Apis mellifera worker larvae were derived from three strong colonies reared in the apiary of the College of Bee Science and Biomedicine, Fujian Agriculture and Forestry University, Fuzhou, China.

We predicted sequence of ame-miR-5119 (5′-TCG​GGG​TCC​TAC​ACT​ACT​C-3′) based on sRNA-sequence (Xiong et al., 2018). Specific stem-loop primers and forward primers (F) as well as universal reverse primers (R) of ame-miR-5119 were designed using DNAMAN software (Table 1) and then synthesized by Sangon Biotech Co., Ltd. (Shanghai, China). The total RNA was isolated from 4-, 5-, and 6-day-old larvae of A. mellifera workers utilizing an RNA extraction kit (Promega, Madison, WI, United States). Reverse transcription was performed with stem-loop primers using HiScript^® III first Strand cDNA Synthesis Kit (Vazyme, Nanjing, China), and the resulting cDNA was served as a template for PCR amplification of ame-miR-5119. The reaction system (20 μL) included 10 μL of 2×Hieff^®PCR Master Mix (Yeasen, Shanghai, China), 1 μL of forward and reverse primers (2.5 μmol/L), 1 μL of cDNA template, and 7 μL of DEPC water. The reaction conditions were set as follows: predenaturation at 95°C for 5 min, denaturation at 95°C for 50 s, and annealing at 56°C for 30 s and extension at 72°C for 50 s for 34 cycles; and then extended at 72°C for 10 min. PCR amplification was conducted and the amplification product was then detected by 1.5% agarose gel electrophoresis. The expected fragment of 6-day-old larvae was extracted, ligated to the pClone 007 vector (Tsingke, China), and transformed into E. coli DH5α competent cells (Tiangen, China). The cloned product was inoculated onto LB agar medium with 50 μg/mL ampicillin, and cultured overnight at 37°C in a biochemical incubator. A single colony was subsequently selected and transferred into LB broth containing 50 μg/mL ampicillin for 12 h of agitational incubation. A sample of the bacterial culture was obtained for PCR analysis, and the PCR-positive samples were sent to Sangon Biotech Co., Ltd. (Shanghai, China) for Sanger sequencing.

Miranda (v3.3a) (Betel et al., 2008), RNAhybrid (v2.1.2)+svm_light (v6.01) (Rehmsmeier et al., 2004), and Targeted Scan (v7.0) (Lewis et al., 2005) software were used to predict the target mRNAs of ame-miR-5119, with default parameters for each software. The common mRNAs predicted by three software were used as the target mRNAs of ame-miR-5119. The predicted target mRNAs sequences were then compared with the GO and KEGG databases using the BLAST software to obtain annotation information for the target genes.

The potential binding sites between Eth and ame-miR-5119 was predicted using the RNA hybrid software (v.2.1.2). Specific primers for the aforementioned binding sites were designed (Table 1), followed by the use of PCR amplification. The amplified fragments were then cloned into pmirGLO vectors and named miR-5119-wt. Concurrently, the mutant sequences of the above binding sites were designed and synthesized, and then cloned into pmirGLO vectors, designated as miR-5119-mut. The bacterial fluids were sent to Sangon Biotech Co., Ltd. (Shanghai, China) for Sanger sequencing, and reverse transcription using reverse transcription using reverse transcription using reverse transcription using those samples that were sequenced correctly were transferred to a fresh liquid LB medium. Plasmids were extracted using an Endotoxin Removal Plasmid Extraction Kit (Beijing Total Gold Biotechnology Co., Ltd., Beijing, China).

The HEK-293T cells placed into a 37°C incubator for 24 h to reach a cell density of 90%–95%. Cell transfection experiments were carried out by following the instructions for the Hieff Trans™ Liposomal Nucleic Acid Transfection Reagent (Shanghai Yeasen Biotechnology Co., Ltd., Shanghai, China), and 4 transfection groups were set up at the same time: 1) mimic-miR-5119 co-transfected with pmirGLO-Eth-miR-5119-wt; 2) mimic-NC co-transfected with pmirGLO-Eth-miR-5119-wt; 3) mimic-miR-5119 co-transfected with pmirGLO-Eth-miR-5119-mut; 4) mimic-NC co-transfected with pmirGLO-Eth-miR-5119-mut. After the transfection was completed, the cell culture plates were incubated in a 37°C incubator for 24 h. Further, the viability of firefly fluoresceinase and Renilla fluoresceinase was detected on a dual-luciferase assay reporter system (Promega, Madison, WI, United States) using a dual-luciferase detection kit (Shanghai Yeasen Biotech Co., Ltd.), and the relative expression folds were obtained by calculating the ratio of firefly fluoresceinase/Renilla fluoresceinase. Following the method described by Matsumura et al. (2022) and Wang et al. (2023). This experiment was repeated in triplicate.

Specific primers of genes were designed using Primer Premier 6 software (Table 1) and then synthesized by Sangon Biotech Co., Ltd. (Shanghai, China). Total RNA treated with DNase from A. mellifera worker larva at 3-day-old (larval stage), 7-day-old and 8-day-old (prepupal stage), 12-day-old (pupal stage) by respectively extracted using an RNA extraction kit (Promega, Madison, WI, United States) (n = 3), and subsequently divided into two portions for reversing transcription using the HiScript^® III first Strand cDNA Synthesis Kit (Vazyme, Nanjing, China). One portion was subjected to reverse transcription with stem-loop primers, and the resulting cDNA was used as templates for RT-qPCR detection of ame-miR-5119; the other portion was reverse transcribed using a 1:1 ratio of Oligo (dt) and Random primers, and the resulting cDNA was used as templates for RT-qPCR of other genes. Following the instructions of the Hifair^® qPCR SYBR Green Master Mix (Low Rox Plus) kit (Yeasen, Shanghai, China), RT-qPCR was conducted on a QuantStudio 3 fluorescent quantitative PCR system (ABI Company, Tampa, FL, United States). U6 (GenBank ID: LOC725641) was used as the internal reference of ame-miR-5119 and actin (GenBank ID: NM001185145) was used as the internal reference of Eth, Ethr, Jhamt, and Kr-h1 genes. The reaction system (20 μL) included 10 μL of Hifair^® qPCR SYBR Green Master Mix (Low Rox Plus) (Yeasen, Shanghai, China), 1 μL of forward and reverse primers (2.5 μmol/L), 1 μL of cDNA template, and 7 μL of DEPC water. The reaction conditions were established as follows: predenaturation at 95°C for 3 min, denaturation at 95°C for 10 s, and annealing and extension at 60°C for 30 s for a total of 42 cycles; the melting curve program was set to the default setting of the system. Three individuals are collected and mixed together as a single biological sample, each of biological sample underwent three technical replicates, and a total of three independent biological samples. The relative expression level of ame-miR-5119 was calculated using the 2^−ΔΔCT method (Livak and Schmittgen, 2001).

According to the method described by Wang et al. (Wang et al., 2019), ame-miR-5119 mimics (mimic-miR-5119) and inhibitors (inhibitor -miR-5119) as well as corresponding negative controls (mimic-NC and inhibitor-NC) were designed using Dharmacon (Lafayette, Colorado, United States) software (Table 1) and synthesized by GenePharma (Shanghai, China). According to previously described method (Guo et al., 2018), 2-day-old larvae of A. mellifera were carefully transferred to 6-well culture plates containing 800 μL of artificial diet, with 40 larvae per well, and the plates were subsequently placed in a constant temperature and humidity incubator (35°C ± 0.5°C, RH 90%) for 24 h. 3-day-old larvae were transferred to 48-well culture plates, each larva was fed 50 μL of artificial diet containing mimic-miR-5119, Mimic-NC, inhibitor-ame-miR-5119 or Inhibitor-NC (40 pmol/g). Every 24 h thereafter, 50 μL of diet was added to each well until 6-day-old (Stopped eating at 7-day-old). 4-, 5-, and 6-day-old larvae (Three biological samples, n = 3) were collected into sterile and RNA-free microcentrifuge tubes. They were immediately frozen in liquid nitrogen and subsequently stored at −80°C for future use. RT-qPCR was performed to evaluate the effects of overexpression and knockdown of ame-miR-5119 from 4-, 5-, and 6-day-old larvae.

Based on the method described by Borsuk et al. (2017), 4-, 5-, and 6-day-old larvae in each group (n = 3 for each age) were rinsed three times with phosphate buffer saline (PBS) to remove diet residue on the body surface, followed by absorption of the fluid on the larval body surface with clean filter paper. The larvae were weighed using FA2004 electronic scales (Shanghai Shunyu Hengping, Shanghai, China).

Following the method described in Section 2.5, 3-day-old larvae were transferred to 48-well culture plates and fed. Each larva was fed 50 μL of artificial diet containing mimic-miR-5119, mimic-NC, inhibitor-miR-5119 or inhibitor-NC (40 pmol/g, n = 72). Every 24 h thereafter, 50 μL of diet was added to each well until day 6-day-old. The number of pupation and survival of larvae were continuously counted from 3-day-old from 12-day-old. The number of pupation and survival of larvae were continuously observed and counted. Furthermore, statistics of pupation and survival rates were visualized using Graph Pad Prism 10 software.

GraphPad Prism version 10 (GraphPad, San Diego, CA, United States) was used for graph construction and statistical analysis. Data are presented as the mean ± SD. Statistical analysis was performed using the one-way ANOVA or Two-way ANOVA, followed by Tukey’s multiple comparisons.

Stem-loop RT-PCR showed that the expected fragment (approximately 100 bp) was amplified from the 4-, 5-, and 6-day-old larvae (Figure 1A). Sanger sequencing obtained a 19 nt sequence that was consistent with the ame-miR-5119 sequence predicted by transcriptome data (Figure 1B). The prediction results of the target mRNA regulated by ame-mir-5119 showed that the top six target mRNAs were all hormone signaling pathway related genes, including ecdysone-induced protein 75 gene (E75), ecdysoneless gene (ecd,), and Eth (Figure 1C).

The temporal expression of ame-miR-5119 and Eth was detected, and four typical development stages were selected: the middle time of larval stage (3-day-old), prepupa preparation stage (7-day-old), prepupal stage (8-day-old) and pupal stage (12-day-old). The results revealed that the expression level of ame-miR-5119 was significantly higher at the 7-day-old compared to other developmental stages, followed by the 8-day-old, and a lower expression level observed at the 3-day-old and 12-day-old (Figure 2A). Conversely, the expression level of Eth was lowest at 7-day-old and 8-day-old, and elevated at 3-day-old and 12-day-old (Figure 2B). As presented in Figure 2C, recombinant plasmids pmirGLO-miR-5119-Eth-wt and pmirGLO-miR-5119-Eth-mut for dual luciferase assay were successfully constructed. Dual luciferase assay showed that ame-miR-5119 significantly inhibited the transcription of Eth compared to the mimic-NC group. Mutating the binding target site in Eth (Eth-mut), the expression of Eth was not affected by ame-miR-5119 (Figure 2D). These results together indicate th at ame-miR-5119 targets Eth and negatively regulates the expression of Eth.

The results of RT-qPCR detection showed that the expression level of ame-miR-5119 was significantly upregulated in the 4-, 5-, and 6-day-old larvae in the mimic-miR-5119 group in comparison with that in the mimic-NC group (p < 0.05) (Figure 3A). Comparatively, the expression level of ame-miR-5119 was significantly downregulated in 4-, 5-, and 6-day-old larvae in the inhibitor-miR-5119 group as compared to that in the inhibitor-NC group (p < 0.05) (Figure 3B). The results indicated that effective overexpression and knockdown of ame-miR-5119 in the A. mellifera larvae were achieved by feeding specific mimic and inhibitor. As presented in Figure 3C, ame-miR-5119 had reverse complementary binding target to Eth mRNA. Subsequent analyses revealed that the Eth expression levels of 4-day-old, 5-day-old, and 6-day-old larvae in the mimic-miR-5119 group were significantly elevated compared to those in the mimic-NC group (p < 0.05) (Figure 3D). Similarly, the Eth expression levels of larvae at these same ages in inhibitor-miR-5119 group were markedly reduced relative to the inhibitor-NC group (p < 0.05) (Figure 3E).

The development of worker larvae was investigated after overexpression or knockdown of ame-miR-5119, the results were indicative of a gradual increase in larval body weight with rearing time. Overexpression or knockdown of ame-miR-5119 did not affect body weight, only the weight of larvae in the inhibitor-miR-5119 group was significantly lower than that of larvae in the inhibitor-NC group at 4-day-old (Figure 4A). Larval pupation was mainly observed on day 10 to day 12, the results suggested that the number of pupate in the mimic-miR-5119 group was remarkedly lower than that of pupae in the mimic-NC group at day 11, and the total number of pupae also decreased (Figure 4B). In contrast, the number of pupae in the inhibitor-miR-5119 group was obviously higher than that in inhibitor-NC group on day 11, and the total number of pupae as well increased (Figure 4C). Additionally, the survival of larvae after overexpression or knockdown of mimic-miR-5119 was analyzed, the results demonstrated that the survival number of individuals in the mimic-miR-5119 group did not differ from that of the mimic-NC group from day 3 to day 9, but that of individuals in the mimic-miR-5119 group decreased remarkedly after pupation at day 10 (Figure 4D). However, the survival number of individuals in the inhibitor-miR-5119 group was significantly higher than that in the inhibitor-NC group. Notably, the number of deaths in the pupal stage was lower in the inhibitor-miR-5119 group (Figure 4E).

The expression of Ethr, the receptor gene of Eth, was detected by RT-qPCR. The results revealed no significant difference in the expression level of Ethr following mimic-miR-5119 overexpression. However, knockdown of ame-miR-5119 resulted in a decrease in Ethr expression. Further investigation of the expression of JH synthesis gene Jhamt and JH receptor gene Kr-h1 downstream of Eth was conducted, the results showed that the expression levels of these two genes were significantly decreased following overexpression of ame-miR-5119 while significantly increased after ame-miR-5119 knockdown. These results indicated that ame-miR-5119 affects the expression of downstream Jhamt and Kr-h1 genes through target gene Eth, which in turn affects the metamorphosis of honeybee from larva to pupa.