Bonferroni post hoc comparisons as necessary. The Mann-Whitney U test was used to test for differences between mouse lines in qPCR RNA expression data for each molecule of interest.