Animals and Procedure

VIP-Cre mice (Zhang et al. 2021a) were bred with Ai32 mice (Madisen et al. 2012) to generate VIP-Cre+/+::Ai32R26EYFP/R26EYFP or VIP-Cre+/−::Ai32R26EYFP/WT mice (VIP-YFP mice), and then VIP-Cre+/+::Ai32R26EYFP/R26EYFPmice were bred with Ai14 (Park et al. 2015) mice to produce VIP-Cre+/−::Ai32R26EYFP/WT::Ai14R26tdTom/WT mice (VIP-tdTomato-YFP mice). VIP-Cre mice were also bred with Ai14 mice to generate VIP-Cre+/−::Ai14R26tdTom/WT mice (VIP-tdTomato mice). VIP-tdTomato mice were used for cell morphology observation and electrophysiological analysis, whereas VIP-tdTomato-YFP mice and VIP-YFP mice were used for cell morphology observation and immunofluorescence staining. The mice were housed under standard housing conditions and provided food and water ad libitum. We checked the pregnant mice every day at 10 am. If there were newborn pups and their skin was bright red, they were considered postnatal day (P) 0 pups; if their skin was light red, they were considered P1 pups. We collected samples from the mice at 10 am on P0, P2, P4, P6, P8, P10, P12, P14, P16, P18, P21, P28 and P35 for the experiment. The animal use and care protocols were performed according to the guidelines of Zhejiang University. Use of the laboratory animals was approved by the Tab of Animal Experimental Ethical Inspection of the First Affiliated Hospital, College of Medicine, Zhejiang University (approval No. 2,021,001).

Intravitreal Virus Injection into Mouse Pups

Time is required for a virus to be expressed after transduction. To ensure that the virus was expressed before observation, it was injected intravitreally into mice on P4. A thin needle (tip diameter of approximately 2–3 μm) drawn with a glass electrode (Borosilicate glass, O.D.: 1.5 mm, I.D.: 0.86 mm, Sutter Instrument) was attached to a 10 mL syringe for injection. VIP-IRES-Cre mouse pups were anesthetized by hypothermia by placing the pups on ice for approximately 30 s, and then 0.5 µL of AAV2/2-hEF1a-DIO-EGFP-WPRE-pA virus was injected into the vitreous cavity to label retinal VIP-ACs. The needle was inserted into the edge of the black part of the eye, indicating the ora serrata. After the needle penetrated the skin, another breakthrough was felt when the needle entered the eyeball. At this time, the liquid level in the needle tip increased, indicating partial liquid outflow from the eye. The virus was slowly and steadily injected, and the needle was kept in place for 20 s to allow the virus to spread before being withdrawn.

Analysis of Retinal VIP Content by ELISA

After anesthesia, mouse retinas were collected on ice as soon as possible and stored at − 80 °C. The retinas were immersed in 200 µL RIPA lysis buffer (PC101, EpiZyme, Shanghai, China) and ultrasonicated to obtain homogenates. The samples were centrifuged at 4 °C and 2000×g for 20 min, and the supernatant was collected. For analysis, samples were added to wells according to the instructions of the ELISA kit manufacturer. Then, 100 µL horseradish peroxidase (HRP)-labeled antibody was added to each well except the blank wells, and the plate was sealed with a sealing membrane and incubated for 60 min in a 37 °C incubator. After the liquid was discarded, the wells were washed with washing solution (350 µL per well) and the plate was allowed to rest for one minute; this process was repeated for a total of 5 times. Substrate A and substrate B (50 µL) were added to each well, and the plate was incubated at 37 °C for 15 min away from light. Finally, 50 µL of stop solution was added to each well, and the OD value of each well was measured at 450 nm within 15 min.

Western Blotting

Western blotting was performed according to our previously reported protocols (Zhang et al. 2022). In brief, retinas were homogenized in RIPA buffer containing complete protease inhibitor cocktail (Roche). The protein concentration of each sample was determined by a BCA protein assay (Bio-Rad Laboratories). The total protein samples were electrophoresed on 2–20% SDS-PAGE gels and transferred to polyvinylidene fluoride (PVDF) membranes (Millipore, USA). The membranes were incubated with an anti-VIP antibody (Immunostar, USA) overnight, followed by a secondary antibody conjugated to HRP. The signals were visualized using ECL-Plus Western blotting detection reagents (Thermo Fisher Scientific 1,863,096 and 1,863,097). All experiments were repeated three times.

Quantitative Real‑Time PCR (qRT‑PCR)

qRT-PCR was performed according to our previously reported protocols (Zhang et al. 2022). Briefly, total RNA was extracted using TRIzol reagent (Takara, Japan). A total of 1000 ng of RNA from each sample was converted to cDNA. The cDNA was diluted 1:5 for qRT-PCR and each sample was analyzed in triplicate. Each reaction contained of 0.8 µL primer (Tsingke Biotechnology, Beijing, China), 2 µL cDNA, 10 µL SYBR 2X Master Mix (Vazyme, Nanjing, China) and 7.2 µL DEPC-treated water and amplification was performed after preincubation. The relative expression of the Vip gene was calculated using the comparative threshold cycle (2 − ΔΔCt) method (Livak and Schmittgen 2001). The primers used were: Vip-F (TGCTGTTCTCTCAGTCGCTG) and Vip-R (GCTCCTTCAAACGGCATCCT).

Immunofluorescence Staining and Fluorescence Imaging

Immunofluorescence staining was performed as described in our previous work (Zhang et al. 2021b). In short, after anesthetization, eyeballs were collected, the cornea, iris and lens were removed, then fixed, dehydrated with 30% (w/v) sucrose solution, and cut into 15-µm-thick cryosections. Then, the sections were washed, blocked with blocking solution, and incubated with primary antibodies overnight at 4 °C. Then, the sections were incubated with Alexa Fluor 488- or 546-conjugated secondary antibody for 1 h at room temperature. DAPI (1:4,000; Beyotime, Shanghai, China; C1002; Hangzhou Dianrui Technology Co., LTD) was used to label the nuclei. Anti-VGAT (1:400; Synaptic Systems; 131,013; Hangzhou Dianrui Technology Co., LTD)(Chen et al. 2022), anti-PSD95 (1:250; Sigma‒Aldrich; MAB1596; ; Hangzhou Dianrui Technology Co., LTD)(Chen et al. 2022), anti-synaptophysin (1:500; Abcam, MA, United States; ab14692; Hangzhou Dianrui Technology Co., LTD)(Scaramuzzino et al. 2022), anti-VIP (1:200; Abcam, MA, United States; ab272726; Hangzhou Hulk Technology Co., LTD)(Casalia et al. 2021) and anti-gephyrin (1:250; Synaptic Systems; 147,021; Hangzhou Dianrui Technology Co., LTD)(Chen et al. 2022) primary antibodies were used and donkey anti-rabbit IgG (1:1000; Thermo Fisher Scientific, MA, United States; A-21,206; Hangzhou Dianrui Technology Co., LTD) (Chen et al. 2022) was used as the secondary antibody.

For determination of the VIP-AC soma number, images of the retinal sections were acquired with a VS120 virtual digital slice scanning fluorescence microscope (6 slice system) (VS120, Olympus, Japan, wide-field, 546 nm, 488 nm and 405 nm filter, software for image capture: VS-ASW-S6). For imaging, we adjusted the excitation intensity of the mercury lamp to visualize the minimum outline of cell soma. When taking images for determination the VIP-AC dendrite number and of some retinal sections in which synapses were labeled, a confocal microscope (FV1000; Olympus, Japan, confocal, 405 and 488 nm filter, software for image capture: FV10-ASW 4.0) with a 20× or 100× oil lens was used. The excitation intensity of the laser was kept constant for the same channel. All the images were taken at room temperature. When we used the spot function of Imaris to match the fluorescence spots representing dendrites, we kept the system threshold constant; this threshold was set to the optimal level for distinguishing weak fluorescence spots from strong fluorescence spots.

Retinal Slice Preparation

To count the cell somas, we cut the whole eyes into 15-µm-thick slices alone the cornea-optic disk axis. The ten slices between every two collected slices were discarded. To count cell dendrites, we collected 15-µm-thick slices containing the optic nerve.

For electrophysiological recordings, mice were euthanized, their eyes were removed, and their retinas were isolated as described previously (Zhao et al. 2018; Li et al. 2017). Briefly, after anesthetization, the eyeballs were rapidly enucleated and placed in ice-cold, oxygenated (95% O2 and 5% CO2) sucrose-based cutting solution (approximately 4 °C) containing (in mM) 124 sucrose, 26 NaHCO3, 10 glucose, 3 KCl, 3 sodium pyruvate, 1.25 NaH2PO4, 0.2 CaCl2, and 3.8 MgCl2 (pH 7.4) for one minute. Then, the eyeballs were transferred to filter paper on ice, and the corneas and lenses were cut under a microscope. The retinas were transferred together with part of the filter paper to cutting solution for another three minutes. The retinas were carefully isolated and embedded in low-melting point agarose (4% in artificial cerebral spinal fluid [ACSF]) for four minutes. Vertical retinal slices were cut at a thickness of 200 μm on a vibratome (VT 1200 S, Leica) and incubated in a holding chamber where they were completely submerged in ACSF containing (in mM) 125 NaCl, 25 NaHCO3, 10 glucose, 2.5 KCl, 2.5 CaCl2, 1.25 NaH2PO4, and 1 MgCl2 (pH 7.4), incubated with 95% O2 and 5% CO2, and maintained at room temperature for at least 30 min before recording.

Electrophysiological Recordings

Whole-cell current-clamp recordings were performed using a patch amplifier (Heka Elektronik, EPC 10, Germany) with Patchmaster software, as described in detail in our previous studies (Chen et al. 2022; He et al. 2019). Individual retinal slices were continuously perfused with oxygenated ACSF at a rate of 1–2 ml/min at room temperature. Infrared-differential interference contrast (IR-DIC) video microscopy (Olympus, Japan) was used to identify VIP-tdTomato-ACs in retinal slices. All recordings were made on VIP-ACs mainly in the inner nuclear layer (INL). A horizontal pipette puller (P97, Sutter) was used to pull borosilicate glass pipettes (8–10 MΩ), which were filled with potassium-based intracellular fluid containing (in mM) 120 potassium D-gluconate, 10 4-(2-hydroxyethyl) piperazine-1-ethanesulfonic acid (HEPES), 4 ATP-Mg, 0.5 ethylene glycol-bis (b-aminoethyl ether) N, N, N’, N’-tetra acetic acid (EGTA), 0.3 Tris-GTP, and 15 KCl2 (adjusted with KOH to pH 7.3, 280–290 mOsm/L). To record spontaneous excitatory postsynaptic currents (sEPSCs) and spontaneous inhibitory postsynaptic currents (sIPSCs), cesium-based intracellular fluid containing (in mM) 100 CsCH3 SO3, 20 KCl, 10 HEPES, 7 Tris 2-phosphocreatine, 4 Mg-ATP, 0.3 Tris-GTP and 3 QX-314 (pH 7.3, 285–290 mOsm/L) was used. All experiments were performed at room temperature under normal light conditions. After establishing the whole-cell configuration, the fast and slow capacitance as well as the series resistance (Rs) were carefully adjusted. The amount that the slow capacitance was adjusted was considered the cell membrane capacitance (MC). The Rs was normally less than 33 MΩ, and recordings with a change in Rs exceeding 20% were discarded (Veruki and Hartveit 2002). Seals with a resistance higher than 8 GΩ were considered good seals (Wilson et al. 2011). For resting membrane potential (RMP) measurement, the potential at I = 0 was recorded in current-clamp model. For current-clamp recordings, no bias current was injected. The input resistance (IR) was calculated as 5mV / (I− 75mV-I− 70mV) in voltage-clamp model. sEPSCs and sIPSCs were recorded at holding potentials of -70 mV and + 10 mV in regular ACSF, respectively. The E/I ratio of each cell was calculated as E/(E + I) (Antoine et al. 2019; Chen et al. 2022). For potassium current recording, cells were held at − 65 mV, and the voltage was increased from − 65 to 75 mV in 10 mV steps. Among the three subtypes of VIP-ACs (VIP1-ACs, VIP2-ACs and VIP3-ACs) (Perez de Sevilla Muller et al. 2019), VIP1-ACs have gap junctions and likely contribute to the currents measured in voltage-clamp model (Park et al. 2015). Cells were selected as described in previous studies: the main considerations when selecting cells were their location, morphology, glossy and fluorescence layers (Perez de Sevilla Muller et al. 2019). Electrophysiological recordings were digitized at 50 kHz and filtered at 2.0 kHz. MiniAnalysis (Synaptosoft) was used to analyze individual events. Single events were strictly selected according to previously reported standards, namely, a stable baseline, a sharp rising phase, an exponential decay and a single peak (Zhou et al. 2009; Shen et al. 2016). At each set developmental time point, 10–15 cells from three mice were recorded.

Data Analysis

This is a descriptive study and according to previous similar researches, 2–5 mice is adequate(Liang et al. 2018). In each experiment, we used three to six mice. Three to five cells were recorded in each retina. The number of somas was quantified in 20× sections using ImageJ (Fiji) software. We used the spot function of Imaris 9.0.1 for masking and to determine the spine number as previously described (Zhang et al. 2021b). Electrophysiological recordings were analyzed using MiniAnalysis (Synaptosoft, Leonia, NJ, USA) and Igor 4.0 (WaveMetrics, Lake Oswego, OR, USA). One-way analysis of variance (ANOVA) with Dunnett’s multiple comparisons test (normally distributed) or the Kruskal‒Wallis test with Dunn’s multiple comparisons test (variance was not heterogeneous) was used as appropriate. A value of p < 0.05 was considered to indicate significance. “*” refers to comparisons of the initial time point with other time points. “#” refers to comparisons throughout the whole development period. The data were statistically analyzed with GraphPad Prism 8, and the means ± standard deviation (S.D.s) are presented. The picture acquiring, cell recording and data analyzing were performed by different people. To minimize subjective bias, the previous handling people was asked to label the sample in a different sequence according to an appointed rule and the rule will not be open until all the data analyzing were finished. The figure panels were organized into multipart figures with Adobe Illustrator CC 2018.