Mouse models

Three- to four-month-old C57BL/6 mice of either sex were used in this study and were assigned randomly to control and test groups. Animals were kept in pathogen-free conditions at 22 °C, 60% air humidity, 12-h light–dark cycle with ad libitum access to food and water. Females stayed in groups of 3–5 mice, males were kept individually. Mice of similar age were assigned randomly to control and test groups. Littermates were evenly distributed among experimental groups and each experimental group contained mice from different litters.

Implantation of a cranial window

A chronic cranial window was implanted over the mouse OB as described previously [19, 36]. Mice were anesthetized by an intraperitoneal (i.p.) injection of ketamine/xylazine (80/4 μg/g of body weight (BW)). Anesthetic depth was monitored by toe pinches throughout the surgery and additional ketamine/xylazine (40/2 μg/g of BW) was injected when necessary. Dexamethasone (2 μg/g of BW) was administered intramuscularly before the surgery. Local anesthetic lidocaine (2%) was applied subcutaneously over the OB for 5–10 min before removing the scalp. The ointment was used to prevent dehydration of the mouse’s eyes. A circular cranial opening (3 mm in diameter) was made by repeated drilling over the two OB hemispheres. Small pieces of bone were removed with a sharp blade and tweezers. Extreme care was taken, not to damage any blood vessels on the surface of the OB. The opening was rinsed with a standard extracellular solution (composition in mM: 125 NaCl, 4.5 KCl, 26 NaHCO3, 1.25 NaH2PO4, 2 CaCl2, 1 MgCl2, and 20 glucose; pH 7.4, when bubbled continuously with 95% O2 and 5% CO2) and covered with a glass coverslip (Ø 3 mm, Warner Instruments, Hamden, CT, USA). The gap between the edge of the coverslip and the skull was filled with cyanoacrylate glue and then strengthened by dental cement. During the surgery and until full recovery from anesthesia the mouse was kept on a heated plate. Postoperative care included an analgesic dose of carprofen (5 μg/g of BW) for 3 days subcutaneously and the antibiotic Enrofloxacin (1:100 v/v) in drinking water for consecutive 10 days. Mice were allowed to recover for at least 3–4 weeks and were subsequently examined for window clarity. Mice were singly housed after cranial window implantation on a 12 h light/dark cycle with food and water available ad libitum.

Construction of viral vectors and production of viruses

All lentiviral vectors were based on the FUGW backbone [39]. The eGFP in the original FUGW plasmid was replaced by a Ca2+ indicator Twitch-2B at BamHI and EcoRI restriction sites to generate FUW-Twitch-2B [40]. Kv1.2wt-T2A and Kir2.1wt-T2A fragments were magnified by PCR (Phusion High-Fidelity PCR Kit, NEB) from pcDNA3-Kv1.2wt (human cDNA template) and Mgi-Kir2.1wt (mouse cDNA template, from Carlos Lois laboratory, Caltech) plasmids and were inserted into FUW-Twitch-2B between XbaI and BamHI restriction enzyme sites. To construct Lenti-Twitch-2B-T2A-H2B-mCherry, H2B-mCherry was inserted after Twitch-2B and T2A to enable nuclear-located expression of mCherry. The sequence of T2A used in this study is 5’-tgggccaggattctcctcgacgtcaccgcatgttagcagacttcctctgccctctccactgcctaccgg-3’.

The second generation of the virus packaging system was used in this study. Briefly, HEK-293 T cells were transiently transfected with plasmids encoding target genes (Lenti-Twitch-2B-T2A-H2B-mCherry, Lenti-Kv1.2-T2A-Twitch-2B, or Lenti-Kir2.1-T2A-Twitch-2B), plus viral packaging helper plasmids pMD2.G (plasmid #12259, Addgene) and psPAX2 (plasmid #12260, Addgene) as described previously [36]. The Lipofectamine 3000 reagent (Invitrogen) was used for transfection. HEK-293 T and 1F8 cells were cultured in Dulbecco’s Modified Eagle’s Medium (DMEM) with 10% heat-inactivated fetal bovine serum and 2 mM L-glutamine at 37 °C and 5% CO2 in air atmosphere. Cell lines were not authenticated. Mycoplasma contamination was controlled by regular PCR tests. To block the K+ channel overexpression-induced reduction of the lentivirus-producing capacity of HEK-293 T cells [41], K+ channel blocker Ba2+ (0.3 mM) was added to the cell culture medium. 48–72 h after transfection, cell culture supernatant containing viral particles was collected and concentrated by centrifugation at 135,000g at 4 °C for 2 h. Concentrated supernatants were resuspended with PBS and titrated in HEK-293 T cells. Titers of about 8 × 109 virus particles per ml concentrated supernatant were used for the following experiments. Retroviral vector Mgi-Kir2.1mut encoding 3 dominant negative site mutations (G144A, Y145A, G146A; Carlos Lois laboratory, Caltech), was packaged in 1F8 cells derived from 293GPG cell line. The vector was derived from a Moloney leukemia virus with an internal promoter from the Rous sarcoma virus. 48 h after transfection, the cell culture supernatant containing the retroviral particles was harvested and concentrated by centrifugation at 135,000g at 4 °C for 6 h. After concentration, the pellet was resuspended with as little volume of ice-chilled PBS as possible. Titers above 108 particles per ml supernatant were used for in vivo injections.

Virus injection into the RMS

Animals with implanted cranial windows were anesthetized with ketamine/xylazine (80/4 μg/g of BW) and fixed in a stereotaxic frame. The ointment was used to prevent dehydration of the mouse’s eyes and 2% lidocaine was applied subcutaneously on top of the injection sites. The skin was carefully removed to expose the skull and two small cranial openings (~ 0.5 mm in diameter) were drilled at the following coordinates: anterior–posterior + 3.0 mm, medial–lateral ± 0.83 mm. Through an opening in each hemisphere ~ 0.8–1 μl of the virus-containing solution was stereotactically injected into the RMS at a depth of -2.95 ± 0.05 mm from the pial surface and a speed of 100 nL/min via a glass micropipette. Thereafter, a metal bar, required for head fixation during the subsequent imaging sessions, was fixed to the caudal part of the skull with dental cement. The other exposed parts of the skull were also covered with dental cement. The mice were returned to the home cage and carprofen (5 μg/g of BW) was injected subcutaneously for 3 subsequent days.

In vivo two-photon imaging

Mice with implanted cranial windows were anesthetized with either isoflurane or the MMF (medetomidine 0.5 μg/g BW, midazolam 5.0 μg/g BW, fentanyl 0.05 μg/g BW) anesthesia and placed on a heating plate. Breathing rate and body temperature were monitored continuously using the animal monitoring system (AD Instruments, Sydney, Australia). The head of the mouse was fixed with the metal bar to the X–Y table, ensuring consistent positioning through imaging sessions. In vivo two-photon imaging was performed using an Olympus FV1000 system (Olympus, Tokyo, Japan) with a MaiTai Deep See Laser (Spectra-Physics, Mountain View, CA, USA) and a Zeiss 20 × water-immersion objective lens (NA 1.0, Carl Zeiss, Jena, Germany). Unless otherwise indicated, cells were imaged using the 890 nm excitation wavelength.

Imaging migration of adult-born JGCs

Mice were anesthetized with isoflurane (2% for induction, 0.8–1.0% for maintenance) and transferred into the imaging setup. The body temperature was kept at ~ 37 °C. Breathing rate was monitored during the whole imaging session and maintained at 110–140 breaths per minute by slightly adjusting the isoflurane concentration in O2. To measure the migration speed, the positions of adult-born neurons were monitored every 15 min for 4 h at 8 and 14 dpi according to the previously established protocol [36]. To create landmarks for single-cell tracking, we labeled blood vessels via i.p. injection of sulforhodamine B (0.1 ml/20 g BW, 1 mM in PBS, Sigma-Aldrich, St. Louis, USA). In addition, the same FOVs were re-imaged with the same imaging settings at 11, 25, and 45 dpi.

Recording spontaneous and odor-evoked Ca2+ transients of adult-born JGCs

The spontaneous Ca2+ transients were recorded in awake mice. Prior to imaging sessions, the mice were trained for head fixation for 10–12 days, as described in ref. [13]. Spontaneous Ca2+ transients of abJGCs were recorded at 12 dpi continuously for 2 min with a frame rate of 7–10 Hz. Twitch-2B was excited at 890 nm and the emitted light was split into 2 channels by a 515 nm dichroic mirror. The emission light of mCerulean3 was filtered with a 475/64 nm band-pass filter and the emission light of cpVenusCD was filtered with a 500 nm long-pass filter. Note that all major characteristics of spontaneous activity (e.g., the fraction of spontaneously active cells, the maximum Twitch-2B ratio, the area under the curve, the fraction of time spent in the active state, etc.) stay constant between DPI 12 and 22 (fig. 5 in ref. [38]).

The odor-evoked responsiveness of abJGCs was measured at 20 dpi. Mice were anesthetized using the MMF anesthesia, the temperature was kept at ~ 37 °C and the breathing rate was ~ 140 breaths per minute during the whole imaging session. Note that under this anesthesia, neither fractions of odor-responding and non-responding, nor the fractions of reliably responding, non-reliably responding and nonresponding abJGCs differ between awake and anesthetized mice (Chi-Square test, p = 0.36 and 0.12, respectively; [13]). Odors were applied through a custom-built flow-dilution olfactometer, positioned in front of the mouse’s snout as described previously [42]. An odor mixture containing 2-hexanone, isoamyl acetate, and ethyl tiglate (purchased from Sigma-Aldrich, 0.6% of saturated vapor each) was applied as a 4-s-long pulse with an inter-pulse interval of at least 2 min. The odor delivery was not timed relative to respiration. Each cell was stimulated at least twice, as described in ref. [42].

Odor deprivation (OD) and two-photon imaging of odor-deprived mice

Cranial window and virus injection were done using the protocol described above. Lenti-Twitch-2B-T2A-mCherry was injected in both hemibulbs. Unilateral naris closure was performed at 5 dpi. Nose plugs were constructed of a 2 mm polyethylene tube (0.58 mm inner diameter, 0.96 mm outer diameter, Portex, UK) and suture thread (size 3-0, Ethicon, Germany) as described previously [36]. Mice were anesthetized using the MMF anesthesia and the plug was accurately inserted into the nostril. Thereafter, mice received an antidote containing flumazenil (0.5 mg/kg BW, Fresenius, Germany) and atipamezole (2.5 mg/kg BW, Alfavet, Germany) and returned to their home cages. After naris occlusion, physiological conditions of experimental animals (e.g., breathing, body weight, and stability of nose plug) were monitored carefully every day. Two-photon imaging of odor-deprived mice was conducted under the MMF anesthesia, as described above.

Immunohistochemistry

Mice were transcardially perfused with 4% paraformaldehyde (PFA) in PBS. The brains were removed and fixed in 4% PFA for 24 h at 4 °C, and then cryoprotected in 25% sucrose in PBS overnight at 4 °C. Next, the brains were embedded in Tissue Tek (Sakura, Zoeterwoude, Netherlands) and frozen at − 80 °C. The immunostaining was performed on free-floating sagittal cryoslices (thickness 30–50 μm) at room temperature. The sections were incubated in a blocking buffer containing 5% normal donkey serum (Jackson Immuno Research, Dianova) and 0.1% Triton-X 100 (Sigma, USA) in PBS for 1 h to prevent nonspecific background staining. After blocking, the sections were incubated with the primary antibodies diluted in the blocking buffer. The following primary antibodies were used: goat polyclonal antibody against GFP (Rockland 600-101-215, 1:1000), mouse monoclonal antibody against Kv1.2 (NeuroMab 75-008, 1:200), rabbit monoclonal antibody against pCREB (Cell Signaling 9198S, 1:400), mouse monoclonal antibody against NeuN (Millipore MAB377, 1:1000), rabbit polyclonal antibody against DCX (Abcam ab18723, 1:1000). After overnight incubation with primary antibodies at 4 °C, the sections were rinsed in PBS three times for 10 min each and incubated with secondary antibodies (2% BSA and 1% Triton-X 100 in PBS) for 2 h in the dark at room temperature. The secondary antibodies were as follows: donkey-anti-mouse or anti-rabbit IgG-conjugated Alexa Fluor 488 (A21202 or A21206, 1:1000), donkey-anti-goat IgG-conjugated Alexa Fluor 594 (A11058, 1:1000), donkey-anti-mouse IgG conjugated Alexa Fluor 680 (A10038, 1:1000), all purchased from Invitrogen (Grand Island, NY). Afterward, the sections were washed three times in PBS for 10 min, transferred to Superfrost Plus charged glass slides (Langenbrink, Emmendingen, Germany), and mounted in Vectashield (Vector Laboratories, USA) or ProLong Gold (Invitrogen) mounting medium. Immunostained slices were imaged using an Olympus Fluoview 300 laser scanning microscope (Olympus, Tokyo, Japan) coupled with a MaiTai mode-locked laser (Spectra Physics, Mountain View, CA, USA). Alexa Fluor 488 and 594 were excited at 800 nm and the emitted light was split by a 570 nm dichroic mirror and filtered with a 536/40 nm band-pass filter as well as a 570 nm long-pass filter. Alexa Fluor 680 was also excited at 800 nm and the signal was collected in the long-pass channel of a 670 nm dichroic mirror.

RNA sequencing

Male 3-month-old mice were bilaterally injected into the RMS with either Lenti-Twitch-2B-T2A-H2B-mCherry, Lenti-Kv1.2-T2A-Twitch-2B, or Lenti-Kir2.1-T2A-Twitch-2B viruses and sacrificed at 9 dpi. Mice were deeply anesthetized by ketamine/xylazine (100/10 μg/g of BW) and transcardially perfused with 20 ml ice-cold PBS to empty blood vessels from the blood. Then mice were decapitated, and the olfactory bulbs were quickly dissected and transferred into an ice-cold 35 mm dish containing 0.5 ml dissection medium (Hank’s balanced salt solution containing 15 mM HEPES, 25 mM glucose, 0.4 mg/ml DNase I, and 80 U/ml RNAse inhibitor). Olfactory bulbs from two animals were pooled to assure the isolation of at least 3000 adult-born cells. Minced tissue was very gently homogenized in an ice-cold dissection medium and passed through a 70 µm cell strainer. The filtrate was spun for 10 min at 250 g in a refrigerated centrifuge at 4 °C. Pellet was resuspended in 37% Percoll and centrifuged for 30 min at 800 g and 4 °C. First, the upper myelin layer and then supernatant were removed, the pellet was resuspended in sorting buffer (dissection medium without DNase I) and centrifuged once again for 10 min at 800 g and 4 °C. Finally, isolated cells were resuspended in a sorting buffer and kept on ice until sorting. Adult-born cells expressing Twitch-2B were separated by fluorescence-activated cell sorting on a Sony SH800Z sorter (Sony Biotechnology Inc, Surrey, UK) immediately after staining samples with propidium iodide (PI). Single cells were selected based on FSC-W/FSC-H gating, dead cells were excluded based on the PI-signals and transduced cells were identified by their Twitch-2B signal (using a 450/50 bandpass filter). Single Twitch-2B-expressing cells were sorted directly into Eppendorf tubes containing 10 µl sorting buffer, immediately frozen in liquid nitrogen, and kept at − 80 °C until the libraries for RNA-sequencing analysis were prepared.

The synthesis of the cDNA was performed using the SMART-Seq v4 Ultra Low Input RNA Kit (Takara Bio). 3000 to 4500 frozen sorted cells were lysed using a concentration of 100 cells per µl of lysis buffer. Lysis was performed by resuspending the cells by pipetting and incubating for 5 min at room temperature. First-strand cDNA synthesis was performed using 5 µl of cDNA for 90 min at 42 °C. Amplification of the full-length double-strand cDNA was monitored by qPCR and was stopped at 17 PCR cycles during the linear amplification phase. The resulting cDNA presented a fragment size distribution of 1500 up to 5000 bp on the Bioanalyzer High Sensitivity DNA Kit (Agilent) and a concentration above 300 pg/µl measured with Qubit dsDNA HS fluorometric quantification (ThermoFisher Scientific). Next-generation sequencing (NGS) libraries were prepared using 150 pg of cDNA input in the Nextera XT DNA Library Preparation Kit (Illumina), followed by 12 cycles of PCR. Final libraries had a mean fragment size of 370 bp on the Bioanalyzer, a concentration > 5 ng/µl, and a molarity of > 30 nmol/l measured with Qubit. Libraries were sequenced as single reads (75 bp read length) on a NextSeq500 (Illumina) with a depth of > 20 million reads. Library preparation and sequencing procedures were performed by the same individual and we chose a design, aimed to minimize technical batch effects.

Quantitative real-time PCR

Reverse transcribed RNA from sorted cells was generated by SMART-Seq v4 Ultra Low Input RNA Kit (Takara Bio). Real-time PCR was performed using a Fast SYBR green PCR Master Mix (Applied Biosystems) and the QuantStudio™ 5 equipment (Applied Biosystems, Germany). Specific primers (Supplementary Table S3) were designed using the NCBI tool primer BLAST selecting those sequences that span an exon-exon junction to avoid amplification of any contaminating genomic DNA. Three biological samples were determined in triplicate. Single-product amplification was confirmed by running the melting curves. Input quantities were normalized to those for glyceraldehyde 3-phosphate dehydrogenase (Gapdh) and the relative mRNA expression was estimated by the efficiency corrected method [43].

Electrical recordings

20 days after virus injection into the RMS (20 dpi), 250 µm thick sagittal OB slices were cut on Microm HM 650 V vibratome (Thermo Fisher Scientific, Dreieich, Germany). After an incubation period of up to 1 h, slices were transferred into the experimental setup and perfused with the extracellular solution containing (in mM): 125 NaCl, 3.5 KCl, 26 NaHCO3, 1.25 NaH2PO4, 2 CaCl2, 1 MgCl2, and 20 glucose, pH 7.4 when bubbled with 95% O2 and 5% CO2. Experiments were carried out at ~ 33 °C. Twitch-2B-positive control or Kv1.2/Kir2.1 overexpressing abJGCs were recorded in whole-cell configuration using an EPC-10 patch-clamp amplifier (HEKA, Lambrecht, Germany). The intracellular pipette solution contained (in mM): K-Gluconate 140, NaCl 10, HEPES 10, EGTA 0.2, Mg-ATP 4, Na-GTP 0.4, Alexa Fluor 594 0.05, pH = 7.3. Current and voltage traces were acquired at a 20 kHz sampling rate. For current-clamp recordings, currents of different polarity and amplitude (usually from − 40 pA to + 100 pA, 100 or 200 ms long) were injected stepwise to assess the cell’s firing ability. The holding current was adjusted to maintain basal membrane potential at − 70 mV.

AnalysesCalculation of the migration speed of abJGCs

The migration speed of abJGCs was analyzed as described previously [36]. The 3D image stacks acquired during the consecutive imaging sessions (containing both abJGCs and blood vessels) were aligned offline using the blood vessel pattern as an anatomical landmark. Each cell received an identification number, and its position (the center of the cells’ soma) was identified in each of 17 stacks (0–4 h, 15 min inter-stack interval), enabling the reconstruction of the cell trajectory. Next, the X–Y coordinates of each cell’s position were read out using the Fluoview 3.0 Viewer (Olympus). For Z-axis coordinates, the depth of the cells was the relative depth from the dura. With 3D coordinates for each position point, migration distance (D) between the two points in space was calculated according to the following formula:

$$D = \sqrt \right)^ + \left( \right)^ + \left( \right)^ }$$

where (X1, Y1, Z1) and (X0, Y0, Z0) were coordinates of the cells’ position in the current and the immediately preceding stacks, respectively. Migration speed was defined as the translocation of the cells’ soma between the two consecutive time points divided by the respective time interval (µm per 15 min). Because the step size of the acquired stacks was 2 μm, a cell was considered moving if the translocation of the cells’ soma between the two consecutive time points was more than 4 μm.

Analyses of spontaneous Ca2+ transients

Data analysis was performed offline with ImageJ and a custom-made routine in Matlab (R2016b, The MathWorks, United States). Circular regions of interest (ROIs) were manually drawn within the soma of each cell. A fluorescence trace for each cell was obtained by averaging all pixels within the ROI. The background signal was obtained from the ROI of a comparable size devoid of fluorescent processes and located near the cell of interest. The image stack was separated into 2 substacks for mCerulean3 and cpVenusCD channels, respectively. The fluorescence traces were calculated separately for both substacks and filtered using a lowpass Butterworth infinite impulse response filter with a cut-off frequency of 0.6 Hz. The Twitch-2B ratio signal was calculated using the formula:

$$Ratio = \frac } \right)Soma - F \, \left( } \right)\,Background}} \right)Soma - F \, \left( \right)\,Background}}$$

Thereafter, the traces were imported into Matlab to analyze the following parameters: basal and maximum Twitch-2B ratio and the area under the curve normalized to the total recording time (AUC/sec, [38]). The basal ratio was the mean of the lowest 10% data points in the histogram of each individual trace. The value for the maximum ratio was calculated as follows: the filtered traces were processed by a sliding average algorithm with a window size of 1.5 s to determine the maximum ratio (maximum average value). Parameters of Ca2+ fluctuations were also analyzed in Matlab. To compare experimental and control groups, the fluctuations in Ca2+ signals were detected using the mid-reference level crossing approach (function midcross in Matlab) and the numbers of crossing points, passing through the mid-reference level, per trace were counted. The Gaussian mixture model [44] was used to explore and detect the naturally existing clusters among the counted numbers of crossing points. The correct number of clusters has been estimated using the Bayesian information criterion [45]. Assigned cluster labels (i.e., with and without fluctuations in Ca2+ signals) have been used to calculate the fraction of each cluster in a given experimental group. Subsequently, these fractions were compared statistically using ANOVA.

Analyses of odor-evoked responsiveness of adult-born JGCs

The odor-evoked Ca2+ transients of individual neurons were detected with a custom-written Igor Pro routine (WaveMetrics Inc., OR 97035 USA). First, background fluorescence, measured in the neuropil surrounding the abJGCs, was subtracted as described above and the Twitch-2B ratio signals were calculated and expressed as relative Twitch-2B ratio changes (ΔR/R). For automatic detection of responding cells, all ΔR/R traces were smoothed with a binomial filter (time window 0.3 s). Each smoothed trace was subtracted from the original ΔR/R trace, resulting in the “baseline noise” trace. ΔR/R transients were automatically detected with a template-matching algorithm, taking into account their sharp rise. A ΔR/R change was recognized as an odor-evoked Ca2+ transient if its amplitude was three times larger than the standard deviation of the corresponding baseline noise.

Sholl analyses

The dendritic morphology of abJGCs was analyzed both in vivo and in situ. The three-dimensional stacks were imported into Neuromantic software (https://www.reading.ac.uk/neuromantic/body_index.php) for 3D reconstruction. Neuronal morphology was manually traced to obtain accurate reconstructions. Digitally reconstructed neurons were imported into Image J and analyzed with the Simple Neurite Tracer plugin. The following morphological parameters were read out: the number of dendritic branches and the total dendritic branch length (TDBL). Sholl analysis was performed by counting the number of intersections between dendrites and centered on the soma concentric spheres with 10 μm radius increments [46].

Analyses of the survival rate

All FOVs including cells and blood vessels were imaged at 14, 25, and 45 dpi under the same settings. The size of each image stack was 635 µm × 635 µm × 200 µm (XYZ). To minimize the effects of cell migration, a safe margin amounting to 100 µm × 100 µm (XY) from the image border was introduced, and cells residing within the margin area were excluded from the analysis. By using blood vessels as a landmark, a cell was considered surviving if its soma was found at the same position in both image stacks (14 and 25 dpi, or 25 and 45 dpi). Based on the resolution of the microscope, an offset of ≤ 4 µm was tolerated.

Analyses of the pCREB expression level and quantification of the Kv1.2 expression

All brain slices were stained and imaged under the same conditions (i.e., antibody concentration, incubation time, laser power, photomultiplier voltage, etc.). Background-subtracted images were generated by estimating background noise from 5 negative control slices, which were stained with secondary antibodies only, and subtracting the median noise value from the original images. Each slice was simultaneously stained with anti-pCREB, anti-GFP (recognizes Twitch-2B), and anti-NeuN antibodies, with secondary antibodies conjugated with Alexa Fluor-488, 594, and 680, respectively. All images were processed using the following protocol in ImageJ: (i) generate 3 substacks for pCREB, GFP, and NeuN staining by splitting the original 3D stack; (ii) draw the ROIs for the abJGCs on the central slice (Z-axis) of the GFP (Twitch-2B) stack; (iii) find the corresponding frame (same depth) in the NeuN stack; (iv) subtract the median background noise value from the NeuN stack; (v) adjust the threshold of the image to highlight all NeuN positive regions and draw ROIs for NeuN-positive cells in the whole field of view; (vi) measure the fluorescence intensity of NeuN-positive mature neurons and GFP-positive abJGCs in the stack of pCREB; (vii) calculate the relative pCREB expression level as the intensity of pCREB fluorescence of each GFP-positive adult-born JGC divided by the median intensity of all NeuN-positive mature neurons located in the same field of view.

$$} = \frac}\left( }} \right)}\,background}}}\left( }\left( }^ }} \right)}\,background} \right)}}$$

For quantification of the Kv1.2 expression, the background-subtracted images were generated as described above and a custom-written Matlab code was used to calculate separately the intensity of immunofluorescence in the cell somata and the surrounding neuropil. The relative Kv1.2 expression level was calculated as the ratio of the somatic fluorescence divided by the fluorescence of the surrounding neuropil.

Transcriptomic analyses

The read quality of RNA-seq data in fastq files was assessed with ReadQC (ngs-bits version 2019_09). Raw reads were filtered to remove sequencing adapters and for quality trimming using SeqPurge (ngs-bits version 2019_09). Filtered reads were aligned against the reference mouse genome of the Ensembl Mus Musculus GRCm38 using STAR (v2.7.0f), allowing gapped alignments to account for splicing. Low mapping quality reads or those mispairing or multi-mapping were removed using Samtools (v1.9) and visually inspected in the Integrative Genome Viewer (v2.4.19). A matrix of raw counts was built using subread (v1.6.4). Low-expressed transcripts were filtered out to minimize the false-positive rate. For each dataset, all transcripts with less than 1 count per million in at least two samples were excluded, leaving 11,611 genes for further differential expression analysis. Differentially expressed genes (DEGs) were identified using edgeR (v.3.24.3) with R (v3.5.2) (https://www.R-project.org/) following the standard workflow. With this method, the size of the library is corrected and differential expression is tested by using negative binomial generalized linear models [47]. Genes with an absolute > 2-times-fold change between the control and Kv1.2/Kir2.1 groups (885 genes) were considered as DEGs and imported into the online portal Metascape (https://metascape.org/gp/index.html#/main/step1) to run GO biological processes, cell components, molecular functions, and KEGG pathway enrichment analyses [48]. In this study, the focus was on signaling pathways downstream of pCREB as well as on the routes leading to CREB phosphorylation.

Electrical recordings

To calculate the threshold of an AP, the first derivative was analyzed and the membrane voltage that corresponds to a threefold S.D. of a baseline value was taken as a threshold [49]. The same results were obtained using two other methods for threshold definition: (a) rate-of-rise point of 50 kV/s [49] or (b) the first peak in the 3rd derivative [50] (not shown). The input resistance was measured in whole cell voltage-clamp configuration with cell membrane voltage set to − 70 mV by applying 500 ms-long hyperpolarizing pulses with amplitudes of 40–20 mV and measuring the respective steady-state currents. To determine resting membrane potentials, the I–V plots were linearly interpolated to measure the voltages at zero holding current. The data points for the I–V plots were obtained by clamping the cells at three different potentials − 110 mV, − 90 mV and − 70 mV and measuring the respective holding currents.

Statistical analyses

Statistical analyses were performed using the GraphPad Prism 9 software (GraphPad Software Inc, San Diego, California USA). The Shapiro–Wilk test was used to check the normality of data distribution within the data sets. P < 0.05 was considered statistically significant. In the case of normally distributed data, the parametric (i.e., Student’s t-test or ANOVA followed by Tukey's multiple comparison test) otherwise nonparametric (i.e., Mann–Whitney U test or Kruskal–Wallis test followed by Dunn's multiple comparisons test) tests were used. For Sholl analysis, we used the generalized mixed-effects model with the Poisson family (R programming language [51] with lme4 package [52]). P-values were obtained using the lmerTest package, which estimates p-values based on Satterthwaite approximation [53]. Post-hoc comparisons were done using the lsmeans package [54]. Box plots were used to present 5 parameters of the dataset: 10th, 25th percentile, median, 75th, and 90th percentile. Unless otherwise indicated, the error bars represent median ± IQR (interquartile range).